Vous êtes sur la page 1sur 424

Chapter 0 Page i September 12, 200810:44

Chapter 0 Page ii September 12, 200810:44
Chapter 0 Page iii September 12, 200810:44




Edited by
Ronald B. Mitchell

Los Angeles London New Delhi Singapore Washington DC

Chapter 0 Page iv September 12, 200810:44

Introduction and editorial arrangement Ronald B. Mitchell 2008

First published 2008

Apart from any fair dealing for the purposes of

research or private study, or criticism or review, as
permitted under the Copyright, Designs and Patents
Act, 1988, this publication may be reproduced, stored
or transmitted in any form, or by any means, only
with the prior permission in writing of the publishers,
or in the case of reprographic reproduction, in
accordance with the terms of licences issued by the
Copyright Licensing Agency. Enquiries concerning
reproduction outside those terms should be sent to
the publishers.

Every effort has been made to trace and acknowledge

all the copyright owners of the material reprinted
herein. However, if any copyright owners have not
been located and contacted at the time of publication,
the publishers will be pleased to make the necessary
arrangements at the rst opportunity.

SAGE Publications Ltd

1 Olivers Yard
55 City Road
London EC1Y 1SP

SAGE Publications Inc.

2455 Teller Road
Thousand Oaks, California 91320

SAGE Publications India Pvt Ltd

B 1/I 1 Mohan Cooperative Industrial Area
Mathura Road
New Delhi 110 044

SAGE Publications Asia-Pacic Pte Ltd

33 Pekin Street #02-01
Far East Square
Singapore 048763

British Library Cataloguing in Publication Data

A catalogue record for this book is available from the

British Library

ISBN 978-1-4129-3005-5 (set of four volumes)

Library of Congress Control Number: 2008923783

Typeset by Mukesh Technologies Pvt. Ltd.,
Pondicherry, India
Printed on paper from sustainable resources
Printed and bound in Zrinski d.d. Croatia
Chapter 0 Page v September 12, 200810:44

Contents VOLUME IV
Part 4: International Environmental Issues and International
Environmental Debates (Continued)
Issues (Continued)

50. Unilateralism and Multilateralism in International Fisheries
Management J. Samuel Barkin and Elizabeth R. DeSombre 3

51. Reducing Pollution of the River Rhine: The Influence
of International Cooperation Thomas Bernauer and
Peter Moser 24

European Union
52. The European Union as an Environmental Governance System
Regina S. Axelrod, Norman J. Vig and Miranda A. Schreurs 49

53. Whalers, Cetologists, Environmentalists, and the International
Management of Whaling M.J. Peterson 74

Marine Pollution
54. Regime Design Matters: Intentional Oil Pollution and Treaty
Compliance Ronald B. Mitchell 113

55. Assessing the Effectiveness of International Environmental
Agreements: The Case of the 1985 Helsinki Protocol
Evan J. Ringquist and Tatiana Kostadinova 146

World Environment Organization
56. Toward a World Environment Organization: Reflections
upon a Vital Debate Steve Charnovitz 173
57. Addressing the Global Governance Deficit Peter M. Haas 194
Chapter 0 Page vi September 12, 200810:44

vi Contents

Free Trade and the Environment

58. Bridging the Trade-Environment Divide Daniel C. Esty 208
59. Beyond Pollution Havens David Wheeler 227

Environmental Security
60. Armed Conflict and the Environment: A Critique of the
Literature Nils Petter Gleditsch 237
61. The Environment and Violent Conflict: A Response to
Gleditschs Critique and Some Suggestions for Future
Research Daniel M. Schwartz, Tom Deligiannis and
Thomas F. Homer-Dixon 259

Sustainable Development
62. Towards Sustainable Development World Commission on
Environment and Development 282
63. The Chimera of Sustainable Development
Wilfred Beckerman 303
64. Sustainable Development and Agenda 21: The Secular
Bible of Global Free Markets and Pluralist Democracy
Timothy Doyle 319

International Financial Institutions and the Environment

65. Delegation to International Organizations: Agency Theory
and World Bank Environmental Reform Daniel L. Nielson
and Michael J. Tierney 336

Concluding Section
International Financial Institutions and the Environment (Continued)
66. The Sovereignty of Nature? Environmental Protection in a
Postmodern Age Paul Wapner 371
67. Earth System Governance as a Crosscutting Theme of Global
Change Research Frank Biermann 396
Chapter 0 Page 1 September 16, 200811:47

Part 4:
International Environmental Issues and
International Environmental Debates
(Continued )
Issues (Continued)
 European Union
 Marine Pollution
 World Environment Organization
 Free Trade and the Environment
 Environmental Security
 Sustainable Development
 International Financial Institutions and
the Environment
Concluding Section
 International Financial Institutions and
the Environment (Continued)
Chapter 0 Page 2 September 16, 200811:47
Chapter 50 Page 3 September 16, 200811:47

Unilateralism and Multilateralism in International
Fisheries Management
J. Samuel Barkin and Elizabeth R. DeSombre

raditional international relations theory suggests that unilateral action

T in pursuit of the national interest should be the first preference of

states in an anarchical international system.1 Empirically speaking,
however, the actions of states are rarely purely unilateral. States often pursue
international relations through bilateral negotiations and multilateral mech-
anisms, such as alliances, treaties, and international organizations. They do
so largely because in a variety of issues they cannot achieve their goals unilat-
erally. Within the broad issue area of international environmental politics,
one of these reasons for multilateralism in particular is far more pronounced.
Transboundary and global environmental issues are almost by definition
problems that cannot be effectively managed by one country alone. For
example, a state cannot guarantee the survival of a highly migratory animal
species because no matter how effective its unilateral efforts at conservation,
the species can always be killed off elsewhere. This logic, inherent to most
international environmental issues, mitigates toward a strong bias in these
issue areas for cooperative mechanisms for international management. We
also see a wide array of multilateral activity undertaken to provide environ-
mental protection. By the time of the Rio Conference on Environment and
Development in 1992, there were over 900 international agreements devoted
wholly or partially to environmental protection,2 and the UN Environment
Programme (UNEP) counted around 150 multilateral agreements devoted
solely to environmental protection.3 The numbers have only increased since
The use of multilateral mechanisms for international environmental mana-
gement is thus the norm, both logically and empirically. Yet we continue to
see cases of unilateral attempts to manage environmental resources that cross
borders or that form part of an international commons. These unilateral
attempts are vastly outnumbered by instances of the more normal multi-
lateral approach, yet they remain important objects of study in the area
of international environmental politics for two reasons. The first relates to
the politics of environmental protection: to more successfully plan attempts

Source: Global Governance, 6(3) (2000): 33960.

Chapter 50 Page 4 September 16, 200811:47

4 International Environmental Issues and Debates

to generate international environmental management, we must know when

countries are likely to engage in unilateral action that can undermine, or
perhaps support, concurrent multilateral efforts to manage the environ-
mental resource in question.
The second reason unilateral actions in support of environmental protec-
tion remain important objects of study involves politics outside of environ-
mental management. Attempts at unilateral action in pursuit of international
environmental goals can prove highly conflictual and can have significant
political implications beyond the immediate environmental issue in ques-
tion. For example, unilateralism in defense of domestic tuna industries by
various South American countries in the 1950s profoundly affected interna-
tional maritime law by providing the impetus for the creation of 200-mile
exclusive economic zones (EEZs). The Turbot War between Canada and
Spain led to the most severe political crisis between these two North Atlantic
Treaty Organization (NATO) allies in the twentieth century. And unilateral
attempts by the United States to protect dolphins from tuna fishers on the
high seas have led to significant confrontations between the United States
and many of its closest trading partners within the core institutions of the
international trading system.
Moreover, none of these unilateral approaches solved the underlying
resource conflict in question. Without acceptance by most of the worlds
states, EEZs could not have been successfully defended, especially by small,
weak Latin American states. Although the Canadians successfully chased off
Spanish fishers in 1995, the ultimate management of the fishery could not
be ensured without Spanish cooperation in the longer term. And even the
economic might of the United States was not powerful enough to change the
behavior of all those killing dolphins in the process of tuna fishing.
Given that multilateral approaches to international environmental poli-
tics are the norm, that unilateral action cannot in the end successfully
manage an environmental resource, and that unilateral action often generates
significant levels of international conflict, why do states choose to engage
in it? Realist theory might suggest that power differentials should be a key
predictor, that unilateral action should be expected when powerful states
confront relatively powerless ones. This approach fails to predict unilater-
alism in international environmental politics. Although the United States can
be found taking unilateral action against its Southern (and weaker) neigh-
bors, we also see Namibia forcibly taking on the European Union (EU) and
winning. Liberal theory might argue that states would engage in this sort
of behavior with issues particularly important to them: those for which the
absolute costs of failure to manage are highest. The importance of the issue,
however, is little better as a predictor of attempts at unilateral action. Some
goals, such as saving fisheries industries from extinction, have indeed been of
considerable domestic economic importance in countries that acted unilater-
ally. Others, such as saving turtles from extinction, have been much less so.
More relevant than the actual importance of the issue will be the domestic
political influence of those concerned about it.
Chapter 50 Page 5 September 16, 200811:47

Barkin and DeSombre International Fisheries Management 5

We argue that the key predictor of attempts by countries at unilateral

action in pursuit of managing particular international environmental goods
is the relative degree of substitutability of the resource. By substitutability
we mean the ease with which a country can find a reasonable substitute for
the environmental good in question. A set of states may share a source of
water, but the one that has alternative water resources has a higher level
of substitutability for that resource. An intact ozone layer may be seen as
a resource no state could do without and for which currently no available
substitutes exist; states share a similarly low level of substitutability on this
issue.4 A country will engage in unilateral action only if the degree to which
it can substitute for the environmental good in question is far lower than
the equivalent degree of substitutability of the other major beneficiaries of
the resource. The actual overall level of substitutability is relatively unim-
portant in this context; only when one country is far less able to substitute
for the good will it consider breaching the dominant norm of interaction in
international environmental politics, that of multilateralism, and engage in
unilateral action.
We examine this process by looking at seven international attempts to
manage conflicts over fishery resources. Two of these cases involve only
negotiated management, whereas the other five are cases in which unilateral
action was taken, although not necessarily to the exclusion of multilat-
eral activities. We shamelessly select on our dependent variable (unilateral
action) to do a first-cut plausibility probe of our hypothesis; if convincing,
it would be useful to then focus more directly on varying levels of resource
substitutability to more thoroughly evaluate our claim. We do not there-
fore investigate the extent to which a differential level of substitutability is
a sufficient condition for unilateral action because we do not choose cases
based on that variable. We limit our focus in this article to the management
of fisheries to provide some degree of comparability within a potentially
immense universe of cases. Fisheries provide a particularly useful set of cases
with which to examine environmental management as well. They represent
the type of resource that practically as well as legally cannot be managed
unilaterally; fisheries are often important economically for the countries that
rely on them, and they are rarely managed well.

Theoretical Framework

First, we provide some definitions of our basic terms. By multilateralism

we mean attempts to address international environmental issues through
methods such as negotiation, treaty mechanisms, or international organiza-
tions and institutions.5 By unilateralism we mean actions by states outside
the context of multilateral or negotiated forums. Particularly relevant are
actions intended either to protect the environmental resource in question
directly, without the consent of other states, or to deter other states from or
coerce them to cease overusing a resource. Such actions can include economic
Chapter 50 Page 6 September 16, 200811:47

6 International Environmental Issues and Debates

measures, such as import restrictions, or administrative measures, such as

fines levied on users of a resource. They can also include the use of force to
physically prevent the use of a resource. These definitions help us examine
different ways states can attempt to influence action internationally and are
consistent with the way unilateral and multilateral action is viewed in envi-
ronmental politics more broadly and in other issue areas within international
relations. We may therefore be able to shed light on the form of international
action states undertake in contexts other than fisheries. It is worth noting,
additionally, that unilateral action is a specific strategy. As such, it fits into
a broader context of action on an issue area, some of which might include
efforts at international cooperation. Moreover, unilateral action may be used
as a tool in support of a bilateral or multilateral outcome, not simply as a
way of avoiding one.6 Unilateralism and multilateralism should therefore
be seen as operating on a continuum and as potential tools for achieving
specific goals that may include a cooperative outcome.
The final term to be defined is substitutability. Following the use of
the term in economics,7 we understand it to mean the degree to which the
consumer of an environmental resource can find reasonable substitutes for
that resource in case of depletion.8 But this definition raises the question,
substitutability for whom? Up to this point, we have been speaking in terms
of state interest, which is broader than is strictly accurate. Substitutability as
used in this article refers to primary consumers: those most directly involved
in extracting the value of the resource from its environment. The fishing fleets
of Canada and Spain, for example, are the primary consumers of turbot. It
is their level of substitutability that is relevant, primarily because they have
the strongest vested interest in the resource in question. The link between
differentials in substitutability and state action will therefore be influenced
by the dynamics of domestic politics, and to determine whether differential
substitutability is sufficient to cause unilateral action, we must look at the
specific domestic political situation.
This returns us to the question of why substitutability and, in partic-
ular, relative differences in substitutability should be the key predictors
of unilateral action. The answer is that significant differences in levels of
substitutability of a resource between countries generate a certain type of
bargaining dynamic. When no difference in levels of substitutability exists
when the levels are similar across countries party to negotiations on the
management of a particular resource all of the countries face a similar set of
incentives. Specifically, the parties are all in a situation in which unsustain-
able consumption of a resource hurts them all equally (or at least similarly).
States may individually face tragedy of the commons type incentives
to consume a greater proportion of the resource than is sustainable since,
absent collective management, there is a disincentive to act alone to preserve
the resource when others may not.9 But should they choose to manage the
resource, they realize both that they can only do it collectively and that
collective management will leave them all better off than the tragedy of the
commons going for their best short-term outcome would create. In this type
Chapter 50 Page 7 September 16, 200811:47

Barkin and DeSombre International Fisheries Management 7

of situation, therefore, no one state can credibly threaten to take an unsus-

tainable amount of the resource in question as a way to gain bargaining
leverage. States are prevented from such action because in the long run it
hurts the unsustainable consumer as much as anyone else and because there
is no reason other states cannot similarly threaten unsustainable consump-
tion as well. Moreover, to the extent that all states involved have an incentive
to manage the resource, none will want to manage it significantly more or
less than the others. And all know they cannot manage it successfully alone.
We would therefore not expect unilateral action in support of management
of a resource for which states have a similar level of substitutability. In other
words, a lower level of substitutability is a necessary condition for a rational
choice to resort to unilateral action.
The same is not true, however, for a resource for which states, or the
primary consumers within states, have widely divergent abilities to substi-
tute. Some still have little incentive for unilateral action. Those actors with
higher relative levels of substitutability for a fishery do not gain from under-
taking unilateral action to preserve that fishery because they are able to
consume some other fishery resource should this one become overly depleted.
In a tragedy of the commons situation, an actors first-choice strategy is to
consume as much as desired (even if this exceeds sustainable levels) of the
resource while others conserve, thus receiving none of the costs and all of
the advantages of conservation. A cooperative outcome becomes possible
(as indicated earlier) when all realize they are worse off by simultaneously
pursuing this strategy than they would be if they all accepted their second-
best outcome and conserved cooperatively.10 But states that can more easily
substitute for the resource in question are hurt less by the degradation of that
resource than are the other users of the resource. As such, they can credibly
threaten to increase consumption as a tool to generate bargaining leverage in
multilateral forums. Or they can simply increase consumption and reason-
ably expect that states with lower substitutability, and thus greater interest
in the well-being of the specific resource in question, will not retaliate in
kind and may even reduce their own consumption.
This dynamic gives states with lower levels of substitutability the incentive
to resort to unilateral action. It is the only available counter to the bargaining
leverage that arrogates to countries with higher levels of substitutability in
multilateral forums as a result of their ability to credibly consume more than
their fair share. States for which the resource is relatively less substitutable
have no bargaining leverage vis--vis other countries within the context of
resource consumption since they have fewer options to use other resources
if collective management fails. States that have fewer substitution options
are therefore much more likely to venture outside multilateral negotiation;
unilateralism is the only recourse for either signaling willingness to escalate
or changing the rules of the game (Figure 1).
In some ways, this argument is about preferences: states able to substitute
other resources for the one in question are not likely to act unilaterally to
preserve it because they have other options should it become exhausted.
Chapter 50 Page 8 September 16, 200811:47

8 International Environmental Issues and Debates

Figure 1: Relative Resource Substitutability for Each Party

High Low

multilateral ,
High Y s unilateral

, multilateral
Low X s unilateral

Note: As these are measures of relative levels, they are also ideal types; it would be possible, for
instance, for both actors to have intermediate levels of substitutability, which would therefore
be equivalent and unlikely to lead to unilateral action.

Those who want more intensely to preserve the resource are those more likely
to resort to unilateral action. By focusing on the issue of substitutability, we
gain the ability to predict the conditions under which states will have more
intense preferences for preservation.
This discussion of substitutability indicates why the standard realist and
liberal arguments fail to work as predictors of unilateral action taken to
manage fishery resources in the international commons. Relative power is
not an accurate predictor because it does not change the basic dynamics
discussed earlier; when a state has either a higher or an equivalent level of
substitutability for a resource compared with another state, it has no partic-
ular incentive for unilateral action despite any relative power advantage. And
states with lower relative levels have little choice but to forgo the benefits of
the resource or to search for those areas in which unilateral action might be
most effective.
We thus expect unilateral action aimed at protection of a fishery to be
rare within environmental politics, but we expect it to occur within certain
predictable parameters. We should be able to determine from characteristics
of the fishery and of the industries and others involved in exploiting it what
the incentives for unilateral action, and by whom, would be. The potential for
unilateral action, of course, would depend on the presence of an actual or a
potential perceived resource shortage. Differences in levels of substitutability
would be inconsequential if none of the actors would face the possibility
of having to substitute for the resource. The potential would also depend
on the ability of the industries involved to capture the attention of their
national governments. But we would expect unilateral action to be taken
only when significant differences in substitutability existed. Examination
of a number of fishery issues that vary on levels of substitutability and on
existence of unilateral action should provide an overview of the usefulness
of this formulation.
Chapter 50 Page 9 September 16, 200811:47

Barkin and DeSombre International Fisheries Management 9


We examine seven fishery resource problems to determine the relationship

between unilateral action and substitutability. The cases are chosen to vary
on potentially relevant factors. We examine the same two states as they
face each other on different fishery issues (Canada and the United States,
addressing both Pacific salmon and the Atlantic fishery off Georges Bank).
We look at a developed state engaged in fishing disputes with another devel-
oped country and with a developing country (conflict between Spain and
Canada over turbot near Canadian shores and between Spain and Namibia
over hake near Namibian shores). We see multiple states in the same fishery
facing different types of resource issues over time (the United States and
Latin America attempting to regulate tuna, concerned first about access to
the stock and later about protection of dolphins). And we look at a varia-
tion on the number of actors involved in the fishery and their proximity to
the coast (with Antarctic fisheries as the quintessential case of many actors
involved in a distance fishery).

The Southern Ocean

The Southern Ocean is among the most biologically productive marine

ecosystems on the planet, and it includes commercially significant stocks of
a variety of species of fish.11 Widespread exploitation of fish stocks did not
begin until the early 1970s because of the remoteness of the region and the
capital investment required to launch commercially successful fisheries there.
Once this investment had been made and commercial fishing had begun on
a large scale, however, the first signs of resource depletion became apparent
in little more than a decade.12
The exploitation of specific stocks of fish in the Southern Ocean is almost
by definition a highly substitutable activity. Fishing takes place far from any
major concentrations of population and thus is accessible only to long-range,
deep-sea fleets that can venture long distances and process their catches
onboard. Vessels that can fish profitably in the Southern Ocean, therefore,
have the capability to fish almost anywhere on the high seas. In fact, most
major fleets exploiting stocks there came from Northern industrialized coun-
tries rather than from those of the much closer Southern Hemisphere, an
indication that mobility and range were simply not a problem.13 The biggest
fleets involved were those of the Soviet Union, Poland, and Japan, although
most signatories of the Antarctic Treaty declared an interest in the long-term
health of the regional ecosystem and the species within it.
Thus, this case illustrates similar and fairly high levels of substitutability
by primary users of fish stocks in the Southern Ocean. We would thus expect
management of the resource, to the extent that it happens, to occur through
multilateral mechanisms, with little incentive for unilateral action. The multi-
lateral mechanism in this case is the Convention for the Conservation of
Antarctic Marine Living Resources (CCAMLR), negotiated by the members
Chapter 50 Page 10 September 16, 200811:47

10 International Environmental Issues and Debates

of the Consultative Committee to the Antarctic Treaty and promulgated in

1980.14 This convention has failed to have a noted impact in moderating
the behavior of the major fishing states in the Southern Ocean, but it
remains the accepted forum for managing those resources and most other
living resources in the region. No states have attempted unilateral action
in managing these fisheries outside the auspices of CCAMLR.15 By the late
1980s, the ineffectiveness of the convention at enforcing conservation of
stocks had led to dangerous levels of depletion of a number of heavily fished
species. The major fleets responded to these developments simply by moving
on to new species or stocks.16 The problem of overfishing was ameliorated
only in the early 1990s by the economic crises brought on by the end of
communism in Eastern Europe, one of the results of which is that the govern-
ments involved could no longer afford the capital required to outfit their
long-distance fleets.17

Georges Bank

Georges Bank, off the coasts of Maine and Nova Scotia, has been an impor-
tant fishery for centuries. It is inhabited by nearly a hundred fish species,
twenty of which have been commercially viable.18 The United States and
Canada have regulated fishing in this region through several treaties, begin-
ning as early as 1818.19 By the mid-1970s, the fisheries off the Gulf of
Maine were dominated by foreign factory trawlers,20 which left fewer fish
available for local fishers. In part to remedy this problem, the United States
passed the 1976 Magnuson Fishery Conservation and Management Act,21
and Canada enacted similar legislation, both countries claiming jurisdiction
over fisheries within 200 nautical miles of their shores. The incentives given
to U.S. fishers to enter the industry, combined with increasing technological
capacity, resulted in a depletion of stocks that brought the North Atlantic
fisheries off the U.S. and Canadian coasts into crisis, with lowering prices
and declining fish stocks.22
Jurisdictional issues had plagued the Georges Bank area, beginning with
issues of access to resources of the continental shelf in the 1960s.23 The
two parties addressed the fishery aspect of this dispute in 1976 with an
interim Reciprocal Fisheries Agreement that allowed fishers from each state
continued access to the undisputed fishery zones of the other to continue
traditional fishing patterns. The agreement expired at the end of 1977, but
the parties agreed to honor it while discussions continued.
Canada and the United States have a similarly low level of substitutability
for the resources of Georges Bank. The fleets that fish in the region largely
comprise small owner-operated boats, and many are seasonal fishers. By the
end of the 1970s, Canadas fleet in the region was composed of around 7,000
vessels. Of these, 6,500 vessels were classified as small, 175 were classified as
medium, and 131 were considered a mid-shore fleet. In addition, 121 large
boats owned primarily by processors and 32 large trawlers constituted
the section of the fleet that could probably have done distance fishing as
Chapter 50 Page 11 September 16, 200811:47

Barkin and DeSombre International Fisheries Management 11

well.24 Large boats and trawlers were responsible for just over half of the
Canadian catch in the region.25 At the time, 67 percent of the U.S. fleet in
the region was owner operated. Most U.S. vessels fishing on Georges Bank
were incapable of processing and freezing fish at sea and therefore could not
move to other areas and certainly could not engage in distance fishing if
they lacked access to the fish on Georges Bank.26
The demise of the Reciprocal Fisheries Agreement led to further negoti-
ations, which resulted in two linked agreements in March 1979. One was
an agreement to submit the boundary dispute to the International Court
of Justice for adjudication, and the other was a fishery treaty contingent
on the adjudication. The two states ratified the boundary settlement treaty,
and a special panel of the court heard the boundary dispute. In 1984, the
panel reached its decision, giving Canada jurisdiction over approximately
one-sixth of the total area including some of the richest fishing grounds
and the United States jurisdiction over the rest.27 Although neither state was
happy with the result, both agreed to abide by it.
The fishery treaty met with a more troubled fate. The U.S. Senate post-
poned ratification discussion when the treaty proved contentious with New
England fishers, and the United States officially withdrew consideration
of the treaty in 1981.28 However, both countries continued to honor the
boundary agreement, which has prevented further significant dispute over
these fisheries. What turned out to be a contentious issue has nevertheless
been addressed largely without unilateral action.

Namibian Hake

The continental shelf off the southwest coast of Africa supports an active
fishery, the most important species of which is hake. The major exploiter
of this resource in these waters in the 1970s and 1980s was the Spanish
long-range fishing fleet, which by some estimates was responsible for as
much as 80 percent of the total catch.29 The Spanish fleet fished in interna-
tional waters outside the 3-mile coastal jurisdictional limit. No international
agreement was in place to regulate the catch in those waters, so there were
no legal limitations on the Spanish catch. This situation changed in March
1990 when what had been the South African mandate of Southwest Africa
became the independent country of Namibia. On gaining independence,
Namibia claimed a 200-mile EEZ in which fishing would be allowed only
by license and then only by states with which Namibia concluded a fisheries
The Namibian fishing fleet was relatively small, accounting for only
about 18 percent of the catch in the newly declared EEZ. Namibia aimed
to develop a fleet capable of fishing within 200 miles of its coast.31 It could
grow, however, only at the expense of either the Spanish fleet or the sustain-
ability of the resource stock. More important, Namibian fishers fish almost
entirely within the EEZ, whereas Spains fishing fleet is almost entirely a
distance fleet. It is the biggest EU fleet, comprising 18,000 vessels.32 One of
Chapter 50 Page 12 September 16, 200811:47

12 International Environmental Issues and Debates

the conditions of Spains accession to the EU in 1986 was that it give up its
rights to fish in European waters for ten years,33 a de facto admission that the
Spanish fleet is capable of finding new waters in which to fish. The Spanish
fleet, therefore, had a high degree of substitutability for the hake caught
in Namibias EEZ. The Namibian fleet, however, had a far lower relative
degree of substitutability for the same stock; without deep-sea capability, it
could not go anywhere else.
On declaring its 200-mile EEZ, Namibia announced a moratorium on
unlicensed fishing in the EEZ and a markedly lower ceiling on the total
catch to preserve stocks. It then entered into negotiations with the European
Community (EC) in which EC vessels would be licensed to fish for Namibian
hake in exchange for compensatory fees. Spanish vessels continued to fish
actively in the Namibian EEZ in contravention of the moratorium, encour-
aged by a lack of enforcement by the Spanish government and a lack of
any meaningful Namibian government naval capability. In the end, Namibia
resorted to dropping fishing inspectors and soldiers onto Spanish trawlers
from helicopters to impound the trawlers and arrest the crews.34 The Spanish
responded by withdrawing the EC from licensing negotiations. Presented
with a choice between accepting business as usual or undertaking fairly
radical unilateral action, the Namibian government opted for the latter.
Presented with the prospect of an end to business as usual, the Spanish
government chose to simply withdraw and fish elsewhere.


Turbot is a species of halibut found in the waters of the northwest Atlantic,

between Canada and Greenland. The harvesting of turbot in interna-
tional waters is governed by the Northwest Atlantic Fisheries Organization
(NAFO), as recognized by treaty by Canada and the EU, among others.
Until the early 1990s, this stock was not actively harvested; turbot is not
a particularly lucrative catch, and the stocks within EEZs were richer than
those in the open ocean. Beginning in the early 1990s, however, demand
for the stock increased markedly for two reasons. The stocks of turbot
within the Canadian EEZ had become significantly depleted, causing the
Newfoundland-based fleet to venture further to sea for healthy stocks. And
much of the Spanish long-distance fleet, forced to leave its traditional fishery
off Namibia, had targeted the turbot stocks in international waters as an
With the increased interest in fishing for turbot in international waters,
the question of an appropriate quota was raised in NAFO. For 1995, it
determined the quota to be 27,000 tons, an amount all parties duly accepted
as a total allowable catch (TAC).35 At this point, it was incumbent on the
parties to divide the TAC into individual national quotas,36 which were then
put to majority vote. A majority of members accepted Canadas claim
based on proximity and a traditional turbot fishery in the area to the
Chapter 50 Page 13 September 16, 200811:47

Barkin and DeSombre International Fisheries Management 13

largest single quota. The majority rejected the EUs similar claim, however,
which was based on the fact that the largest take of the stock in question
during the previous two years had been by the Spanish fleet. As a result,
the national quotas favored Canada, which was allocated 16,300 tons,
compared with 3,400 tons for the EU.37 NAFO rules, however, allow coun-
tries to object to specific quota decisions,38 and the EU, acting on behalf of
the Spanish fishery, did so. It was therefore legally removed from the require-
ment of staying within its assigned quota. Thus, even though the Spanish
and Canadian authorities agreed on a TAC, the fact that they could not
agree on the allocation of quotas made the TAC largely meaningless. Not
surprisingly, it became apparent early in the 1995 season that the total catch
of turbot in international waters would be unsustainably high.
We discussed the relevant Canadian and Spanish fishing fleets earlier.
Canada has no high-seas fleet that fishes outside its EEZ.39 The Spanish
fleet includes many of the same vessels that had been chased from Namibian
waters four years earlier. As such, the Canadian fishing fleet had a substan-
tially lower level of substitutability for the turbot stock in question than did
the Spanish fleet.
As would be predicted, when faced with the inevitable overfishing of
turbot, the Canadian government reacted. It first tried multilateral
approaches, attempting to deal with the issue through the appropriate chan-
nels of NAFO. When these approaches proved fruitless, Canada resorted to
unilateral action. The Canadian Coast Guard claiming Spanish trawlers
were fishing in excess of the TAC, were using illegal equipment, and were
catching fish below the minimum size allowed under NAFO rules seized
and impounded a Spanish trawler outside Canadas EEZ and arrested its
crew. Spain protested that because the trawler had been on the high seas,
this constituted an act of piracy by the Canadian government. In the end, a
compromise was reached in which Canadas quota was decreased somewhat,
Spains was raised, and the TAC was protected.40

Pacific Salmon

The salmon that spawn in the North American Pacific watershed spend
most of their lives in the high seas, then return to Canadian and U.S. rivers
when they have reached full maturity. The salmon stocks in question swim
around the Pacific in a generally clockwise direction, meaning they swim
from international waters into the U.S. EEZ off Alaska. Those that spawn
south of Alaska then swim into Canadian waters, and those that spawn in
Washington state and Oregon then swim back into U.S. waters.
Salmon risk being overfished at sea on the way home to spawn, creating
a risk that not enough new stock will be produced to maintain the fishery.
Recognizing this risk, Canada and the United States signed a treaty in 1985
for management of the Pacific salmon fishery.41 The treaty stipulates as
general principles that the overall salmon catch will not exceed the carrying
Chapter 50 Page 14 September 16, 200811:47

14 International Environmental Issues and Debates

capacity of the stock and that the allocation of the catch between the two
countries will be proportional to the percentage of total stock that spawns
in each countrys rivers. Specific quotas, however, were to be worked out in
negotiation of protocols between the two countries based on these principles.
The two countries were unable to agree on specific quotas between 1992
and 1998.
Canada, on the whole, can substitute for specific Pacific salmon stocks
considerably less effectively than can the United States. The most threatened
commercial stocks are those that come from Canadian rivers. Canada has
an interest in these specific stocks for reasons of general conservation and
because they support an active and important offshore, riverine, and sport
fishing (tourist) industry. The U.S. salmon industry is much more focused on
offshore fishing and thus is much less dependent on specific riverine stocks.42
This difference is reinforced by the fact that U.S. salmon fishers have greater
access in the U.S. EEZ to stocks that spawn in Canada than Canadian fishers
have in their own EEZ to stocks that spawn in U.S. rivers. Thus, one would
expect Canada to face significant pressure to take unilateral action to pursue
improved mutual management of these stocks.
Canada has attempted unilateral action to encourage salmon manage-
ment. In 1994, the Canadian government instituted a temporary law
allowing it to levy a transit license fee of $1,500 on U.S. boats making this
passage, an attempt calculated to force the United States to a negotiated
settlement. In the words of Canadian consul general Bernard A. Gagosz, If
that doesnt work, were going to impose other measures and well finally
get your attention. In 1996, Canada passed a regulation requiring all U.S.
salmon boats passing through Canadian waters to register with the Canadian
Coast Guard and stow their fishing gear, a move ostensibly intended to help
track the fishery but with the effect of inconveniencing U.S. boats.43
The next year, the Canadian Coast Guard used these regulations to
impound four U.S. salmon boats. The boats were seized after failing to
inform Canadian authorities when passing through Canadian waters44 and
were returned after each captain paid a $300 fine. The Canadian Ministry
of Fisheries and Oceans explained that we have stepped up our enforce-
ment activity    to try to signal to the United States our displeasure at our
inability to negotiate a new salmon treaty.45 Canada also attempted to
increase its level of fishing of specifically U.S. stocks as a tool to push for
agreement. At the beginning of the 1997 fishing season, the fisheries minister
announced that boats based in British Columbia would take 12 million of an
expected 18.2 million sockeye salmon stock that would otherwise head into
U.S. waters. The action demonstrated in practical ways to the Americans
our unhappiness with their unwillingness to negotiate an agreement.46 The
geography of the region, in which salmon swim in and out of national waters
multiple times, makes this sort of strategic fishing possible. In the summer of
1999, the two states finally agreed on a cooperative arrangement to assign
catch quotas.47
Chapter 50 Page 15 September 16, 200811:47

Barkin and DeSombre International Fisheries Management 15

Pacific Tuna and EEZs

The Pacific tuna fishery is among the worlds most commercially impor-
tant.48 One-quarter of the total global tuna catch is in the eastern tropical
Pacific Ocean.49 Yellowfin and related tuna migrate long distances along the
Pacific coast of the Americas, from the United States to Chile.50 Although
tuna swim near coasts, they are far enough away that for most of the
twentieth century the area in which most tuna were caught was considered
outside national jurisdiction. Tuna fishing, therefore, took place near coasts
but not within waters controlled by states. The U.S. fleet was responsible
for 90 percent of the tuna catch in this region in the 1960s, and the U.S.
market accounted for by far the largest consumption of tuna from the eastern
Greater mechanization of the tuna fleet, including the addition of freezers
in the 1930s and more powerful engines in the 1950s, increased the catches
by individual vessels, many of which could now travel far from their shores
and stay at sea for months, processing and preserving their catches. As these
vessels grew more profitable, the overall numbers of tuna fishing boats also
expanded. The Inter-American Tropical Tuna Commission began regulating
tuna catches in 1966. By this point, the region clearly could not support as
intensive a tuna harvest as fishers wanted.52
Of the states with fishing vessels in the eastern Pacific, the U.S. fleet
was the most capable of substituting for tuna, particularly from any specific
location, because it was a distance fleet that could fish either for tuna in
different areas of the Pacific Ocean or for other species altogether.53 During
the 1950s, in contrast, fleets from Peru and Ecuador, the states nearest the
largest tuna harvests, were small and accounted for less than 10 percent of
the tuna caught in the region.54 Moreover, many of the boats from Latin
American countries were less mechanized than those of the United States and
were used primarily to fish off their own coasts.55 The exception was Mexico,
which decided to expand and modernize its fishing fleet and outfitted it with
vessels at least technically capable of fishing far from the coast.56
In an effort to gain greater access to and control over tuna (among other
resources), Latin American states began to unilaterally claim jurisdiction
over resources off their coasts. Chile and Peru first declared 200-mile zones in
1947. Ecuador similarly declared jurisdiction over resources within 12 miles
of its coast.57 Other Latin American states did so somewhat later: Costa Rica
in 1975, Mexico and Guatemala in 1976, Colombia in 1978, and Honduras
in 1980.58 The United States did not recognize this expansion of jurisdiction,
particularly with respect to tuna. Even when the United States declared its
own 200-mile fishery conservation zone in 1976, it maintained that tuna,
as a highly migratory species, should be excluded from management within
EEZs. When U.S. fishing vessels continued to fish for tuna near the coasts of
Latin American states, those states began to seize the vessels. In the eastern
Pacific Ocean, Costa Rica captured U.S. boats in 1979, 1980, and 1986;
Chapter 50 Page 16 September 16, 200811:47

16 International Environmental Issues and Debates

Peru captured boats in 1979 and 1980; and Mexico and Ecuador took U.S.
boats in 1980.59
The United States responded with legislation that provided compensa-
tion to U.S. fishers for these Latin American actions, which encouraged U.S.
tuna fishers to continue to fish off Latin American shores.60 Subsequent
U.S. legislation provided for U.S. prohibition of importation of fish prod-
ucts from states that refuse to allow U.S. boats to fish for tuna within their
EEZs.61 Both actions were clearly intended to restore the status quo ante
and to encourage resource extraction rather than management. Interestingly,
however, the Latin American states unilateralism ultimately succeeded, even
in the face of the U.S. response. It is generally accepted that tuna stocks
within EEZs of states are now under the control of those states.

Pacific Tuna and Dolphins

Yellowfin tuna are highly migratory fish that swim, as mentioned earlier,
largely along the Pacific coast of North and South America. In the eastern
Pacific Ocean, the association of yellowfin tuna with dolphins makes
catching tuna cost-effective. Fishers can quickly identify the location of
dolphins and encircle them with tuna nets, assured that a school of yellowfin
tuna will be below. In the process, however, the dolphins that betrayed
the location of the tuna are frequently caught in the nets, where held
underwater they drown.62 Incidental kills of dolphins during tuna fishing
in the Pacific amounted to more than 100,000 per year during the 1970s
and 1980s.63
Saving dolphins from tuna fishers is an issue of existence, rather than
consumptive, benefits.64 For more than two decades, the U.S. public has
expressed an interest in the preservation of cetaceans, including dolphins,
from both direct and incidental effects of commercial fisheries (as in whaling
and the yellowfin tuna fishery, respectively). The Marine Mammal Protec-
tion Act (MMPA) of 1972 placed severe restrictions on the rate of incidental
dolphin kills.65 Those restrictions worked as such, but they also put the
U.S. industry at a competitive disadvantage because fishing for tuna without
setting on dolphins (or with setting on dolphins and allowing them to escape)
is more difficult and expensive than other methods of catching yellowfin
tuna.66 Moreover, U.S. regulations alone did not solve the problem of inci-
dental dolphin catches. The U.S. industry initially accounted for 90 percent
of the total catch of eastern Pacific yellowfin. But as tuna fishing by Latin
American states increased and U.S. dolphin kills decreased, the overall
number of dolphins killed in tuna fishing stayed steady during the 1970s and
1980s even though U.S. dolphin kills declined dramatically.67 The imme-
diate effect of the MMPA on the total number of dolphins killed by the tuna
fishery was therefore marginal.
The nationals of other countries with major Pacific tuna fleets seem to
derive much lower levels of existence benefits from dolphins than does the
U.S. public. The movement to protect marine mammals originated and has its
Chapter 50 Page 17 September 16, 200811:47

Barkin and DeSombre International Fisheries Management 17

strongest support in the United States; citizens in many other states, particu-
larly developing countries, wonder why such effort should be made to protect
a species that is not even endangered. Because existence benefits, particu-
larly when derived from specific species, are inherently difficult to substitute
for, the relative substitutability of dolphins is inherently higher for the U.S.
public than for the publics of major U.S. yellowfin fishing competitors.
After unsuccessful attempts to persuade other countries with major
yellowfin fleets to adopt dolphin-friendly tuna fishing techniques, the United
States took unilateral action intended to force foreign fishing fleets to
adopt practices compatible with the MMPA. It chose to use trade mecha-
nisms, specifically to prohibit the importation of tuna from countries that
caught tuna in ways that killed more dolphins than did U.S. tuna fishers.68
This action elicited two general responses complaint and compliance
concurrently. The United States has twice been taken before General Agree-
ment on Tariffs and Trade panels over this issue, and both panels have ruled
that the extension of the MMPA is illegal under international trade law. At
the same time, total incidental kills of dolphins by tuna fishers have fallen
by a factor of over 10. The decrease reflects foreign fleets having recognized
that the additional costs of dolphin-safe tuna fishing are lower than the costs
of being excluded from the U.S. market for tuna69 the worlds largest and
consumers showing a preference for tuna certified dolphin-safe.70 More
recent progress on protecting dolphins in the course of tuna fishing has been
undertaken through multilateral agreements negotiated under the threat of
continuing U.S. unilateralism.


These seven cases suggest that substitutability is indeed a useful indicator of

when states will undertake unilateral action to promote or enforce interna-
tional fisheries management. The two cases in which levels of substitutability
did not vary greatly among the major users of a fishery resulted in coopera-
tive action to manage the resource through recognized mechanisms of inter-
national law. The effectiveness of these mechanisms at sustainably managing
the resources in question differed significantly in the two cases, but that is not
surprising. As we suggested earlier, there is no reason to assume similar levels
of substitutability will result in successful management of a given resource,
only that any attempts at management will likely be multilateral, even when
management proves difficult. In the five cases in which the relevant levels
of resource substitutability differed substantially, it was the state less able to
substitute for the fishery resource in question that undertook some kind of
unilateral action.
Our substitutability hypothesis suggests nothing about the form such
unilateral action would take, which ranged from the use of force (in three
cases) to the use of economic sanctions (in one) and various legal technical-
ities to harass (in another). It is interesting to note that in two of the three
Chapter 50 Page 18 September 16, 200811:47

18 International Environmental Issues and Debates

cases in which force was used, it was used by the clearly inferior military
power against the clearly superior military power, but in neither case did
the superior power respond in kind. Some aspects of a states capabilities
might help to determine the form the unilateral action will take. For instance,
the United States could impose sanctions on tuna because it was one of the
worlds largest markets for the product; similar action would be unavail-
able to other countries. Canada and Namibia had the jurisdictional ability
to impose whatever regulations they chose on foreign fishers in their waters.
In the remaining two cases, however, states (South American countries and
Canada) resorted to actions that can be seen as acts of war. Perhaps in these
cases there were simply no other available options.
We also need to consider at what point a fishery management problem
becomes sufficiently contentious that unilateral action is likely. The cases we
chose based on unilateral action clearly overrepresent the degree to which
states will act alone to protect an environmental good. Our cases alone
cannot provide an answer to this question, but they do suggest some possi-
bilities. For a fishery resource not perceived to be scarce, differing levels
of substitutability are unlikely to influence behavior. Even when scarcity
becomes an issue, given the impossibility of successful unilateral manage-
ment we would expect to see some effort at joint management of the fishery.
If that proves fruitless, especially for the actor that most relies on the fishery,
we then expect a resort to unilateral action. It is thus worth noting that
there are two stages at which the sort of conflict we saw in five of our cases
can be prevented. First, if the fishery is never over-exploited, it is less likely
to be perceived as scarce. Second, once the fishery begins to show signs of
scarcity, successful multilateral management can head off unilateral action.
Given both that unilateral action can be predicted and that it often succeeds,
even when undertaken by weaker countries against stronger ones, those who
can predict that they would be the targets of such action might be wise to
compromise during the multilateral stage to avoid high odds of significant
international political conflict.
Certainly, some of what determines a resort to unilateral action will
be historically contingent or will depend on factors of domestic politics.
In some cases, unilateral action may be taken earlier in the process of
contentious resource management if the users of the fishery have particularly
high levels of political power, either symbolically or because of economic
clout. Both are prevalent in fishery cases, so the resort to unilateral action
we see in these conflicts may be higher than we would expect in other
issue areas. If, however, the users of the fishery have negligible political
power, unilateral action may never be taken. As is true more broadly, other
elements of domestic politics will influence the political power of inter-
ested actors and thereby influence the likelihood of unilateral action. We
nevertheless believe we have identified an underlying dynamic that explains
the conditions under which unilateral action in environmental management
becomes likely. The dynamics of relative substitutability are insufficient for
Chapter 50 Page 19 September 16, 200811:47

Barkin and DeSombre International Fisheries Management 19

an attempt at unilateralism, but in the case of fisheries politics they are

How applicable are our conclusions to resource issues other than fishery
management? The norm of multilateralism certainly holds in environmental
politics more broadly, yet we also see the use of unilateral economic sanctions
over forestry or endangered species issues. To apply our conclusions about
the predictive power of substitutability, it is important to look at whether
and how the patterns of substitutability in the cases we described can be
We can imagine three major sources of differential levels of substi-
tutability. The first is technological: differences in national means for
exploiting a given resource lead to differing abilities to substitute other
resources. In the cases we examined, technological differences have been
expressed mainly as differences in the ranges of fishing fleets; the more
capital-intensive fleets can go farther and stay out longer, giving them a wider
variety of options of where and what to fish. But in other issue areas, tech-
nological differences need not be ones of range; they can be, for example, the
ability to invent products or processes that substitute for a given resource,
such as the ability to desalinate large quantities of water as an alternative to
depleted rivers.
The second source of differences in levels of substitutability stems from
the ways a resource is used. When one country uses a resource for more
purposes than another does, it will have more difficulty substituting for
that resource. For example, Canada profits from a salmon sport fishery
and from a stock of salmon that to the United States is simply a source
of benefits to a commercial fishery. In substituting for that stock, Canada
would therefore have to find an alternate draw for tourists and alternate
employment for the commercial fishers involved. Similarly, different types
of uses of a resource can lead to differing abilities to substitute for it, as is the
case when one country gains existence benefits from a resource and another
does not.
The third source of differential substitutability is directional. When
resources have a directional flow, the ability of the upstream country to
substitute for the resource at the point at which it is held in common will
almost always be greater than that for the downstream country. A number
of environmental issues involving air, water, and migratory species are likely
to fall into this category.
It remains to be seen how broadly this substitutability hypothesis can
apply. Future research might tell us if it can be usefully applied to pollution
and resource issues, for example, and whether certain types of resource issues
are more likely than others to generate highly differential levels of substi-
tutability. We suggest that the presence of such differential levels can be a
useful predictor of when states will forgo the usual and accepted methods
of managing international environmental resources and instead resort to
unilateral methods in an attempt to enforce a management regime on others.
Chapter 50 Page 20 September 16, 200811:47

20 International Environmental Issues and Debates


1. Kenneth Waltz, Theory of International Politics (New York: Random House, 1979),
p. 109. See chap. 7 for a full discussion of this logic.
2. Edith Brown Weiss, Daniel B. McGraw, and Paul C. Szasz, International Environmental
Law: Basic Instruments and References (New York: Transnational Publishers, 1992); see also
Brown Weiss and Harold K. Jacobson, eds., Engaging Countries: Strengthening Compliance
with International Environmental Accords (Cambridge: MIT Press, 1998).
3. UNEP, Register of International Treaties and Other Agreements in the Field of the
Environment (Nairobi: UNEP, 1991).
4. A similar level of substitutability of a resource for states does not indicate that they
will successfully or nonconflictually manage the resource in question, only that the way they
approach it is likely to be multilateral.
5. For a more thorough discussion of the definition of multilateralism, see John Gerard
Ruggie, Multilateralism: The Anatomy of an Institution, International Organization 46, no. 3
(summer 1992): 561598. Note that our definition includes bilateral cooperation under the
rubric of multilateralism.
6. Elizabeth R. DeSombre, Domestic Sources of International Environmental Policy:
Industry, Environmentalists, and U.S. Power (Cambridge: MIT Press, 2000).
7. The seminal use of the term in international political economy was by Albert Hirschman
in National Power and the Structure of Foreign Trade (Berkeley: University of California Press,
1945), who referred to the substitution of national trading. Our usage differs in that we are
referring to the substitution of specific resources, but the general meaning of the term remains
the same. In the terminology of Robert Keohane and Joseph Nye in Power and Interdependence,
2d ed. (New York: HarperCollins, 1989), we mean the vulnerability dependence of the primary
industry in question to depletion of the resource.
8. We do not attempt to measure substitutability numerically because absolute levels do
not gure into the argument; only relative levels predict unilateralism.
9. Garrett Hardin, The Tragedy of the Commons, Science 162 (13 December 1968):
10. Note that this set of incentives is identical to those identified as Prisoners Dilemma,
which serves as a model for some international relations approaches to understanding incentives
for cooperation.
11. Christopher C. Joyner, Biogeography of the Southern Ocean, in Joyner, Antarctica
and the Law of the Sea (The Hague: Martinus Nijhoff, 1992), pp. 139.
12. Karl-Hermann Kock, Fishing and Conservation in Southern Waters, Polar Record
30, no. 172 (1994): 45; Darryl Powell, Antarctic Fishing and Its Likely Development, in
John Handmer and Martijn Wilder, eds., Towards a Conservation Strategy for the Australian
Antarctic Territory (Canberra: Centre for Resource and Environmental Studies, Australian
National University, 1993), pp. 7589.
13. Ibid.
14. Convention on the Conservation of Antarctic Marine Living Resources (CCAMLR),
done at Canberra, 720 May 1980, entered into force on 7 April 1982. 33 U.S.T. 3476,
TIAS No. 10240, reprinted in John Heap, ed., Handbook of the Antarctic Treaty System
(Washington, D.C.: U.S. Department of State, April 1994), p. 178. Current members of
the CCAMLR Commission include Argentina, Australia, Belgium, Brazil, Chile, European
Economic Community, France, Germany, India, Italy, Japan, Republic of Korea, New Zealand,
Norway, Poland, Russian Federation, South Africa, Spain, Sweden, United Kingdom, and the
United States. States that have acceded to the convention but are not members of the commission
include Bulgaria, Canada, Finland, Greece, the Netherlands, Peru, Ukraine, and Uruguay.
15. Christopher Joyner, Managing Common-Pool Marine Living Resources: Lessons from
the Southern Ocean Experience, in Samuel Barkin and George Shambaugh, eds., Anarchy and
the Environment: The International Relations of Common Pool Resources (Albany: SUNY
Press, 1999).
Chapter 50 Page 21 September 16, 200811:47

Barkin and DeSombre International Fisheries Management 21

16. See, for example, CCAMLR, Report of the Eleventh Meeting of the Commission
(Hobart, Australia: CCAMLR, 1992), pp. 1920.
17. Busy Agenda at Antarctic Fisheries Meeting, Antarctica Project 4, no. 4 (November
1995): 1.
18. Donald W. Bourne, The Fisheries, in Richard H. Backus, ed., Georges Bank
(Cambridge: MIT Press, 1987), p. 406.
19. Convention Respecting Fisheries, Boundary, and Restoration of Slaves, 20 October
1818, TS No. 112, was the first of these. It indicated specific areas where each state was allowed
to fish.
20. Robert DeLello, Fisheries: United States and CanadaConservation or Prosperity,
Suffolk Transnational Law Review 19 (winter 1995): 198.
21. 16 U.S.C. 1801.
22. Michael Knight, Plummeting Prices End New Englands Fishing Boom, New York
Times, 6 July 1980, p. 1.
23. Donna R. Christie, The Georges Bank/Gulf of Maine Boundary Dispute Between the
United States and Canada, in Backus, Georges Bank, p. 469.
24. Joseph Gogh, The Canadian Fishing Industry and Georges Bank, in Backus, Georges
Bank, pp. 465466.
25. Calculated from information in Backus, Georges Bank, pp. 465466.
26. Leah J. Smith and Susan B. Peterson, A Social and Economic View of the New England
Offshore Fishing Industry, in Backus, Georges Bank, pp. 459460.
27. Jan Schneider, The First ICJ Chamber Experiment: The Gulf of Maine Case, the
Nature of an Equitable Result, American Journal of International Law 79 (July 1985): 539;
for a further description of the delimitation, see Christie, The Georges Bank/Gulf of Maine
Boundary Dispute, p. 473.
28. Rushworth M. Kidder, U.S.-Canadian Fishing: Troubled Waters, Christian Science
Monitor, 19 October 1981, p. 1.
29. Debra Percival, Fishing: EU Trawlers Plunder Souths Stocks, Inter Press Service,
21 March 1995.
30. Robert Anderson, Namibia Gets Tough to Defend Its Fishing Grounds, Financial
Times, 30 April 1991, p. 34; EEC/Namibia: Temporary Break in Fisheries Negotiations,
European Report, 20 April 1991, p. 4.
31. Frederick H. Beaudry, William B. Folsom, and David J. Rovinsky, Africa and the
Middle East, vol. 2 of World Fishing Fleets: An Analysis of Distant Water Fleet Opera-
tions: Past, Present, Future (Silver Spring, Md.: Office of International Fisheries, National
Marine Fisheries Service, National Oceanic and Atmospheric Administration, U.S. Department
of Commerce, November 1993), p. 21 (hereinafter cited as World Fishing Fleets). Namibias
fishing fleet, even the vessels capable of fishing on the high seas, fishes entirely within the EEZ
and takes its fish to shore processing facilities.
32. Peter Gruner, The New Spanish Pirates, Evening Standard, 13 April 1995, p. 20
33. Paul Koring and K. Cox, Scots Support Seizure of Vessel, Globe and Mail, 11 March
1995, p. A1.
34. Namibian Court Orders Seizure of Five Spanish Fishing Vessels, Radio Nacional de
Espaa, Madrid, BBC Summary of World Broadcasts, 12 April 1991 (Lexis/Nexis).
35. Canada, Department of Fisheries and Oceans, Backgrounder: Why NAFO Members
Agreed upon a Total Allowable Catch for Greenland Halibut, B-HQ-95-3E, March 1995.
36. The EU joined NAFO as an individual member, so there is an EU quota rather than
quotas for its fifteen constituent members. Notwithstanding this institutional quirk, we will
continue to refer to national quotas.
37. Canada, Department of Fisheries and Oceans, Backgrounder: European Union
Overfishing the Northwest Atlantic, B-HQ-95-E, March 1995.
38. Convention on Future Multilateral Cooperation in the Northwest Atlantic Fisheries,
1978, Article 12(1).
Chapter 50 Page 22 September 16, 200811:47

22 International Environmental Issues and Debates

39. Frank H. Beaudry and William B. Folsom, Executive Summary, in World Fishing
Fleets, p. 5.
40. Edison Stewart, Canadians Seize Spanish Trawler, Toronto Star, 10 March 1995,
p. A1; Jonathan Ferguson, Europeans Charge Piracy, Toronto Star, 10 March 1995, p. A1;
John DeMont, A Partial Victory, Macleans, 1 May 1995, p. 12. Canada agreed to give
approximately 6,000 tons of its quota to Spanish fishers, and in return the EU accepted the
TAC and its division.
41. Treaty Concerning Pacific Salmon, 1985, TIAS 11091.
42. The damming of U.S. rivers on the west coast has made river fishing for salmon much
less feasible; salmon even have to be transported on land past some dams in the region.
43. Gagosz quoted in Sandra L. Lee, Saving the Salmon, Lewiston Morning Tribune, 23
July 1994, p. 5A. See also Janet I-Chin Tu and Danny Wesneat, Albright Joins Fray, Protests
Boat Seizures by Canada, Seattle Times, 29 May 1997, p. A1.
44. Lorraine Woellert, Quest for Salmon Divides Neighbors in the Pacific Northwest,
Washington Times, 1 June 1997, p. A1.
45. Canada Seizes U.S. Salmon Trawlers, Deutsche Presse-Agentur, 27 May 1997
46. U.S.-Canada: Canada Jabs at U.S. with Salmon Quotas, Greenwire, 30 June 1997
47. Canada, U.S. Sign Salmon Pact, Toronto Star, 1 July 1999. The agreement assigns
more of the catch of chinook salmon to the United States and more of the catch of the Fraser
River sockeye salmon catch to Canada.
48. The Tragedy of the Oceans, The Economist (19 March 1994): 24.
49. National Research Council, Dolphins and the Tuna Industry (Washington, D.C.:
National Academy Press, 1992), pp. 1, 13.
50. Alberto Szekely, Yellow-fin Tuna: A Transboundary Resource of the Eastern Pacific,
Natural Resources Journal 29 (fall 1989): 1052. It is not clear that the same groups of fish
migrate all that distance; more likely there are three separate substocks that each keep to a
section of the geographic range. But for practical purposes, that is the area in which these tuna
species are found, and they all do migrate long distances.
51. National Research Council, Dolphins and the Tuna Industry, p. 4.
52. Ibid., p. 15; Alessandro Bonanno and Douglas Constance, Caught in the Net: The
Global Tuna Industry, Environmentalism, and the State (Lawrence: University Press of Kansas,
1996), p. 120; Inter-American Tropical Tuna Commission Annual Report 1966.
53. In fact, as tuna stocks in the Pacific declined and tuna regulation increased, Pacific tuna
fishers would travel through the Panama Canal to fish for bluefin tunaan entirely different
speciesin the Atlantic Ocean during closed Pacific yellowfin seasons.
54. National Research Council, Dolphins and the Tuna Industry, p. 17.
55. James Joseph and Joseph W. Greenough, International Management of Tuna, Porpoise,
and Billf ish (Seattle: University of Washington Press, 1979), pp. 3335; see also Beaudry and
Folsom, Executive Summary, p. 7.
56. Bonanno and Constance, Caught in the Net, p. 136. Since the tuna catch is greatest off
the coast of Mexico, however, most Mexican tuna fishing was likely taking place there. See
Joseph and Greenough, International Management of Tuna, Porpoise, and Billf ish, pp. 2931.
57. These early claims of jurisdiction were actual territorial claims, unlike the later forma-
tion of EEZs.
58. Bonanno and Constance, Caught in the Net, p. 135.
59. See the Federal Register for the years in question; see also Elizabeth R. DeSombre,
Domestic Sources of International Environmental Policy: Industry, Environmentalists and U.S.
Power (Ph.D. diss., Harvard University, 1996), pp. 247248.
60. Fishermans Protective Act (1954), 16 U.S.C. 971. This legislation allows ultimately for
cutting foreign aid budgets for the states in question.
61. Magnuson Fishery Conservation and Management Act (1976), 16 U.S.C. 1825.
62. Bonanno and Constance, Caught in the Net, pp. 124125.
Chapter 50 Page 23 September 16, 200811:47

Barkin and DeSombre International Fisheries Management 23

63. National Research Council, Dolphins and the Tuna Industry, p. 4; some estimates range
as high as 400,000 per year. Mary Russell, House, in Compromise, Eases Porpoise Limit,
Washington Post, 2 June 1977, p. A13.
64. Ronald Mitchell, International Environmental Common Pool Resources: More
Common Than Domestic But More Difficult to Manage, in Barkin and Shambaugh, Anarchy
and the Environment, pp. 2650.
65. After 20 October 1974, no marine mammals could be taken in the course of a commer-
cial fishing operation unless the taking constitutes an incidental catch and is covered by a
permit; these permits were phased out almost entirely. 39 Federal Register 321189.
66. Joseph and Greenough, International Management of Tuna, Porpoise, and Billsh,
pp. 141, 158161.
67. National Research Council, Dolphins and the Tuna Industry, p. 4.
68. For several incarnations of this regulation, see 39 Federal Register 32124; 40 Federal
Register 56904; 42 Federal Register 6455860.
69. See DeSombre, Domestic Sources of International Environmental Policy, chap. 7.
70. Bonanno and Constance, Caught in the Net, pp. 188190. Many elements of the tuna
industry have changed concurrently in a way that also contributed to the protection of dolphins,
although most can be traced back to initial U.S. unilateral action.
Chapter 51 Page 24 September 11, 200819:52

Reducing Pollution of the River
Rhine: The Influence of International Cooperation
Thomas Bernauer and Peter Moser

t is widely believed that the river Rhine is one of the most successful
examples of international cooperation in environmental protection and
the management of international rivers in particular.1 Pollution of the
Rhine decreased dramatically since the beginning of the 1970s while, at the
same time, a web of transboundary agreements developed to provide for
just that a cleaner Rhine. This correlation suggests causation; however, the
connections between international cooperation and reduced pollution are
not straightforward.
In this article we try to disentangle the Rhine story, with the goal of
isolating and bringing into perspective the role that various forms of trans-
boundary cooperation have played in reducing pollution of the river. The
first part of the article reviews international and transnational agreements
aimed at reducing Rhine pollution by heavy metals, which may be regarded
as a typical case; by and large the reduction of other pollutants has followed
the same pattern. The second part discusses the evolution of Rhine pollution
by heavy metals. The third part analyzes the relationship between the evolu-
tion of cooperation and pollution. It does so by examining: (a) processes
by which cooperation may affect the behavior of actors polluting the river;
(b) actors and institutions that may have influenced such processes; and
(c) background variables that need to be controlled when assessing the
impact of transboundary cooperation on river pollution.

Transboundary Cooperation

The International Commission for the Protection

of the Rhine against Pollution

1950 is usually considered the starting point of the modern history of interna-
tional environmental cooperation among the riparian countries of the Rhine
(Kiss, 1985, p. 621). At that time, a coordinating body, the International
Commission for the Protection of the Rhine against Pollution (ICPR), was

Source: The Journal of Environment & Development, 5(4) (1996): 389415.

Chapter 51 Page 25 September 11, 200819:52

Bernauer and Moser Reducing Pollution of the River Rhine 25

established, following several initiatives by the Netherlands (Goppel, 1991,

p. 4).2
The Netherlands draws a large share of its freshwater supply from the
Rhine. Also, it is located at the mouth of the river and is thus unable to
influence directly the quality of water flowing into the country from outside
its jurisdiction hence its strong interest in a clean Rhine.
The ICPR was reconstituted in 1963 by a formal treaty among the
riparian states: Switzerland, the Federal Republic of Germany, France, the
Netherlands, and Luxembourg. The latter is connected to the Rhine basin
through the river Moselle (see Figure 1). The functions of the ICPR were
quite narrowly defined. Its tasks were to monitor the nature, extent, and
sources of pollution; propose measures for the protection of the Rhine; and
to prepare agreements among the parties. The functions of the secretariat
of the ICPR (located in Koblenz, Germany) are those of an advisory body
that has no decision making power of its own. Decisions by the ICPR must
be taken by consensus, which tends to produce agreements at the lowest
common-denominator level. Indeed, the ICPR and its secretariat did not play
a very active role during the 1960s and early 1970s and were unable to exert
political pressure on the riparian governments to cut their pollution of the
river (LeMarquand, 1977). The ICPR concentrated merely on the first of
its three functions during the first 10 years of its existence, gathering and
publishing information on the pollution of the Rhine.3

The Annual Rhine Environmental Ministers Conference

Some progress was made in 1972, when on the initiative of the Netherlands
the Annual Rhine Environmental Ministers meeting was instituted. These
meetings have become the most important international forum for the discus-
sion and resolution of Rhine water quality issues (LeMarquand, 1977). It
seems that this step exerted a reviving influence on the ICPR by giving it
clearly defined objectives. In the following year, the Rhine Ministers meeting
directed the ICPR to elaborate a convention regulating chemical pollution
of the Rhine.
France was among the major advocates of a convention on chemical
pollution. The motives for the French position may have been the low costs
that cuts in chemical pollution would entail for French industry, and the
desire to shift the political pressure for environmental protection by France
to other actors. Since the 1930s, France had been accused of being the biggest
chloride polluter of the Rhine (Bernauer, 1995b). Germany, by contrast,
would bear the heaviest burden if chemical pollution was curbed because it
has the largest industrial base along the Rhine.

Convention for the Protection of the Rhine against Chemical Pollution

These renewed efforts resulted in the signing of the 1976 Convention for
the Protection of the Rhine against Chemical Pollution, the so-called Bonn
Chapter 51 Page 26 September 11, 200819:52

26 International Environmental Issues and Debates

Figure 1: Map of the Rhine Basin


Rotterdam Arnhelm


Sieg t
B Bonn
Me r



SELTZ Karlsruhe




Rekingen = Measurement Stations
B = Belgium

NL = The Netherlands
F = France CH
D = Germany

CH = Switzerland
L = Luxemburg

Convention. Its early existence was fraught with difficulties (ICPR,

19751976). The signing of the accord was retarded by 1 year because, due
to problems unrelated to the agreement, there was no ministerial conference
in 1975 (ICPR, 19751976). The treaty entered into force in 1979. This
delay resulted from concerns of the German Federal Republic that its costs
would be too high.
Chapter 51 Page 27 September 11, 200819:52

Bernauer and Moser Reducing Pollution of the River Rhine 27

The approach used in the Bonn Convention is heavily influenced by the

European Communitys (EC) directive on Pollution Caused by Certain
Dangerous Substances Discharged into the Aquatic Environment ([19]76/
464/EEC [European Economic Community]) (Romy, 1990, p. 54).4 Like the
corresponding EC directive, the Bonn Convention is a framework agreement.
It outlines the general aims of the environmental protection effort, which are
to reduce pollution of the Rhine. It defines black and gray lists of substances
(in Annex I and Annex II) that are to be controlled through different
procedures (see Grjins & Mostert, 1992). It directs the ICPR to elaborate
proposals for threshold values for particular substances and branches of
industry. Only in a second stage would these proposals become binding
rules (once they are adopted unanimously by the contracting parties); that is,
national emission permits could not exceed the agreed thresholds. The Bonn
Convention expanded the geographical scope of cooperation by providing
for the application of the agreement to the entire Rhine basin.

The Rhine Action Program

In October 1987 the Rhine Ministers Conference adopted the Rhine Action
Program (RAP). The reasons are twofold: (a) difficulties with the implemen-
tation of the Bonn Convention (see below); and (b) an accident in November
1986 at the Swiss chemical firm Sandoz that led to a widely publicized
spill of toxic substances into the Rhine, and to a corresponding flurry of
international activity. In contrast to the Bonn Convention, which was effec-
tively replaced by the new agreement, the RAP is not a formal treaty but a
gentlemens agreement. It covers a broader range of Rhine environmental
issues and its approach is different, one might say rather more pragmatic.
The overarching goal, symbolizing the health of the ecosystem as a whole,
is the reintroduction of salmon in the Rhine and its tributaries. As in the
earlier Bonn Convention, the reduction of chemical pollution remains the
centerpiece of the agreement. Specific water quality targets were set, taking
as a baseline the year 1985. A 50% reduction of the total load of priority
substances (among which are heavy metals) over the entire course of the
river should be reached by 1995.5 To achieve the reduction targets, the ICPR
can make recommendations as to the waste-water treatment applicable in
specific industries. Such recommendations have been issued for the cellulose
(1992), waste-water management (1991), and organic chemistry industries.
The member states of the ICPR have to ensure the application of the best
available technology (BAT) and to report to the ICPR about their activities
(see ICPR annual reports). However, as the ICPRs recommendations are
not legally binding, their monitoring and enforcement is based on informal
control (ICPR 1992, p. 30). Additionally, the RAP encompasses not only
point sources but also diffuse sources. (Grjins & Mostert, 1992, p. 35; press
communiqu of the eleventh Rhine Ministers Conference, December 1994).
To achieve the RAPs goals, the Rhine riparian nations agreed to a three-
stage process. The first stage was devoted to fact finding, particularly the
Chapter 51 Page 28 September 11, 200819:52

28 International Environmental Issues and Debates

establishment of detailed national registers of important dischargers. During

the second stage, from 1989 to 1995, the proposals made by the ICPR were
to be put into effect. In the third stage, from 1995 to 2000, the remaining
problems were to be ironed out.

Transnational Cooperation

A relatively new and innovative transboundary measure aimed at reducing

pollution of the Rhine involves nonstate actors, such as firms or munici-
palities. The most widely publicized such case relating to Rhine pollution
by heavy metals involves the city of Rotterdam and chemical industries in
Germany, Switzerland, and France.
The City of Rotterdam is under the obligation, imposed by public law, to
keep its harbor navigable and thus has to remove sediments from it. Because
Rotterdams harbor lies at the mouth of the Rhine, pollution of the river
leads to a contamination of sediments. Dutch and international law forbids
the dumping of contaminated matter into the North Sea, so the sediments
dredged from Rotterdams harbor have to be disposed of in a special reposi-
tory, a so-called slufter. This slufter, built in the mid-1980s at a cost of about
200 million Dutch guilders, will be at full capacity by the year 2000. To
avoid construction of another slufter, heavy metal loads in the Rhine will
have to be reduced substantially, by roughly 70% to 90%. The municipality
of Rotterdam therefore tried to identify the principal dischargers of heavy
metals along the Rhine and to conclude agreements with them. To enhance
its bargaining position, Rotterdam threatened to pursue legal action against
upstream polluters. In return for pollution reductions by upstream polluters,
Rotterdam offered to drop its liability claims for the costs it incurred as
a result of the contaminated harbor sediments. The most important legal
premise backing the credibility of the threat is the fact that Rotterdam could
file a lawsuit either in any upstream state where the pollution originated or
downstream where the damage occurred.6 (For a detailed discussion of the
legal issues, see van Dunn, 1990, 1992; Kernkamps, 1992.)
After extensive research, the Rotterdam authorities began to negotiate
with two Swiss, four French, and seven German companies.7 These negotia-
tions have so far resulted in several agreements. Yet other negotiations seem
to be under way. Characteristically these contracts contain clauses regarding
the time span during which certain emissions are to be reduced, and to what
extent. In the case of a breach of the contract, the claim for tort by the munic-
ipality of Rotterdam will be revived. Table 1 gives an overview of major
agreements, the ICPR and the Sandoz spill.

The Evolution of Rhine Pollution

Until the end of the 1960s pollution of the Rhine increased steadily, with peak
loads of heavy metals in the late 1960s and early 1970s (Stigliani, Jaff, &
Chapter 51 Page 29 September 11, 200819:52

Bernauer and Moser Reducing Pollution of the River Rhine 29

Table 1: Overview of Major Agreements, the ICPR, and the Sandoz Spill

Institution/Incidents Date Brief Description

International 1950 The ICPR started meeting in 1950 and was

Commission for the formally constituted in 1963. Its members
Protection of the are France, Germany, Luxembourg, The
Rhine Against Netherlands, Switzerland, and (since
Pollution (ICPR) 1976) the European Union. Its secretariat
is located in Koblenz, Germany.
Bonn Convention December 3, Framework agreement under which
1976 agreements on emission thresholds for
individual substances and industries are
to be worked out. Even though the ICPR
prepared recommendations for nine
substances, agreement among the ICPR
members was reached only for cadmium
and mercury. Since the early 1990s, the
implementation of the Bonn Convention
has quietly ceased.
Rhine Action Program October 1, This politically binding framework
1987 agreement among the ICPR members
aims at reducing emissions from direct
and diffuse sources as well as risks due
to accidents, and improving the Rhines
ecosystem as a whole. This agreement is
based on a best available technology
approach and is to be implemented in
three phases until the year 2000. The
secretariat of the ICPR can recommend
measures on how to implement the
Rotterdam agreements 1990s Agreements under private law between the
municipality of Rotterdam and chemical
firms in upstream countries. The aim of
these agreements is to reduce pollution of
the river by substances that settle in the
Rotterdam harbor and have to be
dredged and disposed of by Rotterdam at
high costs.
Sandoz spill November 1, In the process of ghting a major fire,
1986 millions of liters of water were sprayed
on a warehouse of Sandoz, a Swiss
chemical firm, at Schweizerhalle near
Basel. About 10,000 to 15,000 cubic
meters of this heavily contaminated water
(insecticides, fungicides, pesticides)
entered the Rhine. The spill had
devastating effects on the rivers fauna
(e.g., 150,000 dead eels) and led to the
interruption of drinking water supply and
irrigation systems downstream.
Chapter 51 Page 30 September 11, 200819:52

30 International Environmental Issues and Debates

Anderberg, 1993, p. 790). Since then, pollution has declined significantly.

The pattern of this decrease is essentially similar for all heavy metals and
indeed other chemical pollutants. We can observe a very steep decrease from
the peak years until about the mid-1980. Since then, the decline has slowed
down with concentrations stagnating at relatively low levels (ICPR, 1994;
van Dunn, 1992, p. 117). Figure 2 shows, as an example, pollution by
Most international efforts to reduce pollution of the Rhine have concen-
trated on point sources. At the same time, pollution by point sources has
declined significantly more than pollution by diffuse sources (Stigliani et al.,
1993, p. 790). To conclude from this correlation that international cooper-
ation must have influenced pollution would, however, be premature.
In looking closer at pollution originating from point sources, it is inter-
esting to note that the iron, steel/coke, and metal refining industries, which in
their peak years were major point sources for all three pollutants examined
by Stigliani et al. (1993), had reduced their emissions dramatically by 1988.
For instance, aqueous cadmium emissions discharged by the metal refining
industries had amounted to almost 40% of the total cadmium load of the
Rhine in the peak year. By 1988, they were reduced to zero. Stigliani et al.
(1993) attribute this progress to the installation of waste-water treatment
as a result of more stringent national regulation and the implementation
of good housekeeping practices. By-products of production processes (such
as cadmium in zinc refineries) that used to be dumped into the Rhine are
now considered valued resources. Finally, metal-producing industries recycle
their waste and have replaced older, more wasteful technologies with cleaner
Other factors, which may have contributed to pollution reductions,
include industrial restructuring processes in the Rhine area, particularly in

Figure 2: Mean Annual Cadmium Loads in the Rhine at Bimmen-Lobith 1975 to


Cd. [mg/l]

1975 1977 1979 1981 1983 1985 1987 1989 1991

Source: Bundesministerium fr Umwelt Naturschutz und Reaktorsicherheit 1995: Abb. 5a

[Figure 5a].
Chapter 51 Page 31 September 11, 200819:52

Bernauer and Moser Reducing Pollution of the River Rhine 31

the context of declining coal and steel production. Whether, and if so how
and to what degree, transboundary cooperation has influenced these devel-
opments will be examined in the next section.

Hypotheses and Methodology

Hypothetical Effects

In the simplest case, a specific transboundary regulation will either be directly

applicable at the domestic level, or will be transformed into national law. To
avoid the consequences of breaching the law, polluters will react to new regu-
lations by changing their behavior, for example by installing waste-water
treatment facilities or using cleaner production technology.8 In the case of
agreements that are not legally binding, such as the RAP, the cause-effect
relationship could work in a similar way. The recommendations by the ICPR
serve as guidelines for national regulations that influence the behavior of
the relevant polluters. In both cases one should expect that Rhine pollution
decreases after the conclusion of transboundary agreements.
In addition to direct effects, such as those just discussed, transboundary
cooperation may have several more indirect effects. The prospect of future
regulation may cast its shadow on the present behavior of polluters. If,
for instance, the expected conclusion of an international agreement makes
stricter regulation probable in the future, polluters might change their
behavior in advance to spread their investment in pollution reduction over
a longer time span and gain a competitive advantage. The same effect may
occur if the shadow stems not from new national and international legisla-
tion, but from the threat of costly lawsuits against polluters. In addition, as
more and more companies believe that stricter regulations will be adopted,
greater competition for a greener image may set in. In all these cases,
we should expect pollution reductions even before the conclusion of trans-
boundary agreements.
The question of time lags underlines the fact that the direction of
causal relationships is crucial to the analysis because it provides prelimi-
nary answers as to whether transboundary cooperation has had an effect on
pollution reduction. It will be noted that causal relationships between trans-
boundary and national and actor-specific pollution reduction efforts can also
work the other way around: National or actor-specific efforts can determine
cooperation at the transboundary level in that states only agree to reduction
measures that have already been agreed on or implemented at the domestic
level. In this case, no new obligations exceeding domestic standards are
created at the transboundary level and, therefore, transboundary efforts have
not affected domestic efforts. The same holds for lawsuits. In the Rotterdam
case, for example, some analysts have claimed that upstream chemical firms
signed on to reduction targets that they knew they were going to implement
irrespective of Rotterdams efforts. Even in such cases, however, we may find
Chapter 51 Page 32 September 11, 200819:52

32 International Environmental Issues and Debates

that transboundary efforts can have other types of effects on the behavior of
polluters at the domestic level, including the following.
First, transboundary cooperation may lead to changes in domestic power
structures. For example, governments or, at a lower level, environmental
agencies that run into implementation problems with domestic regulations
on water pollution may use transboundary obligations to increase their pres-
sure on polluters and the interests supporting them. This strategy may facil-
itate the implementation of agreed or planned domestic measures.
Second, the additional information generated by transboundary cooper-
ation can foster pollution reductions in at least three ways. It can render
riparian actors more willing to engage in cooperation on pollution reduc-
tions because it raises transparency. Incomplete information on present and
future preferences and behavior of the actors in a given group is widely
regarded as one of the major impediments to cooperation. This applies in
particular to cases where pollution reductions by one actor are contingent
on reductions by another actor (Morrow, 1994). Information generated by
transboundary efforts can be used by environmental lobby groups, water-
works, or political authorities to exert pressure on national governments,
individual firms, or other actors (shaming these actors into pollution reduc-
tions). Finally, information generated through transboundary cooperation
can result in new options for reducing pollution at lower cost, which in
turn increases the willingness of polluters to comply with more stringent


Even the preliminary and superficial comparison of important events, such

as the entering into force of a treaty or a follow-up regulation, and the
development of pollution over time (see the Evolution of Rhine Pollution
section) shows that immediately obvious connections between any of them
are lacking. We thus decided that a qualitative approach to the analysis of
our research question was the most appropriate. Additionally, from a policy
perspective, this approach allows us to trace in detail the causal chains that
connect transboundary cooperation with polluter behavior and may thus
yield practical knowledge on which type of effort affects polluters in what
manner, or, conversely, when and why transboundary efforts fail.9

The Effects of Transboundary Cooperation


In the Rhine case, a very important, if not the most important effect of
transboundary cooperation was that institutionalized exchanges of informa-
tion among national government agencies involved in implementing envi-
ronmental standards enhanced their problem-solving capacity by increasing
their knowledge. The Rhine regime, defined as the totality of the interna-
Chapter 51 Page 33 September 11, 200819:52

Bernauer and Moser Reducing Pollution of the River Rhine 33

tional agreements and organizations, thus acted as a learning facilitator. At

present, the governmental authorities responsible for the implementation of
environmental legislation in the riparian states form a close-knit community.
This community shares a coherent view regarding not only the extent and
origins of Rhine pollution, but also the available technological means to deal
with problems as they occur.
This community did not spring into life immediately. Its evolution was
on the contrary a lengthy process. It began in the mid-1950s, when the
first annual pollution data, covering only a few parameters, were published
by the ICPR. These data were collected by national measurement stations.
During the 1960s, information exchanges were restricted to annual pollu-
tion gures collected at the principal national measurement stations along
the Rhine. This approach may have raised the general awareness of Rhine
pollution problems. But it does not seem to have significantly influenced the
behavior of national water agencies, which at that time were weak. It should
be noted that during this early period the ICPR was not dominated, as it is
today, by technocrats but largely by diplomats. Contacts intensified after the
first (1973) meeting of the Rhine riparian nations environmental ministers,
which asked the ICPR to negotiate a convention on chemical pollution.
The 1976 Bonn Convention added an important layer to the flow of infor-
mation. This development led to the further enhancement and unification of
measurement programs.10 It also institutionalized international discussions
about threshold values for particular polluting substances. Therefore, even
though the results of this development in terms of legally binding agree-
ments were rather poor, these information exchanges facilitated discussions
on sensible threshold values for polluting substances and the technical possi-
bilities of dealing with the problems at hand. In the framework of the 1986
Rhine Action Programme, the riparian states must supply information on
the pollution of the Rhine, including data on individual firms and their emis-
sions. They must also indicate for individual substances the reductions that
are possible when applying BAT. Beforehand, such detailed information was
not available to many of the national authorities. The requirement to furnish
this information to the ICPR and the other parties thus made it necessary
for at least some riparians to allocate more resources to the collection and
processing of such information.11
This increase in transboundary information flows has resulted in a
homogenization of perceptions of Rhine pollution problems, and, perhaps
equally important, common views about the available technical means
to solve these problems. The various treaty provisions concerning infor-
mation gathering and exchange thus had the important effect of binding
together national authorities directly involved in Rhine pollution issues. They
helped create a closely connected community of low- to mid-level govern-
ment officials and scientists. This group has met largely in the working
groups of the ICPR to exchange information about new forms of pollu-
tion, polluters, sudden increases in pollution due to accidents, measurement
methods, pollution abatement technologies, and so forth. As a result, it seems
that new information and knowledge has spread virtually as quickly across
Chapter 51 Page 34 September 11, 200819:52

34 International Environmental Issues and Debates

as within national boundaries. Moreover, this development appears to have

strengthened national government agencies dealing with Rhine pollution vis-
-vis coalitions representing the interests of polluting actors, for example,
industry associations, or industry and transportation ministries.
Information networks, such as the one just discussed, have also developed
at the nongovernmental level. The water purification plants in the Rhine
catchment area are under the obligation to secure specific water quality
standards. Not surprisingly, therefore, they have been very much interested
in the quality of the Rhine water they use and measure the latter regularly.
This information is cleared through a nongovernmental organization, the
International Association of Waterworks in the Rhine Basin (IAWR, 1995).12
The IAWR serves to identify polluters and exert pressure on them as well as
on national legislators or, as it seems to be increasingly the case, the European
Commission in Brussels.

Strengthening of Environmental Government Agencies

International obligations concerning national measurement programs and

requirements (for example in the RAP) not only led to a dissemination of
information, in some cases they even seem to have mobilized additional
resources for the reporting agencies. In addition, the regime seems to have
acted as an agent of internal realignments (Levy, 1993) in other ways as
well. Environmental agencies have been able to use international commit-
ments made by their governments as arguments in support of their (often
costly) goals vis--vis other branches of the bureaucracy in intergovern-
mental bargaining processes about resources and policy goals and also the
polluters themselves.
One of the most interesting examples of an environmental agency that
used the existence of an international commitment to enhance its own
standing domestically is the Agence de leau Rhin-Meuse, situated in Metz.
The French water-management authorities, which are organized by catch-
ment area, are designed to act as financial intermediaries between polluters
and consumers of water. They subsidize, but cannot initiate, water improve-
ment projects through fees levied from polluters. In this particular case, the
agency in Metz seems to have tripled the emission levies in its domain by
referring to the French Governments commitment to the RAP agreement.
Besides improving water quality, the increased levies also made the agency
grow and the latter seems to serve as a model which other French water agen-
cies seem keen to emulate. It is a telling detail that the director of the Agence
de leau Rhin-Meuse is now head of the department in the environmental
ministry in Paris.

Anticipation of Regulation by Polluters

It appears that transboundary efforts have exerted more direct positive effects
on the behavior of polluting industries. In particular, the establishment of new
programs for measuring pollution of the Rhine, the commitment to pollution
Chapter 51 Page 35 September 11, 200819:52

Bernauer and Moser Reducing Pollution of the River Rhine 35

thresholds in the Bonn Convention, and the ensuing discussions about them
in the 1970s and 1980s have cast a shadow ahead. These developments were
interpreted by actors potentially affected as a sign of further tightening of
environmental regulation in the future. Hence these actors reacted by
reducing their pollution in advance of anticipated legally binding laws by
installing waste treatment plants or transferring especially polluting activities
out of the Rhine catchment area. The same logic applies to the mandatory
national registers of important polluters and discussions about international
BAT standards after the adoption of the RAP. The intense interest with
which the big chemical firms located along the Rhine have observed devel-
opments in the framework of the ICPR makes this claim at least plausible. It
would also account for the fact that heavy metal loads in the Rhine started
to decrease before international agreements entered into effect.
It is, however, difficult to isolate the exact role of the Rhine regime in this
respect. Even if one concedes that polluters anticipate future legislation (we
will discuss this point further below), it may be difficult to determine what
these polluters actually reacted to: was it developments in the ICPR, national
measures, measures by the European Union (EU), and to what extent does
each of these potential driving forces account for changes in the behavior of
polluters? Because the trend toward stricter environmental regulation seems
to be a universal and largely parallel phenomenon at all levels in highly
industrialized countries, it is very difficult to sort out individual effects. An
additional problem is that the importance of individual factors may vary over
time and across countries or actors. However, the history of environmental
protection efforts by the Rhine riparians suggests the following.
In the 1970s, the focus of attention was mainly national. This applies
in particular to firms located in countries that initiated comprehensive
environmental protection programs during this period as a consequence
of expanding domestic green movements, especially in the Netherlands,
Germany, and Switzerland. This situation has changed somewhat in the
1980s when tangible discussions about pollution thresholds started in the
context of the Bonn convention, and when the EC became increasingly
involved in environmental protection. In recent years, notably, the European
Commission seems to have gained in importance. One of the indications
is the fact that not only international industrial associations but even big
national associations, such as the association of German water treatment
plants, are represented in Brussels. Their aims are to obtain information
about future EU directives and also influence the contents of such directives
as much as possible.

Direct Impacts of Transboundary Measures

The Bonn Convention

Most experts agree that the implementation of the Bonn Convention has
been disappointing, even in a strictly legal sense, let alone the actual effect
on pollution loads. The ICPR prepared nine recommendations (ICPR, 1981;
Chapter 51 Page 36 September 11, 200819:52

36 International Environmental Issues and Debates

Kiss, 1985, p. 628),13 some of which deal with heavy metals. But in the
end only two of these proposals were adopted by the riparian countries:
thresholds for mercury in the waste water of the chloro-alkali electrolysis
industry (1982) and, for a slightly wider range of industries, thresholds for
cadmium concentrations (1986) (Goppel, 1991).14 These measures coin-
cide with the ECs implementation directives mentioned above, particularly
regarding the substances and sources covered by the 1976 directive On
Pollution Caused by Certain Dangerous Substances Discharged into the
Aquatic Environment (76/464/EEC). This parallel development is due to
the fact that Germany, especially, made progress in the Rhine catchment
area conditional on agreement in the wider context of the EU (see below).
Another sign of the ineffectiveness of the Bonn Convention is that by the
time the follow-up agreements on pollution thresholds for particular indus-
tries had entered into force, they were to a large extent already obsolete.
They had been anticipated by independently started programs by national
regulatory authorities and possibly the polluters themselves. The approach of
the convention was obviously not well adapted to the fast-changing circum-
stances prevalent in Rhine environmental matters. Notably, it established
a procedure much too cumbersome and inflexible to keep pace with the
decrease of pollution for reasons unrelated to the international effort and
the development of new technologies.
Consequently, there is reason enough to assume that the Bonn Conven-
tion did not generate any important new obligations, either for national
authorities or polluters. The legalistic makeup of the convention probably
reflects an earlier stage of international environmental cooperation, at least
in Western Europe.15 In the 1970s when environmental protection was just
beginning to move up on political agendas, the Rhine riparian governments
tried to keep the decision-making power on Rhine pollution issues at a polit-
ical level, so as to control even minute technical matters and also to avoid
any costly over-commitment. The greening of politics, the parallel growth
of environmental authorities (in terms of their resources and bureaucratic
power) and the ensuing pressure on polluters led to rapid improvements in
environmental protection technology and cleaner production processes. All
of this created a new situation in which the approach of the Bonn Convention
became virtually useless because its results tended to come too late.
In summary, the ineffectiveness of the Bonn Convention largely stems
from its design in particular its cumbersome procedures for negotiating and
implementing concrete pollution reduction measures paired with fast devel-
opments in pollution reduction at the national level, which at that time were
technically easily achievable. Whether the riparian governments would have
been able in the 1970s and 1980s to conclude and implement a more effective
approach remains an open question. In any event, however, Rhine pollution
by heavy metals and other substances decreased throughout the 1980s, but
this decrease was driven by forces other than the Bonn Convention. This
ineffectiveness of the convention has not escaped the riparian countries. By
the end of the 1980s, the ICPR had clearly reduced its efforts to implement
Chapter 51 Page 37 September 11, 200819:52

Bernauer and Moser Reducing Pollution of the River Rhine 37

the agreement; and in 1992 the implementation of the convention seems to

have been quietly dropped altogether (ICPR, 1992).16


The evaluation of the RAP is somewhat more complicated than that of the
Bonn Convention because in the case of the RAP it becomes more difficult
to apply the straightforward top-down model of legislation (international
rules are first established and then implemented domestically). Negotiations
in ICPR working groups between the representatives of national environ-
mental authorities and negotiations at the national level between polluters
and national authorities have proceeded in a parallel and highly interac-
tive fashion. The reason for this is that environmental authorities in most
riparian states have in recent years emerged as relatively powerful branches
of government. It also stems from the less formal or diplomatic and more
technical approach to dealing with problems in the framework of the ICPR,
which has evolved over time (see above). Finally, the lack of pressure origi-
nating from the need for ratification, which would involve higher political
levels of government, has also contributed to more flexible and informal
forms of cooperation between actors at various levels.
The resulting BAT recommendations therefore do not have the character
of internationally enacted laws, fed subsequently into a national imple-
mentation process, but are rather some kind of well-agreed-upon statements
of intention of all the parties concerned, including the dischargers them-
selves. The latter are usually few in number for each given industry targeted
by a particular BAT recommendation. In the paper and cellulose industries,
which were among the first to be targeted, only about six dischargers were
actually involved.
The RAP seems to have been effective in several respects. First, it
intensified the flow of information. Second, by setting reduction targets
within strict time frames and elaborating BAT recommendations it created
clear points of reference for all the parties. Third, the lack of treaty status
accelerated agreement on the RAP in the first place, and made it possible
to get quick results in terms of BAT recommendations. The time gained
in preparing these soft recommendations probably outweighed the protrac-
tion in their implementation due to their not legally binding nature. That,
according to the Communiqu of the 11th Rhine Ministers Conference, the
goals set particularly in the heavy metals domain were not fully reached
has probably much to do with their strictness and also with the fact that it
becomes increasingly difficult and costly to bring about even small additional
pollution reductions (increasing marginal abatement costs).

The Rotterdam Agreements

The agreements between the Rotterdam municipality and various chemical

firms along the Rhine have been lauded by some analysts especially and
Chapter 51 Page 38 September 11, 200819:52

38 International Environmental Issues and Debates

understandably in the legal community as a new and potentially more

effective approach to reducing pollution of the Rhine than the intergovern-
mental approach within the framework created by the ICPR. The informa-
tion we obtained through interviews suggests, however, that the effect of the
Rotterdam effort on Rhine pollution has been rather small.
The reduction targets in these agreements did not exceed in a substantial
way the commitments already pledged to national authorities in the frame-
work of the RAP. The argument that the Rotterdam agreements added the
extra fear of lawsuits against polluters if agreed goals were not met, and
therefore changed the polluters behavior, should be taken with a grain of
salt. One needs to bear in mind that legal battles of this kind are usually
extremely costly for both sides, whereas the outcome is by no means clear.
Lawsuits in the context of chloride pollution of the Rhine have shown that
it is extremely difficult to quantify pollution damages and attribute this
damage to the emissions of particular actors. Indeed, only in one single
case have lawsuits resulted in a clear-cut verdict and, ultimately, compensa-
tion payments (Bernauer, 1995b). This problem reduces the effectiveness of
the threat of legal action, and therefore also the effect of the threat on the
behavior of polluters (if we assume that polluters know what we have just
said!). This leaves the possibility, however, that the behavior-altering fear
of the polluters can originate not primarily in the prospective damages that
might have to be paid to Rotterdam, but rather in the potentially negative
effect on the image of industries in the case of highly publicized lawsuits.


Governments and Environmental Authorities

The analysis of causal pathways has shown that the effects of cooperation
cannot be explained merely in terms of unitary state actors, that is, govern-
ments as entities. If cooperation influences polluter behavior, it does so
because lower, more technically oriented layers of government such as water
authorities are able to advance their own interests by employing international
commitments, such as the Bonn Convention, the RAP, or institutions like
the ICPR. The involvement of governments as such is neither a sufficient nor
a necessary condition for the effectiveness of the transboundary efforts we
examined. The role of governments is rather to create institutions in the first
place, and to establish commitments as to the goals to be attained. The Rhine
environmental ministers conferences, especially the three conferences held
after the Sandoz incident, have clearly served this purpose. Whether these
commitments can trigger a process that leads to changes in the behavior of
polluters is highly dependent upon other circumstances, especially the agility
of lower levels of government.
The important, perhaps even decisive stage of the process is entered
after international commitments have been established. The problems
encountered in implementing the Bonn Convention indicate that the
Chapter 51 Page 39 September 11, 200819:52

Bernauer and Moser Reducing Pollution of the River Rhine 39

institutionalized inclusion of higher levels of government (mainly in terms of

the need for ratification of follow-up agreements) may be counterproductive
(at least under conditions prevalent in Western Europe; we will come back to
this point). Moreover, if governments as such have become active, they have
usually done so largely to placate public outrage after some spectacular inci-
dent such as the Endosulfan accident in the late 1960s or, more recently, the
Sandoz incident of 1986 the latter led to three Rhine ministers conferences
within one year.
One of the key results of our analysis is that environmental authori-
ties in the riparian countries have probably been the most important cata-
lysts in respect to the domestic impact of transboundary cooperation. For
these actors, international agreements have not constituted obligations to be
shunned wherever possible, but on the contrary, represent an opportunity to
strengthen their own role in various ways. They have managed to enhance
their domestic position by extending their knowledge base through informa-
tion exchanges in the ICPR, and by using international commitments as an
instrument to enhance their domestic bargaining position.
However, the importance of transboundary cooperation in this respect
should not be overestimated. Sufficiently independent and resourceful
administrative structures at the domestic level, which are capable of using
international cooperation to their own advantage, must probably exist in the
first place. International cooperation may enhance their position, but cannot
usually create them. The most important variable determining their strength
is probably the degree of environmental consciousness among the domestic
population. In the end, it is this consciousness that creates the political pres-
sure that forces governments to react by enacting environmental laws and
establishing the bureaucratic structures to implement and enforce them.

Nongovernmental Actors

One of the more striking results of our research is that many of the larger
sources of Rhine pollution, and particularly the big chemical companies such
as Sandoz and Roche in Basel, or Hoechst, Bayer, and BASF in Germany, are
no longer passive and reluctant targets of environmental regulations. To the
contrary, it appears that they have increasingly cut back on their emissions
in anticipation of future regulation and have sought to influence regulatory
processes well before any legislation has been passed.
A telling sign of this development is the fact that in 1992 the German
chemical firm Bayer was awarded the so-called WRK-Rhine-Prize by the
IAWR for its exemplary efforts in the field of waste-water treatment (IAWR,
1995, p. 99). This distinction had hitherto been awarded to scientists or civil
servants engaged in environmental protection. This development probably
implies that the difficulty of implementing regulations on Rhine pollution
has, at least for the larger point sources, eased considerably: The atmosphere
among the key actors relevant to Rhine pollution seems to be one of reluctant
cooperation rather than one of insuperable antagonism.
Chapter 51 Page 40 September 11, 200819:52

40 International Environmental Issues and Debates

This (modestly) cooperative attitude of important parts of industry along

the Rhine seems to contradict the traditional picture of the company as
a single-minded profit maximizer: Measures to prevent pollution from
reaching the environment are usually costly, so one would expect them to be
taken at the last possible moment, just before the cost of sanctions surmounts
the cost of installing, say, a pollution treatment facility. The inconsistency of
observed behavior of industry with this model can be explained as follows.
First, one has to remember that the big companies do not operate in a
societal vacuum. With the possible exception of France, the population of all
Rhine riparian countries has since the beginning of the 1970s grown increas-
ingly conscious of environmental matters. This development has created a
public opinion climate in which being ostracized as a polluter can impose
very heavy costs on industry, as the chemical firm Sandoz had to learn after
the fire in Schweizerhalle in 1986. The reverse side is that a green image
can be an important asset for public relations. Considerable investments of
chemical and other companies in the projection of a green image through
advertisements and other public relations instruments testify to this point.
These developments are also reflected in organizational changes within
firms. Beginning in the 1960s the latter began to establish nonproduction-
related departments for pollution abatement to cope with increasing envi-
ronmental regulatory activity at all political levels. These departments have
grown in importance over time and have often been staffed by environmental
scientists who have the same educational background as their opponents
in state or international bureaucracies. The fact that these people have been
part of the same scientific community and have thus often shared a common
outlook and values on Rhine pollution issues has probably facilitated
The waterworks along the Rhine, the second type of nongovernmental
actors to be discussed here, have been organized in the IAWR. The IAWR,
whose secretariat is located in Amsterdam, is an umbrella association
whose members are associations of waterworks in the Netherlands (RIWA),
Germany (AWR, AWBR), Switzerland, and France. It represents 113 water-
works in the Rhine catchment area. The waterworks can be regarded as a
particular type of industry, one that operates under quite special conditions
which, in turn, create unique incentives. These waterworks have been under
pressure from strict legislation defining thresholds for polluting substances
in their output. This legislation has been increasingly enacted at the EU
level.17 In addition, the waterworks have also been under pressure by their
customers, and private households in particular, which are very sensitive
to the quality of drinking water: We were told that a request in a Cologne
restaurant for a glass of plain water from the tap would meet with incred-
ulous stupefaction, even though this water conforms with public health
Because the room to maneuver for the improvement of technology for
drinking water production is, for technical and cost reasons, rather small and
the directors of waterworks are personally responsible for the quality of their
Chapter 51 Page 41 September 11, 200819:52

Bernauer and Moser Reducing Pollution of the River Rhine 41

output, the waterworks have a very strong incentive to improve the quality
of their water input. At least in certain locations, where their water input
stems directly from the Rhine or wells in the immediate vicinity, notably in
Germany and the Netherlands, the waterworks are strongly affected by the
extent of Rhine pollution, and therefore by emissions upstream. Not surpris-
ingly, therefore, the waterworks have organized themselves internationally
early in the 1970s (IAWR, 1995, p. 97) and engaged in lobbying efforts at
various political levels. They also sought to identify the sources of particu-
larly harmful substances and put pressure on them to stop their emissions.
They have done so through direct negotiations with the firms concerned.
For the reasons mentioned above, many of these negotiations have been
successful, even if they were not highly publicized.
The Municipality of Rotterdam is, as we outlined above, in a very similar
situation as the waterworks as far as its interests are concerned. It differs,
however, in the approach it has used to put pressure on the polluters. Instead
of the rather low-key approach of the IAWR it has led a highly publi-
cized campaign in a style somewhat resembling that of some environmental
nongovernmental organizations (NGOs), such as Greenpeace. In addition
it has relied quite heavily on legal means of threat and coercion. Whether
this approach has been more effective than the approach used by IAWR for
example, is difficult to pin down.
The differences in the approach between Rotterdam and IAWR can prob-
ably be explained by the fact that the Rotterdam municipality, as a polit-
ical body responsible to its own population/voters, experienced a greater
need than IAWR to engage in more tangible efforts to demonstrate to its
constituency that it was active in trying to avert the costly building of another
slufter. One should also consider that, at the time the Rotterdam project was
initiated, it was by no means clear what we know now, that considerable
pollution reductions were lying ahead.

The EU

The influence of the EU as an actor involved in the clean up of the Rhine

is difficult to gauge. Moreover, its role seems to have changed during the
period in question. In the 1970s, EU environmental policy was hampered by
substantial differences among member states about the approach to be used.
The emission approach was favored by Britain, for example, whereas the
emission approach was favored by Germany and the other Rhine riparian
nations (Grijns & Mostert, 1992, p. 28). Member states also held different
positions as to the standards to be applied. There are indications that the
introduction of the EU as an additional contracting party to the ICPR and the
Bonn Convention, both in 1976, slowed down the implementation process.
The principal reason was that some countries (mainly Germany) made the
conclusion of follow-up treaties to the emission-oriented Bonn Conven-
tion conditional on the adoption of equivalent directives within the wider
Chapter 51 Page 42 September 11, 200819:52

42 International Environmental Issues and Debates

framework of the EU. But steps in this direction were blocked by Great
Britain until 1983.
The fact that the 1976 EC-directive on Pollution caused by Certain
Dangerous Substances Discharged into the Aquatic Environment (76/464/
EEC) was mirrored by the Bonn Convention therefore seems to have been
rather unhelpful. The directive on the quality of drinking water (80/778/
EEC), another major piece of potentially relevant EC legislation, has no
counterpart within the cooperative framework of the Rhine riparians. It is
generally acknowledged that the threshold values set by this directive are
very strict (even water treatment plants in highly developed countries such
as Germany were at first unable to fulfill them). This led to implementation
Unlike regulations, directives address not individuals but member states.
As a result, they need to be transformed into national regulations by the
member states to become binding upon individuals and corporate actors.
It seems that member states implemented them with varying degrees of
success.18 Because the drinking water directive regulates water quality at the
very end of the pipe, its influence on polluter behavior and pollution of the
Rhine is necessarily indirect. The existence of this directive has, however,
served as an argument for the waterworks (which are directly concerned) in
their lobbying activities vis--vis polluters and governments.
This rather ambiguous picture notwithstanding, the fact that a growing
number of relevant international interest groups (e.g., chemical and pesticide
industry associations) and even national interest groups (e.g., the German
association of waterworks) are present in Brussels seems to indicate that
Brussels has become increasingly important as a focus of environmental regu-
latory activity in the Rhine basin.

Background Variables

To complete the analysis, and to be able to draw some more general conclu-
sions that may be applicable to other cases, we have to come back to some
background variables that were so far referred to only in passing.

Environmental Awareness

The rising degree of environmental awareness in the population of all Rhine

riparian states during the 1970s and 1980s, which is indicated by the growth
of green NGOs and political parties, was probably the most important
driving force behind the spectacular clean up of the Rhine in general, and
the international component we are interested in here in particular. It would
go beyond the scope of this article to explain this development and its
precise effects on efforts to curb pollution of the Rhine. However, this
section touches on a few aspects in which rising environmental awareness
among constituencies may have spurred pollution reductions. First, growing
environmental awareness forced governments to do something about Rhine
Chapter 51 Page 43 September 11, 200819:52

Bernauer and Moser Reducing Pollution of the River Rhine 43

pollution, namely, to enact national laws for the protection of the environ-
ment. These separate but parallel national efforts to clean the rivers in the
Rhine catchment area are certainly the main contributor to pollution reduc-
tion in the Rhine as a whole.
Second, to cope with the implementation of this legislation, national envi-
ronmental authorities had to be created or strengthened. For the reasons
discussed above, the existence of entrenched and resourceful environmental
agencies is probably an important precondition for making international
cooperation fruitful and effective as well as facilitating the implementa-
tion of national law. Although international cooperation, as we have argued
above, may have strengthened these agencies backs, they already needed to
have strong backs in the first place.
Third, public outrage or concern, especially about spectacular incidents of
a boundary transcending character such as the Sandoz spill, is conditional on
preexisting environmental awareness. Such incidents have thus been impor-
tant triggers for international agreements and declarations by governments.
These declarations and agreements have then served as starting points for
the processes analyzed above.
Fourth, environmental awareness among populations has changed the
conditions under which polluting firms operate (see above). Polluter
behavior tends to be strongly penalized by the public at large, whereas the
projection of a green image is an important asset. The big chemical firms
especially are under close scrutiny by environmental NGOs and the press;
larger-scale discharges (be it on a regular basis or due to an accident) rarely
go unmonitored and uncommented.

Similarity of Structural Conditions

The second group of background variables pertains to structural conditions.

All Rhine riparian countries are Western European democracies at a high
level of socioeconomic development. The industries causing chemical pollu-
tion are more or less evenly distributed among these countries, and their
environmental standards at the outset of transboundary pollution reduction
efforts were about the same. This circumstance has facilitated cooperation
and made the latter perhaps paradoxically less urgent, because neither
the costs nor the benefits of pollution abatement measures would accumu-
late asymmetrically across the riparian countries. There is, in other words,
no clear-cut distinction between those who suffer from pollution and those
who cause it: All riparian countries are to a certain extent both victims and
If, for instance, Germany takes measures to improve the water quality of
the Rhine or its catchment area (covering a substantial part of the country),
this is seen by its environmentally conscious population as being to its own
advantage, as well as having the side effect of benefiting the Netherlands
downstream. Although no water is taken directly from the Rhine for drinking
water purposes in Germany, water from adjacent wells is used, whose quality
Chapter 51 Page 44 September 11, 200819:52

44 International Environmental Issues and Debates

depends critically on Rhine water quality.20 The same is true for Switzerland,
where many of the lakes in the Rhine catchment area are used for drinking
water production.
An additional problem, which prevents states from turning an interna-
tional river into a gutter, with the goal of minimizing the countrys costs
of pollution reductions is the following. A country would have to treat
dischargers on its territory differently. Those at or near the point where the
river crosses the border (e.g., Basel in Switzerland) would have to be allowed
to discharge without limit whereas those in more sensitive regions inland
would have to pay for treatment facilities. In a state where information
flows freely, and where the principle of equality before the law is generally
accepted, such a solution would meet strong opposition from those who are
by their geographical location under the obligation to reduce their emissions
and to bear the costs.
Another structural condition that facilitated the cleanup of the Rhine
was the high growth rates prevalent particularly in the chemical industry
along the Rhine during the 1970s and 1980s. High profits certainly made
the installation of often costly waste-water treatment facilities and other
pollution reduction measures easier.
Taken together, these background variables created a favorable environ-
ment for cleaning up the Rhine. One could imagine this cleanup happening
even without transboundary cooperation, although it would probably have
happened somewhat slower.


Pollution of the river Rhine by heavy metals has declined substantially since
the 1970s. At the same time, we can observe an extensive web of trans-
boundary political and legal activity designed to reduce such pollution. In
this article we have made an attempt at analyzing whether, and if so how,
these two developments are causally related. In other words, did the observed
transboundary efforts contribute to curbing pollution of the Rhine, and if
so, to what extent and how?
The analysis in this article does not produce a simple answer. Causal
chains leading from transboundary efforts to the behavior of polluters and
ultimately to river pollution itself are long and multifaceted. The trans-
boundary efforts whose impact we examined qualitatively include interna-
tional negotiations and agreements aimed exclusively at reducing pollution
of the Rhine, negotiations and agreements between nonstate actors, and
legislative activity by the European Community/Union (EC/EU).
Our analysis shows that transboundary efforts have influenced polluter
behavior in four different ways. First, these efforts have intensified the
exchange of information at various levels (intergovernmental, transnational,
EC/EU). This exchange ultimately created a tightly knit community of
people with a general knowledge of the causes of pollution and methods
Chapter 51 Page 45 September 11, 200819:52

Bernauer and Moser Reducing Pollution of the River Rhine 45

to deal with the problem. Second, transboundary efforts were used by

domestic environmental agencies to strengthen their position with regard
to other government agencies and the polluting actors. Third, the extensive
transboundary political and legal activity with regard to Rhine pollution
led polluters to reduce their emissions in anticipation of future regulation
(domestic and transboundary). Fourth, the legally binding output of trans-
boundary efforts contributed to changing the behavior of polluters. Our
analysis suggests that the first and second types of effects were probably
much more pronounced than the other two.
In drawing any general conclusions from this analysis, we have to keep in
mind two groups of background conditions: First, environmental awareness
in all riparian countries of the Rhine has been relatively high, at least since
the 1980s; second, all Rhine riparian countries are at a high level of socioe-
conomic development. We think, however, that the following hypotheses
might well be useful with regard to efforts for cleaning up other international

One should not expect that the traditional approach to international

environmental politics international treaties and protocols, transla-
tion of these agreements into domestic law, implementation of these
laws will necessarily work better than softer approaches, such as
action programs or transnational policy efforts. This applies in partic-
ular to relatively homogeneous and environmentally advanced countries
in which a stable legal and administrative framework already exists.
Liability rules may contribute toward pollution reductions, but they are
no panacea. Because of difficulties involved in identifying sources of
pollution, estimating damages, and attributing particular portions of the
damage to individual polluters are enormous, the effect of threats of
legal action on polluters often tends to be small. A different case may
be pollution originating from accidents (such as the Sandoz case), where
the source of pollution is easily identified and damages are usually quite
straightforward. Furthermore, public pressure on a firm to reach a face-
saving settlement is usually higher in these cases.
International and transnational negotiations can contribute to learning
processes in which decision makers develop common perceptions and
approaches to dealing with river pollution problems.
International or transnational commitments can strengthen domestic
government agencies favorably disposed toward pollution reductions.
They cannot create such agencies, but can enhance their position vis--
vis government agencies or sources of pollution less favorably disposed
toward cleaning up the river.
A further lesson to be learned from the Rhine case is that background
conditions that are largely outside of the scope of international cooper-
ation are highly important, and even crucial, as they shape the contents
and the forms of international cooperation and its effectiveness much
more than the reverse.
Chapter 51 Page 46 September 11, 200819:52

46 International Environmental Issues and Debates


1. See Stigliani, Jaff, and Anderberg (1993, p. 786); Neue Zrcher Zeitung [a Zrich
newspaper], March 17, 1990, p. 5. For more critical views, see LeMarquand (1977, p. 123),
who attributes progress in cleaning up the Rhine more to domestic political pressure than to
international cooperation.
2. A treaty, concluded in 1885 and regulating salmon fishery along the Rhine, provided for
occasional meetings of government representatives. It served as a framework for the meetings
in which Rhine pollution was first discussed at the end of the 1940s. These meetings ultimately
led to the decision to establish a special commission that would exclusively deal with Rhine
pollution (Romy, 1990, p. 52).
3. The monitoring program of the ICPR was gradually expanded over the years. A contin-
uous measurement program at all Rhine monitoring stations (as compared to only 26 samples
before) has been in place since 1980 (ICPR, 1980).
4. A representative of the European Union (EU) Commission usually takes part in the
Annual Rhine Ministers Conference. The EU is also a member of the ICPR. The EUs activities
in the field of water protection will not be discussed here; for details see Bernauer and Moser,
5. In 1991 the reduction targets for lead and cadmium were increased to 70% (ICPR,
1994). There is, however, a clause in the program that allows for exceptions from the 50% rule
for pollutants that were heavily reduced in the years before 1987 and whose emission is already
subject to best available technology (BAT).
6. This possibility originates in a decision by the European Court of Justice in connection
with a key case in transboundary environmental liability: the case of Dutch nursery firms against
the Mines de Potasse dAlsace (MdPA) (Alsatian potassium mines) over MdPAs discharges of
chloride into the Rhine (Bernauer, 1995b).
7. The companies chosen were those whose discharges of certain pollutants exceeded 1%
of their total annual load at Lobith in the reference year 1985 (van Dunn, 1992, p. 122).
8. Social science theories offer various explanations of why individuals or groups behave
in compliance with norms or rules. For example, actors may comply because they fear costly
punishment (sanctions). They may also comply out of a habit of obedience. See Bernauer
9. The following analysis is based on a critical synthesis of the very sparse secondary liter-
ature relevant to the area of our study and, more important, interviews with a selected group
of experts, including Dr. Schulte-Wlwer-Leidig (Secretariat of the ICPR, Koblenz, Germany),
Dr. Michael von Berg (Federal Ministry for the Environment, Bonn, Germany), Prof. J. Wessel
(Delft Technical University, Delft, Netherlands), Mr. Mark Uilhoorn (Erasmus University,
Rotterdam, Netherlands), Mr. Karel Dieperink (University of Utrecht, Utrecht, Netherlands),
Dr. Walter Jlich (IAWR, Amsterdam, Netherlands), and Mr. Edwin Mller (Department for
the Environment, Bern, Switzerland). In compliance with a request made by most of our inter-
view partners, we do not attribute particular pieces of information to individuals.
10. Bonn Convention, Articles 8 through 13.
11. The fact alone that these measurements (and the improvement of measurement
programs in general) have existed seems to have affected the behavior of polluters in a favorable
manner. As they could no longer assume that the origin of a particular polluting substance
would be unknown to the authorities, they had a strong incentive to curb their emissions.
12. Due to the larger number of measurement stations, the information produced by this
nongovernmental network is geographically more fine grained than the information coming
from the official measurement stations; it thus seems to allow a quite exact identification of
13. According to the ICPR annual reports, the ICPR prepared threshold values for 12
substances, whereas only two were adopted.
14. For details, see ICPR (1985).
15. It will be noted that at the global level the combination of framework conventions with
follow-up protocols is still very popular (see, for example, the case of climate change).
Chapter 51 Page 47 September 11, 200819:52

Bernauer and Moser Reducing Pollution of the River Rhine 47

16. In a last-ditch effort, the ICPR decided in 1990 that the formal ratification of its recom-
mendations was to be replaced by an accelerated procedure, at least for recommendations that
were modeled on European Community (EC) regulations (ICPR, 1991, p. 31). But this does
not seem to have resulted in tangible progress.
17. See, for example, the EC drinking water directive of 1980.
18. The unsatisfactory implementation of EC environmental legislation seems to be a
widespread problem, at least if one uses the large number of legal cases against member states
as an indicator (Grijns & Mostert, 1992, p. 29; Vogel, 1993).
19. The example of chloride pollution by the Mines de Potasse dAlsace shows that this may
indeed be an important factor; in that case the costs of reducing pollution accumulated mainly
in France, whereas the Netherlands benefited from it. The tortuous history of the international
efforts to solve this particular problem indicates that similarity in this respect matters (Bernauer,
20. See Bundesministerium fr Umwelt Naturschutz und Reaktorsicherheit (1995, p. 7).


Bernauer, T. (1995a). The effect of international environmental institutions: How we might

learn more. International Organization, 49(2), 351377.
Bernauer, T. (1995b). The international nancing of environmental protection: Lessons from
efforts to protect the river Rhine against chloride pollution. Environmental Politics, 4(3),
Bernauer, T., & Moser, P. (1996). Reducing pollution of the Rhine river: The inuence of inter-
national cooperation. (IIASA Working Paper No. 967). Laxenburg, Austria: International
Institute for Applied Systems Analysis.
Bundesministerium fr Umwelt, Naturschutz und Reaktorsicherheiti [Federal Ministry for the
Environment, Environmental Protection and Reactor Security]. (1995). Umweltpolitik: Der
Rheinein europischer uss, Beispiel fr erfolgreiche Gewsserschutzpolitik [Environ-
mental policy: The RhineA European river as an example of successful water protection
policy]. Bonn, Germany: Author.
Goppel, J. M. (1991). Legal aspects of international water management. Article Prepared for
Waterscapes  91, an international conference on water management for a sustainable devel-
opment, Saskatoon, Canada.
Grijns, L. C., & Mostert, E. (1992, January). The Rhine after 1992: Towards a method
for comparing and weighing two conicting interests (Working Paper No. 1). Delft, the
Netherlands: RBA Centre for Comparative Studies on River Administration, Delft Univer-
sity of Technology.
IAWR [International Association of Waterworks in the Rhine Basin]. (1995). Rheinbericht
[Rhine report]. (pp. 9193). Amsterdam, Netherlands: Author.
ICPR [International Commission for the Protection of the Rhine against Pollution].
(19751976). Ttigkeitsberichte [Annual report]. Koblenz, Germany: Author.
ICPR. (1980). Ttigkeitsberichte [Annual report]. Koblenz, Germany: Author.
ICPR. (1981). Ttigkeitsberichte [Annual report]. Koblenz, Germany: Author.
ICPR. (1985). Ttigkeitsberichte [Annual report]. Koblenz, Germany: Author.
ICPR. (1991). Ttigkeitsberichte [Annual report]. Koblenz, Germany: Author.
ICPR. (1992). Ttigkeitsberichte [Annual report]. Koblenz, Germany: Author.
ICPR. (1994). Vergleich der gewssergte des Rheins mit den zielvorgaben 19901993
[Comparison of the Quality of Rhine Water 19901993]. Koblenz, Germany: Author.
Kernkamps, F. H. (1992, October). The Rotterdam Rhine contracts. In J. M. van Dunn (Ed.),
Environmental contracts and covenants: New instruments for a realistic environmental
policy (pp. 111115). Proceedings of an international conference held in Rotterdam, the
Kiss, A. (1985). The protection of the Rhine against pollution. Natural Resources Journal, 25,
Chapter 51 Page 48 September 11, 200819:52

48 International Environmental Issues and Debates

LeMarquand, D. G. (1977). International rivers: The politics of cooperation. Vancouver,

Canada: Westwater Research Centre.
Levy, M. A. (1993, March 2327). The effectiveness of international institutions: What we
think we know, and how we might learn more. Paper presented at the annual convention
of International Studies Association, Acapulco, Mexico.
Morrow, J. D. (1994). The forms of international cooperation. International Organization, 48,
Rhein-Ministerkonferenz [Rhine Ministers Conference]. (1994, August 12). Communiqu
[Press release].
Romy, I. (1990). Les pollutions transfrontires des eaux: L exemple du Rhin, Moyens daction
des lss [Transboundary water pollution: The example of the Rhine, means of action for
the damaged]. Lausanne, Switzerland: Payot.
Stigliani, W. M., Jaff, P. R., & Anderberg, S. (1993). Heavy metal pollution in the Rhine basin.
Environmental Science and Technology, 27(5), 786793.
van Dunn, J. M. (Ed.). (1990). Transboundary pollution and liability. Rotterdam, the
Netherlands: Institute of Environmental Damages, Erasmus University Rotterdam.
van Dunn, J. M. (Ed.). (1992, October). Legal issues arising from the Rhine contracts.
Environmental contracts and covenants: New instruments for a realistic environmental
policy. (pp. 117141). Proceedings of an international conference held in Rotterdam, the
Vogel, D. (1993). Environmental policy in the European Community. In S. Kameniecki (Ed.),
Environmental politics in the international arena: Movements, parties, organizations, and
policy (pp. 181197). Albany: State University of New York Press.
Chapter 52 Page 49 September 11, 200819:53

The European Union as an Environmental
Governance System
Regina S. Axelrod, Norman J. Vig and Miranda A. Schreurs

he creation of the European Union (EU) has transformed western
Europe. The addition of ten new member states on May 1, 2004,
advances the political integration of Europe even further. The objec-
tive of establishing a common internal economic market has contributed
to the openness of national borders and the harmonization of many poli-
cies once in the exclusive domain of individual member states. The EU also
has established some of the strongest and most innovative environmental
protection measures in the world and has increasingly taken the lead on inter-
national environmental issues such as global warming. In principle, environ-
mental protection now enjoys equal weight with economic development in
EU policymaking.
Political will and public support have been the keys to EU success in
approaching the environment from an integrated perspective. First, the legal
foundations have been firmly established so that the EU has an unchallenged
right to protect the environment. Second, all states recognize that without
common environmental policies, barriers to free trade develop. Third, polit-
ical, economic, and geographic diversity have challenged policymakers to
develop innovative strategies for overcoming differences and sharing burdens
The EU is therefore an important model to study, both as the most
advanced regional organization of states and as a comprehensive environ-
mental policy regime. But the EU also has become an important actor in
global environmental diplomacy. In addition to individual member states,
the EU is a party to most international conventions negotiated in the 1990s.1
Since the 1992 United Nations Conference on Environment and Develop-
ment, the EU has played a leading role in pushing the United States and
other nations to adopt more stringent environmental agreements on matters
such as climate change. As a result of enlargement, the EU will play a role in
shaping environmental policy from the Baltic to the Aegean. New member
states are required to transpose into their national systems the entire body of
EU laws, regulations, and directives, known as the acquis communautaire.

Source: Regina S. Axelrod, David L. Downie and Norman J. Vig (eds), The Global Environ-
ment: Institutions, Law, and Policy, 2nd edition (Washington: CQ Press, 2004), pp. 20024.
Chapter 52 Page 50 September 11, 200819:53

50 International Environmental Issues and Debates

The chapter for the environment in the acquis communautaire is known as

the environmental acquis. The EU will face a number of important problems
and opportunities as it becomes a larger and more integrated governance
system. This chapter explores the history, institutions, current environmental
policies, and future challenges of this unique organization.

The Political Origins of the European Union

and Environmental Policy

The quest for political and economic union in Europe has its origins in the
1920s and 1930s, when it was recognized that some kind of supranational
organization was needed to avoid brutal competition, protectionism, and
war. But it was the experience of World War II that convinced statesmen
to seek a new type of unity. U.S. economic assistance under the postwar
Marshall Plan also called for regional cooperation.
The first step toward building a more integrated Europe was the forma-
tion of the European Coal and Steel Community (ECSC). The idea of French
economic planner Jean Monnet and foreign minister Robert Schumann, the
ECSC was created by the Treaty of Paris on April 18, 1951. The orig-
inal members were Belgium, France, Germany, Italy, Luxembourg, and the
Netherlands. Its economic goal was to pool the production of coal and steel
for the benefit of all six countries. Its other purpose was to lock Germany
politically and economically into a stable partnership with western Europe.
Other cooperative activities were slow to develop, but in June 1955 the six
ECSC members decided to move toward closer economic integration. They
saw a European free trade area or common market as a means to increase
industrial and agricultural exports, to redistribute resources to economi-
cally depressed areas, and to encourage travel among countries. The result
was the 1957 Treaty of Rome, which established the European Economic
Community (EEC) and the European Atomic Energy Authority (Euratom).
In the 1970s Denmark, Ireland, and the United Kingdom joined the EEC,
and Greece, Portugal, and Spain followed suit by 1986. Austria, Finland, and
Sweden became full members in 1995, bringing the membership to fifteen.2
In 2004 the membership of the EU expanded to twenty-five, with the acces-
sion of Cyprus, the Czech Republic, Estonia, Hungary, Latvia, Lithuania,
Malta, Poland, the Slovak Republic, and Slovenia. Bulgaria and Romania
will be candidates for accession at a later date.3
The Treaty of Rome contained no explicit provisions for protection of
the environment. EEC policy on the environment dates from the 1972 Paris
summit of the Communitys heads of state and government, which was
inspired in part by the United Nations Conference on the Human Environ-
ment held earlier that year in Stockholm. Under Article 235 of the Treaty
of Rome, which permits legislation in new areas if consistent with EEC
objectives, the summit proposed the creation of an Environmental Action
Programme, in effect adding an environmental agenda to the Treaty of Rome.
Chapter 52 Page 51 September 11, 200819:53

Axelrod et al. EU as an Environmental Governance System 51

Over the next dozen years, the EEC adopted three environmental
action plans (for 19731976, 19771981, and 19821986, respectively) and
enacted more than twenty major environmental directives covering air and
water pollution, waste management, noise reduction, protection of endan-
gered flora and fauna, environmental impact assessment, and other topics.
It took most of these actions first under Article 235 of the Treaty of Rome
and later under Article 100a (added to the treaty by Article 18 of the Single
European Act), which authorizes actions that directly affect the establish-
ment or functioning of the common market. The motivation for these laws
was to avoid trade distortions caused by different environmental standards
while dealing with problems that were inherently transboundary in nature.4
The next milestone in the development of the treaties was the Single
European Act of 1986, which accelerated the integration process by calling
for establishment of a single internal economic market by the end of 1992.
Equally important, the act added a new section to the Treaty of Rome that
formally defined the goals and procedures of EEC environmental policies
and called for balanced growth by integrating environmental policy into
all other areas of decision making. These goals and procedures can be found
in Articles 174, 175, and 176 of the Consolidated Version of the Treaty
Establishing the European Community (Consolidated Treaty), which was
published in 2002. The volume of new environmental legislation reached a
peak during the 19871992 period.
The Maastricht Treaty (also called the Treaty on European Union), which
entered into force in 1993, advocated closer political and monetary union,
including development of a common European currency. The euro was put
into circulation by twelve member states in January 2002. The Maastricht
Treaty also created two new pillars to promote common foreign and
security policies and cooperation in justice and home affairs that, together
with the first pillar, the European Community (EC, until this time called the
EEC), now form the EU. The ability of the EU to forge a common foreign
and security policy has been sorely tested in relation to both Kosovo and
Iraq. It has been more successful in the environmental realm.
In addition, the treaty further strengthened the legal basis and procedures
for environmental policymaking. This trend was continued with revisions
made by the Treaty of Amsterdam in June 1997. Article 3d of the Amsterdam
Treaty (which became Article 6 of the Consolidated Treaty) states explicitly
that environmental protection requirements must be integrated into the
definition and implementation of Community policies and activities    in
particular with a view to promoting sustainable development.5
Expansion of the EU was addressed by the Treaty of Nice, which came
into effect on February 1, 2003. Although the treaty deals primarily with
the effects of enlargement on EU institutions, it also reaffirms the Unions
commitment to environmental policy. At the December 2000 Nice confer-
ence that led to the treatys formation, a declaration was adopted affirming
member states determination to see the European Union play a leading role
in promoting environmental protection in the Union and in international
Chapter 52 Page 52 September 11, 200819:53

52 International Environmental Issues and Debates

efforts promoting the same objective at the global level.6 Further changes
are likely in the future, especially if a constitution is eventually agreed upon.7
The EU has created the most comprehensive regional environmental
protection regime in the world. Although scholars still debate whether the EU
is primarily an intergovernmental organization dominated by the inter-
ests of individual member states or a functional regime that represents
common transnational interests and actors, it is increasingly regarded as a
multi-level governance structure in environmental policy.8

EU Institutions and Policymaking Processes


The EUs primary institutions are the European Council and the Council
of Ministers, the European Commission, the European Parliament, and the
European Court of Justice (ECJ). There are also fifteen secondary agencies,
including the European Environment Agency (EEA).9
The European Council and the Council of Ministers are often simply
called the council. Technically, when the heads of government meet, which
they must do at least twice a year, it is known as the European Council or
summit. Broad policy directions are set at these summits. More specific
policy decisions are made in meetings of the relevant Council of Ministers.
Thus, for example, environmental decisions are reached by the Council of
Environment Ministers, which as the name suggests is made up of member
states environment ministers. The presidency of the council rotates among
the member states every six months, and the country in charge can shape
the agenda of all council meetings. The council is the most important EU
body because it must approve all legislation. Its directives must be adopted
by the individual member states and incorporated into national law within
a specified period of time, usually two years. The EU can enact regulations
that automatically apply to the states, but they are less common. The council
also makes decisions on international treaties and agreements. In general,
the councils actions reflect the national interests of the states. The Single
European Act introduced qualified majority voting on some environmental
matters. This is a special procedure that gives greater weight to states with
large populations but protects smaller states by requiring a majority of votes
(71 percent) for a measure to pass. The Treaty of Nice assigned new voting
weights to member states but kept the 71 percent rule. It also extended
qualified majority voting to thirty other areas.
The European Commission is a body of commissioners (and their staffs)
who head twenty-three directorates-general (DGs). With enlargement, the
number of commissioners is to be kept below twenty-seven. DG XI, now
called the Environmental Directorate, is responsible for the environment,
nuclear safety, and civil protection. The commissions task is to initiate
EU legislation and to oversee its implementation by member states. The
Chapter 52 Page 53 September 11, 200819:53

Axelrod et al. EU as an Environmental Governance System 53

Treaties of Amsterdam and Nice strengthened substantially the powers

of the commission president. Romano Prodi has held this position since
1999. A multinational bureaucracy of somewhat less than 20,000 serves the
commission and its directorates in Brussels.10
The European Parliament, in contrast, is elected directly by voters in
each country and tends to reflect the diverse interests of political parties
and groupings across Europe. The Treaty of Nice restricts the number of
parliamentarians in the enlarged EU to a maximum of 732. The parlia-
ment holds plenary sessions in Strasbourg, France, but much of its staff is in
Luxembourg, and most of its committee meetings are held in Brussels. Draft
legislation from the commission is submitted to the parliament, which can
either accept the draft as is or propose amendments. The parliament also
must approve the commission budget and EU treaties, and it votes on the
appointment of the president and commissioners.11
The European Parliament is not regarded as a true legislature because it
cannot initiate measures. However, the Maastricht Treaty allows a majority
of members to request that the commission develop a proposal if it concerns
implementation of the treaty, and parliament committees also can infor-
mally influence policy formation in other ways. Under the Amsterdam Treaty
revisions, a co-decision procedure was extended to many more areas of
environmental legislation; if the parliament does not agree with the council
position after a second reading, a conciliation committee is formed to resolve
differences. If agreement still cannot be reached, the parliament can reject
the proposal by majority vote, giving it a de facto veto. The new proce-
dure has increased the parliaments role in policymaking, resulting in more
transparent decision making and reducing the so-called democratic deficit.
The European Court of Justice, located in Luxembourg, considers cases
brought before it by the commission, the council, or member states con-
cerning the application of EU treaties. ECJ decisions are binding on member
states, but the ECJ must depend on national courts to carry them out, making
enforcement difficult. Under the Maastricht Treaty, the ECJ can levy fines
on member states that fail to comply with its decisions. The ECJ also has
made it easier for citizens to enforce their EU rights in national courts.12
Some ECJ decisions have helped define the rights of member states
to enact environmental legislation that may violate EU treaty provisions
prohibiting restraints of trade. For example, in the 1988 Danish bottle case,
the ECJ upheld Denmarks law requiring the use of returnable bottles for beer
and soft drinks on the grounds that its environmental benefits were sufficient
to justify a minor restraint on trade.13 Like courts in the United States, the
ECJ is emerging as an important policymaker in balancing economic and
environmental interests. Significantly, the acquis communautaire includes
the judgments passed by the ECJ.
The European Environment Agency, approved in 1990, was established
in 1994 in Copenhagen after a long battle over its location. Although it
does not have the regulatory and enforcement powers of the Environmental
Protection Agency in the United States, the EEA is becoming an important
Chapter 52 Page 54 September 11, 200819:53

54 International Environmental Issues and Debates

actor in EU policymaking. Its membership includes, in addition to the EU

member states, the three European Economic Area states (Iceland, Norway,
and Liechtenstein) and Bulgaria, Romania, and Turkey. Its mission is to
compile a scientific database on environmental conditions in Europe and
develop analytical models for understanding environmental processes and
improving decision making. It collects and distributes data through the Euro-
pean Environment Information and Observation Network (EIONET), which
consists of about 300 environmental bodies, agencies, and research centers.14

The Policy Process

Policymaking within the EU is more political than a description of the

institutions might suggest.15 Because the EU is a fluid and developing insti-
tution, policymaking is complicated by uncertainty over roles, powers, and
decision rules. As we have seen, the council, commission, and parliament
perform functions different from those of the three branches of the U.S.
The commission and the parliament can be viewed as supranational
bodies, whereas the council remains essentially intergovernmental. The
commissioners and their staffs are international civil servants who are
not supposed to serve any national interest; therefore, the commissions
proposed legislation tends to favor greater harmonization of European
policies. Parliament also tends to favor stronger EU policies, especially in
fields such as environmental and consumer protection that are popular with
the electorate. The council, in contrast, is usually more cautious because of
its sensitivity to national political interests and the costs of implementing
EU policies (which largely devolve on national governments). The council is
more likely to invoke the principle of subsidiarity, under which actions are
to be taken at the EU level only if they cannot be carried out more efficiently
at the national or local level.16
Conflicts of interest among the states are evident in the council. In the
past, a fluid coalition of Denmark, Germany, and the Netherlands pushed
the hardest for environmental protection. These green countries often had
higher regulatory standards than the EU and tried to get the EU to adopt these
standards a process that has been called regulatory competition.17 For
example, Germany was influential in proposing tough air pollution controls
on large combustion plants, while the Netherlands convinced the council to
adopt its high standards for small car and truck emissions. More recently,
Austria, Finland, and Sweden have joined Denmark and the Netherlands as
the environmental leaders, with Germany taking a somewhat more cautious
position because of economic difficulties at home.18 At the other end of
the spectrum, the poorer countries, such as Greece, Ireland, Portugal, and
Spain, have been more reluctant to carry out EU environmental policies,
while Belgium, the United Kingdom, France, Italy, and Luxembourg tend to
fall in between. One of the principal challenges facing the EU has been to
find ways of accommodating different levels of environmental commitment
Chapter 52 Page 55 September 11, 200819:53

Axelrod et al. EU as an Environmental Governance System 55

and regulatory capacity without weakening ultimate goals. This will become
an even bigger challenge because of the accession of the central and eastern
European states, which are much poorer and have far less experience with
environmental protection programs than the older member states.
Lobbying by private interests is also omnipresent in the EU.19 Industry
is very concerned about the impact of new environmental legislation on
competitiveness and maintains an army of lawyers in Brussels. Both the
commission directorates and parliamentary committees regularly consult
such interests, which tend to represent the largest companies and trade asso-
ciations. Environmental, consumer, and other public interest groups also
have representation. An umbrella organization in Brussels, the European
Environmental Bureau, represents over 140 environmental organizations
from the EU, accession states, and some neighboring countries.20 It closely
monitors DG XI and tries to influence proposed legislation. Other interna-
tional environmental nongovernmental organizations (NGOs), such as the
Worldwide Fund for Nature and Greenpeace, also lobby intensely and are
regarded as among the most effective pressure groups.21 A broad range of
stakeholders and policy networks influence the EU policy process at all levels.
Environmental policy is closely related to other issues such as economic
competition, taxation, research and development, energy, agriculture, and
transportation.22 Effective policymaking therefore requires interaction and
cooperation among many EU directorates and parliamentary committees.
For example, the development of efficiency standards for electrical appli-
ances involved a working group of members from the environment and
energy directorates. The divergent perspectives of these directorates often
lead to different policy preferences. The requirement introduced by Article
3d of the Amsterdam Treaty that environmental protection must be inte-
grated into all fields of EU policy gives weight to the issue. As relatively new
and understaffed, however, the environment directorate is often in a weak
position in negotiating with its counterparts, especially the leading economic
directorates for industry and trade. Final policy resolution by the commission
and council usually involves extensive political compromise, which some-
times takes the form of side payments. For example, in an effort to gain
approval for an overall EU target for reducing carbon dioxide, a burden-
sharing plan was worked out, under which some states agreed to exceed
EU targets so that other states would have lower burdens to meet. Such
differentiated obligations, along with derogations allowing some coun-
tries more time to comply with EU directives, are creating a multispeed
Europe despite efforts to integrate and harmonize policies.

Harmonization of Environmental Standards

One general rationale for creating common environmental policies and

harmonizing standards across member states has been to level the
economic playing field. A danger associated with this, however, is that the
Chapter 52 Page 56 September 11, 200819:53

56 International Environmental Issues and Debates

lowest common denominator will prevail and, in the case of environmental

standards, result in community norms that are considerably weaker than
those of the leading states. To mitigate this problem, the Single European
Act added to the Treaty of Rome three new articles: 130r, 130s, and 130t.
These are now articles 174, 175, and 176, respectively, of the Consolidated
Treaty. Article 174 guarantees that the EU will take action for preserving,
protecting, and improving the quality of the environment, protecting human
health, prudent and rational utilization of natural resources, and promoting
measures at (the) international level to deal with regional or worldwide
environmental problems. Article 175 allows the council to decide which
measures can be decided by qualified majority voting. Article 176 specifies
that protective measures taken at the community level shall not prevent any
Member State from maintaining or introducing more stringent protective
measures so long as these measures are compatible with treaty law.23
Lead states are able to retain higher environmental standards than other
countries so long as the commission or the ECJ does not find them in viola-
tion of other treaty rules. For example, half of the EU 15 introduced national
carbon taxes (Denmark, Finland, Germany, Italy, the Netherlands, Sweden,
and the United Kingdom). Naturally, they would rather bring the EU norms
up to their level so they are not at a competitive economic disadvantage.24
The Maastricht Treaty allowed most environmental legislation to be
enacted by a qualified majority in the council, whereas previously unanimity
was normally required. In an offsetting provision, the treaty placed greater
emphasis on the principle of subsidiarity, under which actions should be
taken by the member states unless the objectives can be better achieved
through EU actions. Since 1992 some states have used this principle as a
rationale to challenge some community-wide environmental legislation. The
Amsterdam Treaty did not alter the provisions on majority voting and failed
to clarify the meaning of subsidiarity, leaving this conflict unresolved. Effec-
tive implementation of subsidiarity also has been a matter of debate in the
discussions of a European constitution.

The Sixth Environmental Action Programme

As noted, since 1972 the commission has developed environmental action

programmes to guide its activities for a multiyear period. Although these
programs are not legally binding, they have had substantial influence on
policy development at the EU level and among member states. The Fifth
Environmental Action Programme, Towards Sustainability, focused on
the need to integrate environment into other economic and sectoral poli-
cies. In line with this, the commissions Communication on Environment
and Employment, issued in November 1997, spelled out for the first
time how environmental protection and job creation can be mutually rein-
forcing.25 The Sixth Programme, Environment 2010, Our Future, Our
Choice, continues this theme. It calls for the EC to integrate environmental
Chapter 52 Page 57 September 11, 200819:53

Axelrod et al. EU as an Environmental Governance System 57

concerns into all its policies and to promote sustainable development within
the enlarged Community. The programme is based on the polluter pays
principle, the precautionary principle and preventive action, and the prin-
ciple of rectification of pollution at source. Its four priority areas are:
climate change, nature and biodiversity, environment and health, and natural
resources and waste. Progress in achieving goals is to be measured in relation
to targets and timetables covering a ten-year period.
The programme also places great importance on the achievement of
sustainable development in the accession countries through implementation
of the environmental acquis, promotion of the transfer of clean technolo-
gies, exchange of information, and promotion of civil society. In responding
to criticism that the Community needs to do more to fulfill its own envi-
ronmental goals, the programme encourages greater emphasis on effective

Legislative Action

The EU has more than 200 pieces of environmental legislation in force.27

Most of this legislation was enacted during the 1970s and 1980s to address
the major sources of pollution. Since 1992 there has been a marked slow-
down in the passage of legislation. One reason for this is that the principle
of subsidiarity was elevated to a prominent place in the Maastricht Treaty
to gain the support of several states that were reluctant to move toward
further economic integration. According to Article 5 of the Consolidated
Treaty, the EU can take action only if and in so far as the objectives of the
proposed action cannot be sufficiently achieved by the Member States and
can therefore, by reason of the scale or effects of the proposed action, be
better achieved by the Community.28 Although the exact meaning of this
wording is unclear, one consequence is that the commission has been more
cautious in proposing new regulations and directives since 1992. Another
is that there has been substantially increased pressure from industry and
from some governments for deregulation or at least greater flexibility in the
design and implementation of policies. Sluggish economic growth and high
unemployment have made pivotal states such as Germany and France more
reluctant to impose new costs on industry. Finally, there is growing accep-
tance of the need to work to integrate environmental policies into existing
sectoral policies.
The EU has shifted its emphasis toward consolidation of existing policies,
improving implementation and enforcement, and promoting more integrated
decision making before extending regulation in new areas. The EU has also
taken on more of a global environmental leadership role. It has taken a
relatively aggressive stance on certain global issues, championing the Kyoto
Protocol on climate change despite U.S. opposition to the agreement and
differing with the United States in agricultural trade negotiations on the issue
of the safety of genetically modified organisms.29
Chapter 52 Page 58 September 11, 200819:53

58 International Environmental Issues and Debates

Regulations, Directives, and Framework Directives

Legislation in the EU can take the form of either regulations or directives.

Regulations are directly binding on member states and require no further
legislation at the national level; they are used when technical standardiza-
tion is necessary. Directives are more commonly used for environmental
There have been some efforts to establish new directives. In November
2000, for example, the commission adopted a Green Paper on Security of the
Energy Supply, which raises issues regarding the position of nuclear energy
vis--vis other energy sources in lieu of concerns about meeting goals for
reduction of greenhouse gas emissions, on the one hand, and concerns about
aging nuclear power plants on the other. Since December 2001 the council
has requested regular reports on nuclear safety, and in December 2002 the
ECJ confirmed in a judgment that the Community has the authority to legis-
late in relation to nuclear safety. In January 2003 the commission adopted
two proposals for directives for a Community approach to issues of nuclear
plant safety and radioactive waste disposal.30
Much of the commissions work in the past decade, however, has been
focused on consolidating existing directives with the establishment of frame-
work directives. Under Article 249 of the Consolidated Treaty, directives
shall be binding, as to the result to be achieved, upon each Member State   
but shall leave to the national authorities the choice of form and methods.31
This means that directives must be transposed into national law each
country must pass or amend legislation to achieve the stated objectives. This
allows states to adapt EU policies to their particular legal and administra-
tive traditions, but choice of form and methods gives states considerable
discretion in deciding how to achieve EU policy goals. The result has been
wide variations among countries in the implementation of EU environmental
laws and severe difficulties in monitoring and assessing progress toward
common goals.
Framework directives attempt to deal with this problem by establishing
comprehensive long-term environmental quality goals and standards that
can be used to measure progress across a wide range of specific policy instru-
ments and actions. They provide a mechanism for consolidating, integrating,
and simplifying related pieces of legislation (for example, separate directives
on drinking water, bathing water, and protection of shellfish) to encourage
more comprehensive and efficient management of resources. While allowing
countries greater flexibility in pursuing these goals, they can also serve as a
catalyst to force states to adopt a more integrated approach to environmental
protection. Finally, an explicit purpose of the new directives is to increase
transparency in environmental regulation by ensuring public access to
information in a timely fashion.
Despite these laudable objectives, many environmental groups and
members of parliament fear that the trend toward more general framework
directives will result in a weakening of existing environmental controls. They
Chapter 52 Page 59 September 11, 200819:53

Axelrod et al. EU as an Environmental Governance System 59

are concerned that minimum standards may be set too low and that EU
enforcement will become even more difficult than it already is. Therefore,
in the process of reviewing the directives drafted by the commission, the
parliament has adopted the role of preventing any weakening of existing
environmental controls.
In addition to numerous directives addressing such matters as environ-
mental impact assessment, nature protection, chemicals in the environment,
and genetically modified organisms, there are now environmental frame-
work directives addressing air quality, integrated pollution prevention and
control, waste, and water quality.
Integrated Pollution Prevention and Control. The Integrated Pollution
Prevention and Control Directive (96/61/EC), issued in 1996, is intended to
provide much of the operational foundation for the other directives because it
imposes common requirements for issuing permits to large industrial sources
of pollution throughout the EU. Member states must require all new and
existing facilities to obtain operating permits that ensure that all appropriate
measures are taken to prevent or minimize pollution of the air, water, and
land. The directive calls for use of both environmental quality and emis-
sion standards (the combined approach), which accommodates different
national systems; for example, the British rely on ambient quality standards,
but the Germans insist on strict emission limits. Emission standards are to
be based on the best available techniques, but state authorities are given
discretion to determine specific technologies appropriate to local conditions.
However, the directive reserves the councils right to set community-wide
emission limit values for certain categories of installations and pollutants if
The larger significance of the directive is that states are encouraged to
take a comprehensive, integrated approach to pollution reduction at the
source, including waste minimization, efficient use of energy, and protection
of soil and groundwater as well as surface waters and air. This approach
is in line with the shift evident in Europe from end-of-the-pipe controls to
pollution prevention; more integrated, long-term environmental manage-
ment; and greater flexibility in the use of policy instruments. The subsequent
air and water directives are designed to consolidate existing legislation and
provide a legal framework for integrating policy across these media.
Ambient Air Quality and Auto Emissions Standards. EU legislation to
protect air quality goes back to 1970, when the first directive to regu-
late emissions from automobiles was passed (70/220/EEC). Since then, the
Council of Ministers has enacted more than thirty directives on air pollution
covering, among other things, diesel engine emissions, the lead and sulfur
content of fuels, and emissions from large industrial facilities, power plants,
and waste incinerators. Ambient air quality standards also have been set
for sulfur dioxide, nitrogen dioxide, particulates, and lead; and regulations
to limit chlorofluorocarbons (CFCs) and other ozone-depleting gases have
been implemented under the Montreal Protocol. Europe generally followed
the lead of the United States in setting these standards but lagged somewhat
Chapter 52 Page 60 September 11, 200819:53

60 International Environmental Issues and Debates

behind, for example, in not requiring the installation of catalytic converters

in cars until 1991. Air pollution has become a severe problem in many Euro-
pean cities; however, methods for measuring, assessing, and reporting air
quality data have varied greatly from country to country, making it difficult
to apply common standards.
The commission proposed a new air quality framework directive in July
1994, which was adopted by the council in September 1996. The primary
goals of the directive are to define and establish objectives for ambient
air quality in the Community designed to avoid, prevent or reduce harmful
effects on human health and the environment as a whole and to assess
the ambient air quality in member states on the basis of common methods
and criteria.32 Additionally, three daughter directives have been drafted
by expert working groups that set limit values for nitrogen oxides, sulfur
dioxide, lead, particulate matter, carbon monoxide, benzene, and ozone.
Public release of this information is also required. A fourth daughter directive
is under development.33
At the same time, the commission has developed separate legislation to
further reduce acid precipitation (primarily from sulfur dioxide and nitrogen
dioxide) and to cut pollution from vehicles. Increasingly, stakeholders are
being brought into the decision-making process. In 1993 the Energy, Industry
and Environment commissioners jointly launched the Auto-Oil Programme
(AOP), inviting auto manufacturing and petroleum industries to participate
in the development of a plan for meeting EU-wide air quality standards. Both
industries felt that they were being disproportionately burdened in efforts
to reduce traffic-related emissions. The AOP involved an intense series of
negotiations and scientific studies that led to a commission proposal in 1996
to regulate the content of fuels and set far more stringent emission limits
on new cars. However, the proposal received strong protests from manufac-
turers (for being too expensive) as well as from environmentalists (for not
being strict enough).34 The decision-making process was also criticized for
failing to include other stakeholders.
One of the major points of contention was whether the higher standards
set for 2005 would be voluntary, as the commission proposed, or manda-
tory. Germany, the Netherlands, and the Nordic countries supported tighter
exhaust standards, while countries such as France, Italy, and Spain opposed
mandatory controls.35 The parliament largely sided with the former states,
passing more than 200 amendments in early 1997 that would further tighten
and make mandatory the emission controls for 2005, while offering addi-
tional time to the poorer southern states to meet the fuel standards.36 Then in
June 1998 the council and the parliament reached a compromise agreement
on the first Auto Oil Directive that mandates a 70 percent reduction in tail
pipe emissions by 2005.37
A second stage of the program (Auto-Oil Programme II) was launched
in 1997. In addition to the auto and petroleum industries, local authori-
ties, member states, and nongovernmental organization representatives were
invited to join in the work of this program. AOP II assessed the applicability
Chapter 52 Page 61 September 11, 200819:53

Axelrod et al. EU as an Environmental Governance System 61

of the standards set for 2005, considered further standards for fuels and
motorcycles, and proposed nontechnical and local measures for the imple-
mentation of emissions reductions. Also proposed was the increased use of
taxation and other fiscal measures to alter consumer behavior. The AOP II
decision-making process was more inclusive of different viewpoints and thus
has helped pave the way for multi-stakeholder assessments of measures in
other policy areas as well.38
Building on the success of the AOPs, for example, in May 2001 the
commission established the Clean Air for Europe (CAFE) Programme. It uses
the multi-stakeholder approach in assessing measures to address particulate
matter, tropospheric ozone, acidification, eutrophication, and damage to
cultural heritage.39
Water Resources Management. EU water quality has been protected since
1975 by directives covering drinking and bathing water, fish and shellfish,
groundwater, urban waste water, and protection against nitrates from fertil-
izers and various dangerous chemicals. Other policies covered pollution of
European seas and rivers under various international maritime conventions
and agreements. A new drinking water directive was adopted in 1998.
In February 1996 the commission called for a water framework direc-
tive that, like the air framework, would establish broad guidelines for the
protection and management of all freshwater resources. In October 2000
the EU Water Framework Directive (Directive 2000/60/EC), which replaced
seven existing directives, was finally adopted after years of sometimes tense
discussion and debate. The directive is based on a river basin management
approach, with the idea that water quality can best be protected if an entire
ecological systems pollution problems are dealt with in an integrated fashion
combining emission limits and quality standards. States were to have trans-
posed the directive into national policy by 2003. In the coming years, they
are to conduct studies of the characteristics of river basins, establish a moni-
toring network, finalize their river basin management plans, and introduce
pricing policies. The ultimate goal is to obtain good status for water
quality throughout the EU by 2015.40
Proposal for an Environmental Liability Directive. The White Paper
on Environmental Liability, issued in February 2000, called for the estab-
lishment of a Framework Directive on Environmental Liability that would
introduce the possibility of NGOs bringing suits for environmental harms.41
The subsequently proposed EU Environmental Liability Directive, however,
dropped this provision in favor of a system that focuses on environmental
restoration and cost recovery.42

Other Legislation

The EU has enacted legislation on many other aspects of environmental

protection, including environmental impact assessment, control of chemicals
and other dangerous substances, hazardous waste transfer and management,
Chapter 52 Page 62 September 11, 200819:53

62 International Environmental Issues and Debates

development of renewable energy, and protection of forests, wildlife, and

biodiversity. Space does not permit analysis of all of these policies, many of
which continue to undergo amendment and revision. The following section
briefly examines two areas in which the EU has tried to take the lead: pack-
aging reduction and eco-labeling. In both cases individual states had begun
to adopt national legislation that threatened to create barriers to trade, which
could trigger action by the European Commission.
Packaging and Recycling. Beginning in the 1970s, a number of European
countries began to enact laws to reduce the volume of solid waste by
requiring the use of returnable beverage containers, encouraging recycling
of materials, and limiting waste in packaging. Denmark led the way by
banning the use of aluminum cans and requiring that beer and soft drinks
be sold in reusable bottles. Denmarks action eventually led to the seminal
ruling by the ECJ in 1988 that such restrictions on trade may be justified on
environmental grounds, provided that they do not unfairly discriminate in
favor of domestic producers. Other countries have since passed legislation
mandating the reduction or recycling of certain materials, including pack-
aging. In 1991 Germany gained international attention for its novel pack-
aging ordinance (Verpackungsverordnung), which required retail stores to
take back all used packaging materials from consumers and process them.
The ordinance allowed business and industry to set up a private collection
system (the green dot system) on condition that it could meet ambitious recy-
cling targets for various materials; otherwise a mandatory deposit would be
levied on the sale of relevant products. The Netherlands also established an
ambitious recycling program that required industry to reduce its volume of
packaging by 2000.43
The EC had adopted a directive on beverage containers in 1985, but
in the wake of the German and Dutch laws and the Danish bottle deci-
sion it was moved to draft a packaging directive that would accommodate
recycling of other materials while preventing the development of potential
trade restrictions. After much haggling over German and Dutch approaches
mandating higher recycling targets and a coalition led by the United Kingdom
that objected to such rigid quotas, a compromise was reached that lowered
the mandatory targets to 50 percent recovery, 25 percent overall recycling,
and 15 percent minimum recycling for each material. While states were
allowed to exceed these targets, Germanys experience suggested that too
high recovery rates could lead to excessive accumulation and export of waste
materials because of inadequate processing capacity. Therefore, the final
EU packaging and packaging waste directive (94/62/EC), passed in 1994,
also set maximum recovery rates of 65 percent and recycling rates of
45 percent, over the objections of Denmark, Germany, and the Netherlands.
According to some analysts, this compromise illustrates that the establish-
ment of an integrated market still takes precedence over environmental
protection.44 It is noteworthy, however, that despite industry opposition, the
commission presented a proposal to amend the 1994 directive in December
2001. The proposal set new targets for recovery and recycling to be met
Chapter 52 Page 63 September 11, 200819:53

Axelrod et al. EU as an Environmental Governance System 63

by midsummer 2006 for all pre-enlargement member states except Greece,

Ireland, and Portugal, which were given three additional years to meet the
targets. The proposal, which through a co-decision procedure was approved
by the parliament on September 3, 2002, requires overall recovery and recy-
cling targets of between 60 to 65 percent and 55 to 70 percent, respectively.45
In February 2001 the commission adopted a Green Paper presenting ideas
for strengthening product-focused environmental policies. Two years later
the commission held meetings with stakeholders in order to evaluate the
implementation of the directive and to assess ways to reduce packaging and
encourage reuse of packaging.46 The commission also is currently studying
the possibility of introducing an Integrated Product Policy (IPP) that would
address the entire life cycle of a product and its environmental impacts, an
approach that is consistent with promoting sustainable development.47
Eco-Labeling. Another approach to limiting waste and environmental
damage generally is to encourage consumers to purchase more ecologically
benign products by providing better information. Germany had introduced
an eco-labeling system, and several other countries were planning to do so
when in 1990 the council asked the commission to prepare a regulation
establishing criteria for an EC labeling scheme. The initial criteria used for
granting a green label took into account the environmental impact of
the product throughout its entire life cycle, including the materials used,
manufacturing technologies, health and safety of workers, and ultimate
disposal costs. Under the council regulation (92/880/EEC) of March 23,
1992, member states were authorized to appoint a competent body to award
the EC eco-label to manufacturers or importers whose products met the
criteria. Participation by industry was voluntary, but it was hoped that
consumer demand for green products would drive producers to compete
for the label (symbolized by a flower) by designing better products.
In July 2000 a revised eco-labeling regulation (1980/2000/EC) was
passed. It authorizes a new EU Eco-Labeling Board, consisting of eco-
labeling bodies in each member state, consumer groups, environmental
NGOs, trade unions, industry, and small and medium-sized enterprises, to
jointly develop eco-label criteria for different product groups. There are now
nineteen product groups, ranging from furniture and household appliances
to electrical equipment and bedding. Consumers can learn about products
bearing the eco-label on the eco-label catalogue Web site.48 National eco-
labeling bodies remain in charge of implementing the system.

Major Challenges Facing the European Union

Implementation of Environmental Laws

As the examples suggest, the success of the EU commitment to environmental

protection depends on the extent to which member states transpose EU law
into national law. However, they must also apply and enforce the law in
Chapter 52 Page 64 September 11, 200819:53

64 International Environmental Issues and Debates

practice.49 Although there are few thorough analyses of EU treaty compliance

and enforcement to date, scattered studies indicate that the record is uneven
and that in some areas implementation may be deteriorating.50
The European Commission monitors policy implementation and seeks
to detect violations of EU law, but it has no authority to investigate or
inspect specific facilities. In 2001 the commission brought seventy-one envi-
ronmental cases before the ECJ, an increase of approximately 40 percent
over the previous year. Environmental cases represented one-third of all
infringement cases and complaints. The reasons for the growing number
of complaints include greater awareness on the part of citizens that they
can bring complaints to the commission, greater environmental awareness,
the commissions monitoring of national conformity to EU law, the lack of
adequate institutions at the national level, and the complexity of EU direc-
tives and regulations. The rate of alleged violations varies greatly among
states: in 2001 France and Italy both had twenty-two infringement cases
referred to the ECJ, compared with five for the Netherlands and two for
The EU has some legal enforcement mechanisms at its disposal. Citi-
zens, local authorities, businesses, or interest groups have the right to lodge
complaints on the inadequate application or transposition of EU law directly
before the commission. Once a complaint has been brought, efforts are made
to mediate the dispute or to informally persuade the national government to
take appropriate action. If a party is found to be in violation of EU law, the
commission can issue a formal notice to the state. If all else fails, an infringe-
ment case can be brought before the ECJ to force compliance. However,
resolution of cases can take many years, and even if a government is found
guilty, compliance is not automatic. Article 171 of the Maastricht Treaty
allows the ECJ to levy financial penalties on states that fail to carry out its
decisions, but the ECJ has only recently begun to use this power. As part
of the trend in the EU to increase transparency, as of January 2001 formal
notices to refer cases to the ECJ and to terminate cases are published on the
secretariat-generals Web site.52
Some of the variation in compliance among states is related to different
levels of awareness of citizens and interest groups.53 Some states may have
proportionately more complaints lodged because their citizens are more
alert, informed, and able to bring matters to the attention of the EU. But
differential enforcement is also the result of variations in the budgets and
other resources of governments to carry out EU mandates. Because states
choose their own means of compliance, differences are inevitable in the
instruments used and in the severity of penalties levied against violators.54
For example, the ECJ found Italy guilty of nontransposition and noncompli-
ance with an EU directive on protection of wild birds (86/411/EEC) and fined
the government. In another example, the ECJ censured France for failing
to comply with EU directives on air pollution. As a consequence, France
incorporated the directives directly into its laws.55
Chapter 52 Page 65 September 11, 200819:53

Axelrod et al. EU as an Environmental Governance System 65

A growing volume of such cases has led to calls in the parliament and else-
where for an EU inspectorate, possibly under auspices of the EEA. However,
states have resisted any such extension of supranational powers, and its
establishment does not appear likely. The commission instead works with an
informal network of national environmental officials, known as the Imple-
mentation and Enforcement of EU Environmental Law (Impel) network.56
In October 1996 the commission also issued a communication on imple-
mentation, which proposed guidelines for states to follow in carrying out
inspections, handling public complaints about legal enforcement, and guar-
anteeing access by NGOs to national courts.57
The commission is interested in the introduction of a standard clause
on sanctions that would require national governments to enforce effective
penalties and sanctions for noncompliance of environmental laws. To date,
however, the commission has failed to win adequate support for this or even
for a comprehensive review of fines and penalties as they are used by various
member states.58

New Instruments and Approaches

In part because of the ineffectiveness of some existing legislation but also

because of looming new problems that will require different regulatory
approaches, the Fifth Environmental Action Programme called for broad-
ening the range of instruments for attaining sustainable development. The
Sixth Environmental Action Programme has continued this theme, calling for
the active involvement of all sectors of society, and has issued a user-friendly
publication, Our Future, Our Choice.
There is growing interest in the EU in market-based instruments (inclu-
ding taxes and economic incentives, environmental auditing, and voluntary
agreements), horizontal supporting instruments (research, information, and
education), and new financial support mechanisms.59 Such new approaches
are now widely advocated to improve the economic efficiency of regulation
and to involve all sectors of society in shared responsibility for the environ-
ment. As the costs of environmental regulation rise, they are also supported
(at least in theory) by business and industry as an alternative to traditional
command and control regulation.
Several countries in northern Europe have enacted extensive green
taxes to promote waste reduction and energy saving.60 Although the EU does
not have the legal competence to impose new taxes directly, it can encourage
states to do so on a coordinated or harmonized basis; for example, the Euro-
pean Commission issued a communication in early 1997 setting out guide-
lines for such taxes.61 The environmental tax base as a share of the total EU
tax base is on the rise. It went from 5.87 percent in 1980 to 6.17 percent in
1990 and 6.71 percent in 1997. Ninety-five percent of this comes from taxes
on energy and transport but less than 5 percent from taxes on emissions,
chemical substances, waste products, and natural resources.62
Chapter 52 Page 66 September 11, 200819:53

66 International Environmental Issues and Debates

Particularly as a reaction to concerns about how to meet reductions

in greenhouse gas emissions under the Kyoto Protocol, the EU has moved
rapidly in recent years to adopt new policy instruments. We comment briefly
on four: energy taxes, emissions trading, renewable energy portfolios, and
voluntary industry agreements.
Energy Taxes. Concern over global climate change due to the accumu-
lation of carbon dioxide and other greenhouse gases in the atmosphere has
been particularly strong in Europe and has led to a search for effective means
of reducing dependence on fossil fuels. For many years, the commission
sought to win acceptance of the idea of a carbon tax and made numerous
proposals to this effect, but to no avail. In 1997 the commission for a
third time proposed the establishment of a directive on energy taxation.
The plan would have extended the existing harmonized excise tax duties
on mineral oils to coal, natural gas, and electricity, while allowing national
governments to offer rebates for environmentally friendly forms of energy
production. In the spring of 2003, EU finance ministers finally agreed on a
watered-down version of the commissions proposal. They called for a direc-
tive on minimum rates of energy taxation to curb the use of fossil fuels and
encourage sustainable transport.
Emissions Trading. The idea of a carbon emissions trading scheme was
originally proposed by the United States during the negotiations leading
to the Kyoto Protocol but was initially strongly resisted by the EU. After
the Bush administration announced it would withdraw from the Kyoto
Protocol in the spring of 2001, however, the EU unanimously resolved
to move forward in trying to put the agreement into force even without
the United States and began to look more positively at emissions trading
Under the Kyoto Protocol, the EU agreed to reduce its combined emis-
sions of greenhouse gases (including carbon dioxide) by 8 percent of 1990
levels by 20082012. Under an internal burden-sharing arrangement, coun-
tries such as Germany and the United Kingdom are required to reduce their
emissions by substantial amounts, while countries such as Greece, Spain,
and Portugal are allowed to increase emissions. A report issued by the EEA
found that, as of 2000, the EU as a whole had met its interim goal of stabi-
lizing carbon dioxide emissions by 2000, but that some member states were
behind in meeting their respective reduction targets. Steep cuts in emissions
in Germany and the United Kingdom largely explain the stabilization of
emissions within the EU. The report went on to note that if the EU is to meet
its 8 percent reduction target, other measures, such as emissions trading, are
In October 2001 the commission proposed a greenhouse gas emissions
trading scheme to the parliament and the council. Modeled on the successful
sulfur dioxide emissions trading system employed in the United States, the
EU scheme will be the first international trading system in the world. It is
to come into effect in 2005 and will cover the member states of the EU as
well as members of the EEA. It will establish absolute limits on emissions
Chapter 52 Page 67 September 11, 200819:53

Axelrod et al. EU as an Environmental Governance System 67

of carbon dioxide and then set up a system allowing installations to trade

emission allowances. For example, if a firm wants to expand and thus pollute
more, it will have to buy more emission allowances. Firms that can reduce
their emissions at relatively low cost will have an incentive to do so and to
sell off their allowances.64 This kind of scheme may in fact work well in the
EU, where there is a long history of burden sharing, an underpinning for
voluntary agreements.
Renewable Energy. Another policy measure that has been proposed to
reduce greenhouse gas emissions is the Directive on the Promotion of Elec-
tricity from Renewable Energy Sources. The directive obliges member states
to ensure access to the grid of renewable energy sources and sets a goal
of 12 percent of all electricity being produced from renewable energy by
2010.65 The remarkable growth of wind energy in the EU suggests strong
public interest in promoting renewable energy.
Voluntary Agreements. Yet another innovative strategy calls for volun-
tary agreements between government and industrial sectors or individual
companies to pursue pollution prevention and sustainable development.
Such agreements or covenants have been a central feature of Dutch plan-
ning and were endorsed in the Fifth Environmental Action Programme. The
idea is that voluntary cooperation can supplement (but not replace) legal
obligations in many sectors in setting and achieving goals and can at times
be more cost efficient.66 Examples are industry agreements in Denmark on
the phasing out of organic solvents used in paints and varnishes; in Belgium
on elimination of CFCs; and on waste management and improving energy
efficiency in several states. While such agreements can improve communica-
tion and cooperation between regulated industries and governments, envi-
ronmental groups have been skeptical of them on the grounds that they may
not be transparent and may amount to a form of backdoor deregulation
that allows companies to circumvent existing laws. Nevertheless, their use
is becoming increasingly common.
The commission issued a communication on voluntary agreements in
late 1996 that attempted to clarify the legal and other considerations that
should guide such agreements. Under these guidelines, agreements must take
contractual form, be published and open to the public, have quantified
objectives and deadlines, and be monitored, with performance reported to
competent authorities. When appropriate, they may include sanctions for
nonfulfillment. When the agreements are used to implement EU policy, the
commission must be notified of all details and will then scrutinize them for
conformity with Community law and certify their transparency and cred-
ibility.67 In some cases the commission also makes voluntary agreements
directly with industry. For example, in 1997 and 1998 the commission
conducted negotiations with the European auto manufacturing association
on reduction of carbon dioxide emissions of future models, leading to a
voluntary agreement to cut emissions of new cars by about one-quarter by
2008. This agreement was accepted by the Council of Ministers in October
1998, eliminating the need for new legislation in this area.
Chapter 52 Page 68 September 11, 200819:53

68 International Environmental Issues and Debates

In June 2002 the commission published an Action Plan on Simplifying and

Improving the Regulatory Environment, which endorses the use of voluntary
agreements at the European level as well. Whereas the 1996 communiqu
dealt only with voluntary agreements at the national level, the Action Plan
outlines how voluntary agreements can be used at the EU level as a means of
achieving the Sixth Environmental Action Programmes goal of improving
the effectiveness and efficiency of implementation.68


Perhaps the greatest challenge to EU environmental policy of the future will

be to bring the ten new member states up to existing pollution standards.
Under the treaty, all new members of the EU must comply with the full
acquis communautaire, including all environmental directives and regula-
tions. The EU increasingly shies away from giving concrete estimates of the
cost of bringing the accession countries into compliance with the acquis. It is,
however, possible to ascertain costs to date. For the 19902003 period, close
to 20 billion euros were expended in the accession countries. The amount
committed for the 20042006 period is 49.5 billion euros, most of which
are for agriculture and structural operations.69 As the environmental acquis
is just one of many chapters of the acquis communautaire that accession
countries must meet, the competition among programs for limited financial
resources is likely to be great.
The central and eastern European countries have far higher levels of
pollution than western European countries because of the lax environ-
mental standards under their former communist governments. In addition to
outdated coal-burning power plants and factories, several of these countries
rely heavily on Soviet-designed nuclear plants that pose substantial safety
risks. Although some of the worst facilities have been closed down, safety
concerns remain. Many of the accession countries lack adequate sewerage
and waste disposal systems. Estimates of meeting the environmental acquis
for the ten accession countries typically range from 80 to 120 billion euros,
or approximately 10 billion euros annually. While these gures are fraught
with uncertainty, the cost of coming into compliance will certainly place
a great strain on national budgets. Nevertheless, there are benefits to be
accrued from compliance in terms of a reduction in the hidden costs to the
economy of pollution and human health problems.70 A World Bank report
also suggests the potential for reducing the costs of coming into compliance
through effective planning and use of innovative policy measures.71
The EUs Agenda 2000 recognizes that there is a strong possibility that the
new members will enter the EU with compliance levels well below even the
worst current offenders, but emphasizes the importance of working toward
the long-term improvement of environmental conditions, especially air and
water quality. Cooperation at all levels, including both public and private
financing, will be necessary to assist the accession countries in transposing,
interpreting, and implementing the environmental acquis.
Chapter 52 Page 69 September 11, 200819:53

Axelrod et al. EU as an Environmental Governance System 69


The EU is at a critical stage in its evolution because of the further deep-

ening of integration through the new monetary union and the largest
widening in its history.72 Without question, the EU has made great strides
toward environmental protection over the past quarter-century, but it has
also entered a transition phase in this policy area. After two decades of
imposing increasingly detailed environmental directives and regulations from
Brussels, the EU has begun to revise its approach since adoption of the
Maastricht Treaty in 1992 and its emphasis on the principle of subsidiarity.
Member states have pressed for greater freedom in implementing EU legisla-
tion while supporting the general principles of sustainable development. The
European Commission has responded by turning toward the use of broader
framework directives that set long-term environmental goals while allowing
more flexibility in the choice of means to achieve them; it also has encour-
aged the introduction of new policy instruments at both the national and
EU levels to improve environmental performance and cost-effectiveness. At
the same time it has backed the Kyoto Protocol over U.S. opposition and is
developing the worlds first international carbon dioxide emissions trading
Despite these accomplishments, the EU faces daunting challenges both
in extending its policies eastward and in maintaining its standards in the
member states. Economic stagnation and high levels of unemployment have
dampened public and government enthusiasm for increased environmental
protection. Implementation of EU environmental laws at the national level
leaves much to be desired, and there is still a large gap between the northern
green states (including the three members that joined in 1995) and the less
wealthy southern countries. The addition of the ten primarily central and
eastern European states, which are economically farther behind the western
European states, threatens to shift the balance of power toward the laggards.
On the other hand, from a global perspective, enlargement of the EU to
include these countries could improve conditions in the region as a whole if
the new members can be convinced to significantly raise their standards.
Within the EU generally, sustainable development will require much
greater integration of environmental perspectives into other policy areas such
as energy, transportation, agriculture, and tourism.73 Article 6 of the Consol-
idated Treaty legally obligates all EU bodies and member states to pursue
such integrated sustainable development strategies. The real test of the EU
governance system therefore still lies ahead.


1. The EU negotiates on behalf of the member states insofar as they are in agreement and
is a signatory to most recent conventions, but it does not have exclusive jurisdiction. Individual
member states can sign separately, as a member of the EU, or both. See Angela Liberatore,
Chapter 52 Page 70 September 11, 200819:53

70 International Environmental Issues and Debates

The European Union: Bridging Domestic and International Environmental Policy-Making,

in The Internationalization of Environmental Protection, ed., Miranda A. Schreurs and Eliza-
beth C. Economy (Cambridge: Cambridge University Press, 1997), 204206.
2. On the general history and development of the EC/EU, see Clifford Hackett, Cautious
Revolution: The European Community Arrives, rev. ed. (New York: Praeger, 1996); Desmond
Dinan, Ever Closer Union? An Introduction to the European Community (Boulder, Colo.:
Lynne Rienner, 1994); David M. Wood and Birol A. Yesilada, The Emerging European Union
(White Plains, N.Y.: Longman, 1996); Neill Nugent, The Government and Politics of the Euro-
pean Union (Durham, N.C.: Duke University Press, 2003); Michelle Cini, ed., European Union
Politics (Oxford: Oxford University Press, 2003).
3. See European Council, Presidency Conclusions, Copenhagen European Council, 12
and 13 December 2002, press release, 13/12/2002, Nr: 400/02.
4. For a summary of the programs, see Stanley P. Johnson and Guy Corcelle, The Envi-
ronmental Policy of the European Communities, 2d ed., (London: Kluwer Law International,
1995); and David Judge, ed., A Green Dimension for the European Community (London:
Frank Cass, 1993). See also John McCormick, Environmental Policy in the European Union
(Basingstroke, U.K: Palgrave, 2001).
5. For the Consolidated Treaty, see http://europa.eu.int/eur-lex/en/treaties/dat/EC_consol.
6. See Treaty of Nice Amending the Treaty on European Union, The Treaties Establi-
shing the European Communities and Certain Related Acts, 2001/C 80/01, Official Journal
of the European Communities, 10.3.2001; available at http://europa.eu.int/eur-lex/en/treaties/
7. The European Convention, The Secretariat, Draft Treaty Establishing a Constitution
for Europe, CONV 850/03, (Brussels, July 18, 2003). The changes to the draft treaty, agreed
upon in June 2004, can be found in Conference of the Representatives of the Governments
of the Member States, CIG 85104, PRESID 27 (Brussels, June 18, 2004). All member states
must ratify the Constitution for it to go into force.
8. See, for example, Albert Weale et al., Environmental Governance in Europe: An Ever
Closer Union? (Oxford: Oxford University Press, 2002); Andrew Jordan, ed., Environmental
Policy in the European Union (London: Earthscan, 2002); Gary Marks et al., Governance
in the European Union (London: Sage Publications, 1996); and Alan W. Cafruny and Carl
Lankowski, eds., Europes Ambiguous Unity: Conict and Consensus in the Post-Maastricht
Era (Boulder, Colo.: Lynne Rienner, 1997).
9. A useful introduction to the institutions of the EU is Nugent, The Government and
Politics of the European Union.
10. Michelle Cini, The European Commission (Manchester: Manchester University Press,
1996); Geoffrey Edwards and David Spence, eds., The European Commission, 2d ed., (London:
Catermill, 1997).
11. Richard Corbett, Francis Jacobs, and Michael Shackleton, The European Parliament,
3d ed. (London: Catermill, 1995); and Martin Westlake, A Modern Guide to the European
Parliament (London: Pinter, 1994). On the role of political parties, see Simon Hix and
Christopher Lord, Political Parties in the European Union (New York: St. Martins, 1997).
12. L. Neville Brown and Tom Kennedy, The Court of Justice of the European Communi-
ties, 4th ed., (London: Sweet and Maxwell, 1994); Han Somsen, ed., Protecting the European
Environment: Enforcing EC Environmental Law (London: Blackstone Press, 1996).
13. Commission of the European Communities v. Kingdom of Denmark Case 302/86,
Report of Cases Before the Court, vol. 8 (Luxembourg: Office for Official Publications of the
European Communities, 1988).
14. See the EEA at http://org.eea.eu.int.
15. See, for example, Jeremy J. Richardson, ed., European Union: Power and Policy-
Making (London: Routledge, 1996); Helen Wallace and William Wallace, Policy-Making in the
European Union (Oxford: Oxford University Press, 1996); Stephen George, Politics and Policy
in the European Community (Oxford: Oxford University Press, 1991); and Carolyn Rhodes
Chapter 52 Page 71 September 11, 200819:53

Axelrod et al. EU as an Environmental Governance System 71

and Sonia Mazey, eds., The State of the European Union, Building a European Polity?, vol. 3
(Boulder, Colo.: Lynne Rienner, 1995).
16. See Regina S. Axelrod, Subsidiarity and Environmental Policy in the European
Community, International Environmental Affairs 6 (spring 1994): 115132.
17. Adrienne Hritier et al., Ringing the Changes in Europe: Regulatory Competition and
Transformation of the State: Britain, France, Germany (Berlin and New York: Walter de
Gruyter, 1996).
18. Mikael Skou Andersen and Duncan Liefferink, eds., European Environmental Policy:
The Pioneers (Manchester, England: Manchester University Press, 1997); and D. Liefferink and
M. S. Andersen, Strategies of the Green Member States in EU Environmental Policy-Making,
Journal of European Public Policy-Making 5 (June 1998): 254270.
19. Sonia Mazey and Jeremy Richardson, eds., Lobbying in the European Community
(Oxford: Oxford University Press, 1993); Sonia Mazey and Jeremy Richardson, The Logic of
Organisation: Interest Groups, in Richardson, European Union.
20. See the European Environment Bureau, http://www.eeb.org/Index.htm.
21. Pressure Groups Become a Political Force, European Voice, June 1117, 1998.
22. See Andrea Lenschow, ed., Environmental Policy Integration: Greening Sectoral Poli-
cies in Europe (London: Earthscan, 2002).
23. See http://europa.eu.int/eur-lex/en/treaties/dat/EC_consol.html.
24. European Environment Agency, Environmental Signals 2002, Environmental Assess-
ment Report no. 9, Copenhagen, European Environment Agency, May 14, 2002, 127.
25. Commission of the European Communities, Communication on Environment and
Employment, COM (97) 592 Final (Brussels, November 18, 1997).
26. Decision No. 1600/2002/EC of the European Parliament and of the Council of 22 July
2002 laying down the Sixth Community Environmental Action Programme, Official Journal
of the European Communities, L242/1 vol. 45, September 10, 2002.
27. Johnson and Corcelle cite as many as 400 acts in Environmental Policy of the European
Communities; see also Sevine Ercmann, Pollution Control in the European Community: Guide
to the EC Texts and Their Implementation by the Member States (London: Kluwer Law Inter-
national, 1996).
28. See http://europa.eu.int/eur-lex/en/treaties/dat/C_2002325EN.003301.html.
29. See Norman J. Vig and Michael Faure, eds., Green Giants? Environmental Policies of
the United States and the European Union (Cambridge: MIT Press, 2004).
30. See http://europa.eu.int/rapid/start/cgi/guesten.ksh?p_action.gettxt=gt&doc=IP/03/132/
31. See http://europa.eu.int/eur-lex/en/treaties/dat/EC_consol.html.
32. Council Directive 96/62/EC of 27 September 1996 on Ambient Air Quality and Assess-
ment, Article 1.
33. See the European Commission, http://europa.eu.int/comm/environment/air/ambient.htm.
34. Outcry Over Plan for Car Emissions, European Voice, May 30June 6, 1996.
35. Battle Lines Drawn for Confrontation Over Car Emissions, European Voice, January
30February 5, 1997.
36. MEPs Take Hard Line on Car Emissions, European Voice, May 29June 4, 1997;
Delays Hinder Progress of Auto-Oil Deal, European Voice, July 39, 1997.
37. Press release, Auto-Oil Programme Is on the Road (Parliament-Council Conciliation
Committee), General Secretariat of the Council of the European Union, 992498 (PRESS 230)
(Brussels, June 29, 1998).
38. Frazer Goodwin, Controlling Traffic Pollution and the Auto Oil Programme,
European Federation for Transport and the Environment, T&E 99/8.
39. See http://europa.eu.int/scadplus/leg/en/lvb/128026.htm.
40. Directive 2000/60/EC of the European Parliament and of the Council of 23
October 2000, Official Journal L 327, 22/12/2000, pp. 00010073. See also http://europa.
41. Commission of the European Communities, White Paper on Environmental Liability,
COM (2000) 66 (Brussels, February 9, 2000).
Chapter 52 Page 72 September 11, 200819:53

72 International Environmental Issues and Debates

42. Timothy Swanson and Andreas Kontoleon, What Future for Environmental Liability?
The Use of Liability Systems for Environmental Regulation in the Courtrooms of the US and the
EU, in Vig and Faure, Environmental Policies of the United States and the European Union.
43. Markus Haverland, Convergence of National Governance under European Integra-
tion? The Case of Packaging Waste (paper presented at the Fifth Biennial Conference of the
European Community Studies Association, Seattle, May 29June 1, 1997).
44. Thomas Gehring, Governing in Nested Institutions: Environmental Policy in the
European Union and the Case of Packaging Waste, Journal of European Public Policy 4
(September 1997): 337354.
45. Commission of the European Communities, Proposal for a Directive of the European
Parliament and of the Council Amending Directive 94/62/EC, COM (2001)729-Final, Official
Journal C 103, April, 30, 2002. See also http://europa.eu.int/scadplus/leg/en/lvb/121207.htm.
46. See http://europa.eu.int/comm/environment/waste/events_packaging_270303.htm.
47. Commission of the European Communities, Green Paper on Integrated Product
Policy, COM (2001) 68 Final (Brussels, February 7, 2001).
48. See www.eco-label.com.
49. See Peter M. Haas, Compliance with EU Directives: Insights from International Rela-
tions and Comparative Politics, Journal of European Public Policy 5 (March 1998): 1737.
50. Alberta Sbragia, Environmental Policy in the European Community: The Problem
of Implementation in Comparative Perspective, in Towards a Transatlantic Environmental
Policy (Washington, D.C.: European Institute, 1991); Jeremy Richardson, Eroding EU Policies:
Implementation Gaps, Cheating and Re-Steering, in Richardson, European Union; Wyn Grant,
Duncan Matthews, and Peter Newell, The Effectiveness of European Union Environmental
Policy (London: Macmillan, 2000).
51. Commission of the European Communities, Nineteenth Annual Report on Monitoring
the Application of Community Law (2001), COM (2002) 324 Final (Brussels, June 28, 2002).
52. See http://europa.eu.int/comm/secretariat_general/sgb/droit_com/index_en.htm#infr
53. See also Tanja A. Brzel, Leaders and Laggards in European Environmental Policy
(Cambridge: Cambridge University Press, 2003).
54. See also Jonathan Golub, ed., New Instruments for Environmental Policy in the EU
(London: Routledge, 1998), and Matthieu Glachant, ed., Implementing European Environ-
mental Policy (Cheltenham, England: Edward Elgar, 2001).
55. Europe Energy, no. 367, November 15, 1991, 1, 7.
56. Inspectors Agree to Expanded Enforcement Role, ENDS Environment Daily,
May 20, 1997.
57. Commission of the European Communities, Implementing Community Environmental
Law, COM (96) 500 (Brussels, October 22, 1996).
58. See http://europa.eu.int/comm/environment/env-act5/chapt6.htm.
59. Commission of the European Communities, Toward Sustainability, (Brussels: Commis-
sion of the European Communities, 1992) chap. 7.
60. Mikael Skou Andersen, Governance by Green Taxes: Making Pollution Prevention
Pay (Manchester: Manchester University Press, 1994); Timothy ORiordan, ed., Ecotaxation
(London: Earthscan, 1997).
61. Commission of the European Communities, Environmental Taxes and Charges in the
Single Market, COM (97) 9 (Brussels, January 29, 1997).
62. European Environment Agency, Environmental Taxes Gaining Importance in EU
Member Countries, press release, Copenhagen, January 20, 2000.
63. European Environment Agency, Greenhouse Gas Emissions Trends and Projections in
Europe, Environmental Issue Report, no. 33 (Copenhagen: European Environment Agency,
64. Commission of the European Communities, Proposal for a Directive of the European
Parliament and of the Council establishing a scheme for greenhouse gas emission allowances
trading within the Community and amending Council Directive 96/61/EC, COM (2001) 581
Final; 2001/0245 (COD) (Brussels, October 23, 2001).
Chapter 52 Page 73 September 11, 200819:53

Axelrod et al. EU as an Environmental Governance System 73

65. See Commission of the European Communities, Proposal for a Directive on the Promo-
tion of electricity produced from renewable energy sources in the electricity market COM
(2000) 0884 Final, and Commission of the European Communities, Communication from the
Commission to the Council, the European Parliament, and the Economic and Social Committee
of the Regions on the implementation of the Community Strategy and Action Plan on Renewable
Energy Sources (19982000), COM (2001) 69 Final (Brussels, February 16, 2001).
66. Voluntary agreements are seldom entirely voluntary and may be a tool for enforcing
existing regulations. See Duncan Liefferink and Arthur P. J. Mol, Voluntary Agreements as a
Form of Deregulation? in Deregulation in the European Union: Environmental Perspectives,
ed. Ute Collier (London: Routledge, 1998).
67. Commission of the European Communities, Communication from the Commission
to the Council and the European Parliament on Environmental Agreements, COM (96) 561
Final (Brussels, November 27, 1996).
68. Commission of the European Communities, Action Plan on Simplifying and Improving
the Regulatory Environment, COM (2002) 278 Final (Brussels, June 5, 2002).
69. See http://europa.eu.int/com/enlargement/faq/faq2.htm#22.
70. See EcoTec Research and Consulting Limited, The Benefits of Compliance with the
Environmental Acquis, DGENV Contract: Environmental Policy in the Applicant Countries
and their Preparation for Accession. Service Contract B7-8110/2000/159960/MAR/H1, Final
Report, Executive Summary, July 2001, C/1849/PtB.
71. World Bank, Meeting the Environmental Acquis: Cost Estimates for Accession
Countries, (Washington, D.C.: 2002); available at http://europa.eu.int/comm/environment/
72. See Pierre-Henri Laurent and Marc Maresceau, eds., The State of the European Union,
Deepening and Widening, vol. 4 (Boulder, Colo.: Lynne Rienner, 1998).
73. See Lenschow, Environmental Policy Integration: Greening Sectoral Policies in Europe.
Chapter 53 Page 74 September 11, 200819:53

Whalers, Cetologists, Environmentalists, and the
International Management of Whaling
M. J. Peterson

anagement of whaling, like all international policymaking,

M involves both adopting decisions through mutual accommoda-

tion and turning those decisions into effective outcomes. Since
1949, international efforts to regulate whaling have been channeled through
the International Whaling Commission (IWC), an intergovernmental body
established by the International Convention for the Regulation of Whaling
(ICRW).1 The history of IWC regulation has been marked by two major
changes of policy. The first, which began in the mid-1960s and was consol-
idated in the adoption of the new management procedures in 1974, focused
on reducing whale capture to levels deemed consistent with ensuring the
survival of stocks in light of improved biological knowledge. The second,
which was initially proposed in 1972 and adopted in 1982 to take effect
in 1986, imposed zero quotas on the species most exploited by commercial
whalers for a period of up to ten years.
Both changes in policy occurred as different groups acquired prepon-
derant influence in the policy process. Their ability to do so was shaped
not only by the merits of their proposals and the influence of governments
persuaded to adopt them as their own policy but also by the organizational
character of the IWC itself. The IWC consists of one member appointed by
each state that is a party to the ICRW. The commissions membership is
open-ended because Article X of the ICRW permits any state to become a
party. Although the negotiators who wrote the convention probably assumed
that only states possessing an indigenous whaling industry would join, this
open accession clause means that actual participation in whaling is not
required; any state with sufficient interest to pay the relatively low assess-
ments and to send a commissioner to the IWC meetings can accede. Since the
ICRW established no central monitoring or enforcement bodies, implemen-
tation of regulations is left to the national fisheries agencies of each member
Most of the commissioners have been officials from government fisheries
agencies. Yet they have not operated in a vacuum; they have responded to
the arguments of other groups attentive to or affected by their decisions.

Source: International Organization, 46(1) (1992): 14786.

Chapter 53 Page 75 September 11, 200819:53

Peterson Whaling 75

Three groups have been particularly prominent: whaling industry managers,

cetologists, and environmentalists. Despite their differences of emphasis, the
first two groups share the view that whaling is a matter of resource manage-
ment. Cetologists are more conservationist, stressing the need to maintain
whale populations at levels that ensure reproduction over the long term,
while industry managers are more consumptionist, giving greater weight to
the immediate economic return from taking whales. Even so, both groups
regard killing whales and using their bones, oil, meat, and other parts as an
acceptable activity. Influential factions among the environmentalists flatly
reject this approach and argue that whales merit strict preservation. They
believe that people may undertake nonconsumptive uses, such as behav-
ioral research and whale watching, but should not kill whales to use their
parts as food or raw materials. Other environmentalists have less clear-cut
policy views. They accept resource use but apply strict standards of conser-
vation meant to avoid not only the extinction but also the depletion of stocks
to the point that their natural patterns of group behavior are disrupted.2
Adoption of the new management procedures was a victory for the more
conservationist cetologists over the more consumptionist industry managers.
Adoption of the zero quotas was as a victory for the environmentalists
but not necessarily for the preservationists among them. The resolution
regarding the quotas can be read as either conservationist or preservationist
in inspiration.
The history of whaling management by the IWC is not an example
of strong epistemic community influence over policy. A group of experts
sharing a set of causal beliefs, canons of validity, principles, and a common
policy project only briefly predominated in the policy process. A community
of cetologists largely meeting these criteria does exist. Yet it did not determine
outcomes either by the indirect method of defining the terms of debate so
that only its preferred policy appeared reasonable or by the direct method of
placing members or former students into enough deciding and implementing
posts. Over time, however, it has articulated positions accepted by enough of
those involved to prevent both the most short-term profit-motivated industry
managers and the most ardent preservationists from securing everything they
wanted. A static analysis of each separate choice reveals an epistemic commu-
nity usually having only secondary influence; a dynamic analysis of trends
reveals an epistemic community constantly affecting policy by bounding the
range of its fluctuation.
The pivotal role of the U.S. government after 1970 might tempt observers
to begin and end with the conclusion that U.S. preponderance explains every-
thing.3 Yet doing so begs significant questions. First, it cannot explain why
the U.S. government waited so long to use its capability. Most scholars
agree that U.S. predominance was already eroding in 1970; yet the most
significant U.S. actions on whaling issues occurred after 1970. Second, it
cannot explain how or why the U.S. government decided to use its capability
as it did. The choices and steadiness of purpose exhibited by a predomi-
nant power are shaped by the individuals or groups having direct control
Chapter 53 Page 76 September 11, 200819:53

76 International Environmental Issues and Debates

or indirect influence over its policy. Here, too, the conclusion is similar: the
epistemic community helped push U.S. policy in a conservationist direction
but seldom fully determined it.

The Contending Groups

Whaling industry managers and cetologists were active even before the IWC
held its initial meeting in 1949, while the environmentalists became orga-
nized around 1970. The cetologists, the only group to qualify as an epistemic
community, suffered enough internal dissension and outright defections in
the 1970s to lose that status temporarily. Neither the environmentalists
nor the industry managers formed an epistemic community. Although the
members of each group have shared policy preferences, the environmental-
ists have been too divided on principles and canons of reasoning, while the
industry managers have been insufficiently oriented toward revising causal
beliefs in light of new or changed evidence to be regarded as epistemic
The networks of influence during three periods are summarized in
Figures 1, 2, and 3 and discussed in subsequent sections of the article.

Whaling Industry Managers

The industry managers formed an economic interest group whose leverage

stemmed from the ability to provide jobs, from personal ties to polit-
ical leaders, and, in some cases, from contributions to campaign funds.
The industry managers, like members of other economic interest groups,
usually focused their attention on national governments and organized in
national associations. The exact form of their activities and organizations
depended on the relation of whaling to other industries and to the govern-
ment. Japanese commercial whaling, for example, is carried on by firms
that also engage in fishing. Though whaling is a relatively small part of the
combined industry, the link gives whaling companies considerable clout in
Japanese politics.4 Soviet commercial whaling is done by state-owned enter-
prises reporting to the Ministry of Merchant Marine. Norwegian and British
whaling companies began as independent specialized enterprises and then
expanded into other maritime activity as whaling prospects declined.5
Transnational cooperation among industry managers took several forms.
The British and Norwegian firms were linked through the International
Association of Whaling Companies. In the 1960s and 1970s, the Japanese
industry forged a set of transnational links with firms in other countries
through purchases and investments. The large Japanese market for whale
meat was supplied by Brazilian, Chilean, Icelandic, South Korean, Soviet,
and Taiwanese as well as Japanese firms. This network became impor-
tant during the intense competition for IWC votes in the 197682 period
because it gave the Japanese managers local allies in a number of other
Figure 1: Influence Networks, 194960
Chapter 53 Page 77


Association Scientific
Norway Committee

Ministry USSR US Smithsonian

Japan Australia

Nether- New Zealand

Other Other

Association whaling nonwhaling

states states

Industry influence Government agencies


Cetologist influence National private groups

September 11, 200819:53
Figure 2: Influence Networks, 196874
Chapter 53 Page 78


UK Committee Norway


Association Japan Australia

Other New Zealand

International Environmental Issues and Debates

Whalers states

FAO Other
UNEP IUCN states

Industry influence Government agencies

Cetologist influence National private groups
International govern-
mental organizations
September 11, 200819:53
Figure 3: Influence Networks, 197590
Chapter 53 Page 79


UK Committee
Ministry Norway

USSR Local
Local IWC environmental
Iceland Other Local
whalers organizations

Japan Third World

Association states
Other IUCN
states FAO UNEP

Industry influence Government agencies
Environmentalist National private groups

International govern-
mental organizations
September 11, 200819:53
Chapter 53 Page 80 September 11, 200819:53

80 International Environmental Issues and Debates

countries. Until the mid-1970s, managers concentrated on influencing their

own government and relied on their foreign counterparts to do the same.
Later, the Japanese and then the Icelanders and Norwegians got their own
governments to influence other governments and also began hiring American
lobbyists to help influence the U.S. government.


The cetologists formed an epistemic community whose political influence

was based on claims to relevant scientific expertise. They had a shared policy
preference of imposing greater restrictions on all types of whaling whenever
stocks required protection, a shared set of principles reflecting a conserva-
tionist approach to whale taking, a shared set of causal beliefs stemming from
their biological studies, and shared canons of validity. The exact content of
the causal beliefs and the canons of validity changed over time as research
and theoretical argument proceeded. The community members orientation
toward theory conformed to Imre Lakatoss concept of scientific progress.6
They worked with one admittedly imperfect model after another, shifting
when a better one explaining a wider set of observational findings was
Cetologists have always been more transnationally oriented than industry
managers because of the long tradition that science is a global activity. This
tradition pervades their everyday conduct; open publication, international
conferences for discussion of ideas, and inter-institute staff exchanges are
all prominent parts of scientific life. Scientists have well-developed transna-
tional ties through the International Congress of Scientific Unions and
disciplinary transnational organizations, and they regard service on expert
committees advising intergovernmental organizations as a normal activity.7
Contact among cetologists has also been facilitated by small numbers. In
1972, for example, there were no more than thirty full-time cetologists in
the world.8
The epistemic community of cetologists suffered some internal conflict
in the late 1950s, followed by major internal conflict and political eclipse in
the 1970s and early 1980s. In the late 1950s, cetologists had to be supple-
mented by fisheries biologists able to bring more sophisticated mathemat-
ical models to bear.9 Yet this change did not cause major rifts within the
cetologist community because of the common commitment to conserva-
tion and better science. The later experience, however, was different. The
unity on causal beliefs and most appropriate models for analysis broke
down temporarily. Continued research involving not only cetologists but
also the wider community of marine biologists led to the development
of new models. These were adopted quickly by some but questioned by
others. Cetologists thus split into contending schools rather than presenting
consensus advice derived from agreement on a particular model. Some of
those who embraced the newer ecosystems models also shifted principles.
No longer content with a conservationist view, they adopted language far
Chapter 53 Page 81 September 11, 200819:53

Peterson Whaling 81

closer to that of preservationists seeking to prohibit rather than to restrict



The environmentalists concerned with whaling have not qualified as an

epistemic community. While they have agreed on policy, they have not
shared causal beliefs, canons of validity, or principles. Rather, they have
formed an issue-based lobbying group that draws a range of supporters.
Some are scientists who serve the environmentalist cause by their ability
to argue with other scientists as equally qualified experts. Most are drawn
from other occupations. While these nonscientists share a vaguely holistic
approach to nature, they do not share or feel constrained by any one canon
of reasoning. They mobilize for political action best when an issue can be
framed in fairly simple terms indicating a clear policy preference without
the need for a highly detailed explanation of why the chosen policy is
better than others. The time-honored devices of making stark contrasts and
dividing the world into good guys and bad guys work as well for the
environmentalists as they do for other organizers of mass movements. A
temporary ban on whaling has been supported by all wings of the envi-
ronmentalist movement. It has appealed simultaneously to conservationists
concerned with wiser resource use, to radical decentralizers believing that
industrial society must be replaced by simpler life-styles if people are to
develop their full potentials, to preservationists and animal rights activists
objecting to treating whales as a resource under any circumstances, and to
protectors of biological diversity worried that continued whaling will lead
to the extinction of some or all whale species.
The environmentalists can avoid fragmentation of their coalition only
as long as all whale populations appear to be small enough that treating
them as endangered species remains plausible scientifically. Under these
circumstances, the deep differences of principle between preservationists and
conservationists will not come to the fore. So far, the environmentalists have
operated effectively. They have had strong national followings in most of
the Western industrialized states, growing affiliates in a number of Third
World countries, and new opportunities for organizing in the Soviet Union
and other East European nations. They have also had strong and effec-
tive transnational links. Some of these links, such as those among national
groups and international offices of Greenpeace and Friends of the Earth,
are formalized. Others, such as those evident in frequent cooperation at
intergovernmental conferences, are improvised when needed. The United
Nations Environment Programme (UNEP), the UNs Food and Agricul-
ture Organization (FAO), and the International Union for the Conservation
of Nature (IUCN) all have environmentalist-minded staff who have used
organization facilities to keep in touch with each other and to encourage
environmentalist activity. Individual members and foundations have been
able to provide enough money and people to maintain a high level of
Chapter 53 Page 82 September 11, 200819:53

82 International Environmental Issues and Debates

publicity, lobbying activity, and direct action undertaken by individuals

or nongovernmental groups to supplement or substitute for government

Toward the New Management Procedures

Anyone seeking major changes in whaling policies had to contend with a

number of organizational hurdles imposed by the nature, rules, and proce-
dures of the IWC. Unlike many other specialized intergovernmental institu-
tions, the IWC has always been a thing to be captured rather than an actor
in its own right. Its small secretariat, consisting of two people until 1974 and
only a handful thereafter, has not been able to persuade the commissioners
(government delegates) to adopt an organizational ideology that would
dilute national influences or serve as a set of master concepts for framing
regulations.11 Nor has it been able to create a clear enough sense of institu-
tional mission to build its own transnational network of national contacts
able to influence government policies in particular directions.12 The commis-
sioners themselves have not formed a cohesive group. There has long been
a like-minded conservation-oriented group of commissioners led initially
by Australia and Britain and subsequently by the Netherlands as well, and
this group has been capable of following the complicated mathematical
arguments involved in using scientific advice for quota setting. Icelandic,
Japanese, Norwegian, Soviet, and U.S. commissioners have also generally
been able to understand the arguments, with the first four taking a whaler
perspective. Yet many members, particularly the newer ones, send delegates
whose primary duties lie elsewhere and who often have difficulties following
the science-based arguments.13
The regulations defining the seasons for whaling, specifying the maxi-
mum number of whales that can be taken in a season, listing prohibited
gear, prohibiting the taking of immature whales or of pregnant and lactating
females, and requiring report of catch and effort data are specified in the
schedule (annex) of the ICRW. Changing the schedule entails a two-step
process. Proposed changes are first considered by the Scientific Committee,
the Technical Committee, or both. All IWC member governments have the
option of sending representatives to the committee meetings. The Scientific
Committee, consisting of qualified experts, provides advice about the state
of whale stocks and safe levels of taking. The Technical Committee, which
has evolved into preliminary meetings of the commissioners acting as a
committee of the whole, considers the full range of regulatory possibilities
and then recommends policy.14 Although a simple majority vote in either
committee allows its members to forward their proposal to the commis-
sioners, they always seek wider agreement because adoption of a schedule
change requires assent by three-fourths of those present at an IWC meeting.
Particularly ambitious programs of change necessitate amending the
ICRW, and this in turn requires unanimous agreement.15 The ICRW places
Chapter 53 Page 83 September 11, 200819:53

Peterson Whaling 83

significant limits on the ability of the IWC to manage whaling. The IWC
cannot directly limit entry, the policy now widely regarded as the only way
of ensuring optimal utilization of a fishery, by regulating the number of
factory ships and landstations.16 Nor can it encourage governments to do so
by parceling the overall quota into national subtotals.
Because of the many loopholes in ICRW enforcement provisions, control
over IWC decisions has not automatically translated into substantial
influence on outcomes. Whalers can escape IWC-imposed limits in any of
three ways, one derived from general international law and two from the
ICRW itself. First, a states membership in the IWC is optional. No rule of
international law requires states to become or remain parties to a particular
treaty, and no provision of the ICRW can be regarded as having become
customary law binding nonparty states.17 Thus, whalers operating under the
jurisdiction of nonparty states are bound by their countrys rules but not by
IWC decisions. Second, any individual IWC member can legally exempt itself
from applying some or all of the rules. By filing a formal objection under
Article V, paragraph 3, a member state sheds the obligation of enforcing a
rule to which it has objected. By exercising the right to withdraw from IWC
membership, the state can shed all obligations. Third, any IWC member
can fail to enforce even those rules it accepts. Although the organization
receives reports of infractions and may recommend more effective methods
of enforcement, the ICRW leaves the actual detection and punishment of
infractions to individual member governments.
The IWCs monitoring capacity and ability to muster peer pressure
against lax enforcers have been extremely limited. The IWC has never been
authorized to appoint its own inspectors.18 Whaling inspectors were initially
appointed by the territorial or flag state that licensed a particular whaling
expedition or landstation. There was no independent way to determine
whether a lack of reported violations stemmed from compliance or from lax
inspection; suspicions of laxity were rife and hobbled efforts to cooperate.19
A system of appointing an inspector from one country to monitor the activ-
ities of whalers from another country was first proposed in the mid-1960s
but was put into effect only on a limited basis in 1972.20
The IWC decision-making process was strongly affected not only by
the organizations lack of enforcement authority and weak central moni-
toring capacity but also by its members concerns about free riding, concerns
stemming from the fact that whales are a common pool resource open to
all takers. Because no member could be ensured that its own self-restraint
would be matched by that of others, policy tended toward the least common
denominator until ways of applying pressure to recalcitrant governments
and reducing free riding could be found.
IWC decisions in the 194959 period were also affected by a relatively
low sense of urgency. The desire to avoid repeating the obvious overex-
ploitation of the 1930s, coupled with the consensus that simply reviving
the interfirm agreements then in existence would be ineffective,21 had
led governments to agree that explicit policy coordination was necessary.
Chapter 53 Page 84 September 11, 200819:53

84 International Environmental Issues and Debates

They even agreed that cetologists had to be consulted in order to frame

workable policy options. Once these initial agreements were in place, the
matter seemed solved as long as evidence for continued overexploita-
tion was weak. There was evidence of overfishing, particularly in reports
from firms about declining catch per unit of effort,22 but it was sufficiently
weak and contestable that problems of recouping investment and protecting
national industries against foreign rivals loomed larger in many participants
minds. Disgruntled participants had little success in their attempts to change
national debates by mobilizing hitherto uninvolved groups because there was
little public interest in whaling issues.23
The inertia stemming from low concern was reinforced by two features
of the initial ICRW rules that hindered efforts to develop greater restric-
tions. First, the schedule institutionalized a primitive management tool, the
blue whale unit (BWU), since the more sophisticated concept of maximum
sustainable yield had not yet been developed. The BWU was used in the
1930s as the basis of interfirm agreements to limit competition in order to
maintain profits.24 It was not based on any estimate of relative abundance
of different species, for such data was unavailable, but was based instead on
calculations of how much oil could be extracted from an individual of each
species of whale exploited. Although it was sharply criticized by virtually
all cetologists and some government fisheries regulators in the 1950s and
was tacitly modified over the years as bans on taking certain species were
incorporated into the schedule, the IWC discarded it completely only with
adoption of the new management procedures in 1974.
Second, the decision to set the initial annual Antarctic quota at 16,000
BWU created industry expectations that drove investment decisions. Once
significant investments had been made, it proved hard to persuade managers
that lower quotas were needed. The 16,000 gure was proposed in 1944 by
three cetologists Birger Birgersen of Norway, Remington Kellogg of the
United States, and N.A. Mackintosh of the United Kingdom and based
on educated guesses about how whale populations had fared during the
years of light taking during World War II.25 A few cetologists, including
Birgersen, feared the number was too high. Yet the three were being far
more cautious than others involved in or following the negotiations. Most
participants knew that the 193839 catch had been about 30,000 BWU and
felt that any stock depletion had been reversed during the enforced cessation
of commercial whaling in the 194046 period. Industry managers thought
higher takes were possible, and the newly created FAO was urging rapid
resumption of high-catch whaling as the quickest way to meet a severe global
shortage of edible fats and oils.
During the IWCs first decade, industry managers and cetologists devel-
oped distinctive networks through which they attempted to wield influence.
The intergovernmental nature and weak authority of the IWC helped the
managers and hindered the cetologists because managers were able to secure
strong backing from their respective national governments, and their govern-
ments were able to use the threat of objection or withdrawal to secure
Chapter 53 Page 85 September 11, 200819:53

Peterson Whaling 85

deference from other members. Their competing networks are outlined in

Figure 1.
Although government fisheries regulators, industry managers, and cetol-
ogists were distinct communities, they could communicate with one another
because they shared a number of basic beliefs. All three groups agreed on
the basic principle that whales are a renewable resource legitimately used for
human purposes. The problem, then, was one of ensuring rational manage-
ment of a renewable common pool resource. They also agreed on three causal
and policy beliefs: both industry profitability and whale stock viability can
be maintained if whale taking does not exceed safe levels; safe levels of
taking can be determined from certain quantitative indicators by using fairly
simple mathematical models; and managing whaling is essentially the same
as managing a fishery, despite the differences in life cycles and reproductive
rates of whales and fish.
There was, however, a major difference between the perspective of
managers and that of cetologists, and the difference became more significant
as time passed. Managers placed greater emphasis on short-term profit and
recovery of investments. Cetologists were more concerned with long-term
survival of populations at sizes permitting commercial and other whaling.
This difference of emphasis lay at the bottom of the long disagreements that
preceded adoption of lower and stock-specific quotas.
Until the late 1950s, the case for restriction also suffered from the cetolo-
gists inability to present detailed consensual advice or compelling arguments
that uncertainty should always be resolved by erring in the more restric-
tive direction. Cetologists lacked not only the agreed models that would
produce quantifiable estimates of safe catch levels but also the stock size
data needed to make conclusions from such models persuasive to regula-
tors or whalers.26 The concept of maximum sustainable yield (MSY), which
was developed in work on fish population dynamics in the 1950s, appeared
capable of overcoming the first problem.27 However, the second remained
because the necessary data were not available. There were data on catches
and catch per unit of whaling effort, but these could not provide the popula-
tion estimates required for caculating MSY. As Martin Holdgate noted later,
The IWC did not reach the stage of reasonably thorough statistical analysis
until 1960.28 The IWC itself lacked sufficient funds to sponsor research,
and major governments not only opposed supplying the money but some-
times also failed to provide timely or complete catch data as required by the
schedule. Even though the new models provided clear canons of validity,
their use did not lead to consensus in the Scientific Committee. Appointees
from more industry-oriented governments, who did not share the policy
preferences of the epistemic community of cetologists, could still effectively
challenge arguments that there was a compelling need to reduce catches.29
The Dutch member, E.J. Slipjer, was the most adamant opponent of lower
quotas; Soviet and Japanese members were also slow to accept them.30
The influence of cetologists was also limited because their models could
not address the sticky problem of persuading governments and firms to
Chapter 53 Page 86 September 11, 200819:53

86 International Environmental Issues and Debates

reduce industry capacity quickly enough. Firms or governments that had

a large investment in equipment resisted leaving the industry unless and
until they recouped it. Industry managers had sufficient influence with
enough governments to retain their influence over IWC decision making.
The employment, income for suppliers, payments to shareholders, and tax
revenue provided by whaling were all contingent on securing access to a
sufficient number of whales each year to make investments in whaling equip-
ment yield a profit. Until, and in some cases even after, the scientific argu-
ments for reducing catches became compelling, governments of whaling
states tended to accept their industry managers view that greater restrictions
on whaling were undesirable.
Governments not subject to direct industry influence accepted the argu-
ment that overly stiff restrictions would precipitate the filing of objections
or even withdrawal from the IWC. The small landstations and ocean-going
operations functioning outside IWC member states were a constant reminder
of this threat. This government deference was most vividly manifest in the
late 1950s and early 1960s, when there was already considerable evidence
that quotas were too high. They remained high not because governments
were unaware of the cetologists concerns but, rather, because there was
no other way to sustain the interwhaling state agreement on informal
national quotas that provided the real management of Antarctic whaling at
the time.31
The simple overall quota system established in the ICRW encouraged
overinvestment in whaling equipment because each firm had to try to take
as many whales as it could before the season ended.32 Although the IWC set
a maximum season length, operations could be halted earlier if weekly catch
reports indicated that the overall catch quota would be reached before that
time.33 The result was a highly competitive rush in which larger numbers
of catcher boats pursued the same number of whales. While the Antarctic
season had lasted the full 121 days in 194546, it lasted only 58 days in
195556 because of the proliferation of expeditions and catcher boats.34
Firms and government regulators alike recognized that this guaranteed over-
investment in equipment, but the major whaling states preferred to nego-
tiate among themselves rather than give the IWC authority to set national
While these often stormy parallel negotiations over national shares
proceeded, some whalers reacted to the combination of overfishing, overin-
vestment, and declining prices for whale oil by leaving the industry.36 When
the IWC first assembled in 1949, the global whaling industry consisted of
a central oligopoly supplemented by several smaller-scale operations. The
central oligopoly, then accounting for about 70 percent of the total whale
catch,37 consisted of British, Japanese, and Norwegian cartels and a Soviet
monopoly, all of which used factory ships and catcher boats in the Southern
Ocean. These Antarctic operations were initially the main focus of IWC
concern. The smaller operations used landstations or smaller processing
boats in non-Antarctic waters.
Chapter 53 Page 87 September 11, 200819:53

Peterson Whaling 87

British and Dutch whalers had quit altogether by 1965, and the last
Norwegian Antarctic expedition operated in the 196768 season. Both the
British and the Norwegian exits were eased by sales of used equipment
to the still-eager Japanese, who were willing to pay well because portions
of national quota were often transferred at the same time.38 The Japanese
and the Soviets remained in whaling because, unlike their competitors, they
had strong economic incentives to do so. The Soviet fleet received large
state subsidies, and the Japanese had active home markets for whale meat
providing a lucrative alternative to the declining whale oil business.39
The arguments about national quotas also provided the setting for a
major cetologist breakthrough. As part of a broader effort to end the impasse
about the overall quota, the British government proposed that the IWC
appoint a committee of three (later four) experts in fish or whale popu-
lation dynamics drawn from countries not engaged in pelagic whaling in
the Antarctic.40 This would bring in individuals who were better trained
in newer fisheries management techniques than were many on the IWC
Scientific Committee41 and would insulate the negotiating process from
the normal contention within the Scientific Committee and the IWC. The
members of the new committee D. Chapman from the United States, Sidney
Holt from the FAO, F. R. Allen from New Zealand, and (later) John Gulland
from the United Kingdom were leaders of the more quantitatively oriented
wing of the epistemic community and used the newer models of population
dynamics to present far more compelling evidence that immediate adoption
of lower quotas was necessary to ensure the long-term survival of whale
populations at sizes permitting commercial taking.
Growing frustration with the slow pace of change even after the
Committee of Four reported led some cetologists to develop additional
channels of influence in the mid-1960s. The most prominent channel was
provided by the FAO, whose secretariat was persuaded to make continued
cooperation in whale stock assessment contingent on IWC adoption of poli-
cies that more closely reflected the growing scientific consensus on quotas.
A few cetologists, including J. L. McHugh of the United States, advocated
calling public attention to the issue as a way of pressuring governments into
more restrictionist policies.42 But the public interest necessary for the success
of such initiatives did not yet exist. By the time it did in 197072, the cetol-
ogists were being overtaken by the wider environmentalist coalition.
These changes in industry structure and IWC approach led to some
change in the competing networks of influence in the late 1960s. Although
the Japanese and Soviet Antarctic operations supplied 80 percent of the
worlds reduced demand for whale products (mainly oil and meat), smaller-
scale operations often directed at species not initially covered in the ICRW
schedule became a more significant threat to the survival of stocks. They
began receiving systematic attention as the IWC shifted toward regulation
by ocean area and species in the late 1960s. This change made the activi-
ties of nonmember states that licensed whaling most notably Brazil, Chile,
Chapter 53 Page 88 September 11, 200819:53

88 International Environmental Issues and Debates

Panama, Peru, South Korea, and Spain a greater concern. Even so, industry
managers remained able to forestall IWC adoption of the proposals that they
found least acceptable. Cetologists still found their advice diluted by internal
bickering and challenged in the IWC. In the late 1960s, then, the networks of
influence looked very much as they had in the 1950s, as shown in Figure 2.
The rapid decline of stocks in the 1960s, which was evident in the fact
that the level of catches no longer kept pace with the quotas (see Figure 4),
lent urgency to the problem of stock survival. Even the industry managers
recognized the problem and could no longer discredit the cetologists case
for quota reductions on scientific grounds. As James Scarff noted, the 1967
reduction to 3,200 BWU meant that after nearly two decades the IWC had
finally    agreed on a quota that was below scientific estimates of the
sustainable yield of the stocks.43 In this period, managers accepted not only
lower overall quotas and their informal division into national subquotas but
also new methods of defining the quotas. The changes began with banning
the taking of certain species and then extended to setting separate quotas for
individual species or identified distinct populations (stocks) of each species.
As the importance of Antarctic whaling declined and the importance of
whaling in other areas rose, the IWC began debating and using increasingly
precise definitions of stocks. This trend culminated in the adoption of the
new management procedures in 1974.
The new management procedures made the Scientific Committee far
more important because they mandated far more data and more accurate
models of whale population dynamics. The division of all whale stocks into
three classes initial management stocks, sustained management stocks, and
protected stocks rested on comparing the current stock population size to
the size that would supply the maximum sustainable yield (MSY). As refined
in 1975, the three categories were defined as follows: An initial management
stock is one that has not already been subjected to intense exploitation. The
quota for such a stock could be higher than MSY until the total population
was within the parameters established for a sustained management stock, at
which point it would be shifted to that category. A sustained management
stock is one whose population is from 10 percent below to 20 percent above
the size that would support taking at the MSY. Quotas for this category
are set at MSY. A protected stock is one that has been depleted below the
size that would support taking at MSY. The stock is to be protected from
commercial taking until it returns to a size allowing reclassification as a
sustained management stock and resumption of taking.44
The new management procedures did not immediately lead to much
greater restrictions. They did, however, raise the level of scientific argumen-
tation that went into decision making. Before 1974, the members of the
Scientific Committee had given the IWC a unanimous best estimate resting
as often on political as scientific grounds without giving any explicit account
of the criteria actually used in making the estimate. Spurred by pressures
from outside cetologists and from members of the IUCN and the FAOs
Figure 4: Antarctic Pelagic Quotas and Catches, 194950 through 197778
Chapter 53 Page 89

Blue whale





194950 5354 5758 6162 6566 6970 7374 7778


Note: Quotas for the 195960 through 196162 seasons were set by governments of states involved in Antarctic pelagic whaling; quotas for other seasons were

set by the IWC.

Source: J. N. Tonnessen and A. O. Johnsen, The History of Modern Whaling (Berkeley: University of California Press, 1982), p. 749, Table 65.
September 11, 200819:53
Chapter 53 Page 90 September 11, 200819:53

90 International Environmental Issues and Debates

Advisory Committee on Conservation of Marine Renewable Resources, the

Scientific Committee used adoption of the new management procedures to
establish a more open process in which papers were published, commentary
was sought, and the scientific basis of conclusions was made explicit. This
did not abolish political deals, but it made their use more obvious.45 Certain
decisions did inspire controversies.46 However, they were soon overshad-
owed by a far greater one, the controversy sparked by proposals to impose
a moratorium on all commercial whaling.
In the mid-1970s, then, cetologists were trying to expand their newly
acquired influence and thought that IWC adoption of the new management
procedures would help them. Yet the changing political dynamics of whaling
debates soon swept these expectations aside. The broader political environ-
ment was transformed to the disadvantage of both the industry managers
and the cetologists in the mid to late 1970s. The sense of urgency about over-
exploitation spread beyond whalers, cetologists, and IWC member govern-
ments to the wider public in Western countries. Public interest was also
stimulated as changes in ideas about the relation of nature and humankind
began to affect attitudes on a broad range of questions. These developments
encouraged emergence of the third group seeking to affect IWC decisions and
their enforcement: the environmentalists. Lacking insider status in most
countries, they used outsider strategies of bringing local public, transna-
tional public, and transgovernmental influence to bear.

The Zero Quotas

Momentum for an across-the-board moratorium was building even before

the new management procedures were adopted. Two developments of the
1970s, the UN Conference on the Human Environment (Stockholm Confer-
ence) and the Third UN Conference on the Law of the Sea (UNCLOS III),
favored the nascent environmentalist coalitions efforts to influence whaling
policy. The Stockholm Conference adopted three recommendations on the
issue, the most significant of which called for a ten-year moratorium on
commercial whaling.47 While indicative of the environmentalists ability to
influence international conferences, the recommendations themselves were
the least important result of the Stockholm Conference. Far more important
was the impetus it gave to the spread of environmentalist ideas and organi-
zations from the Western industrialized states to other areas of the world.
It spawned UNEP, which took up where the conference left off and worked
to make conference resolutions a reality. Because wide areas of the ocean
remained outside the jurisdiction of individual states and because both the
control of marine pollution and the good management of ocean resources
could be ensured only through intergovernmental cooperation, the UNEP
secretariat quickly developed a particularly strong interest and competence
in ocean issues. It also quickly joined the FAO as a source of well-informed
and significant criticism of IWC policies and procedures.
Chapter 53 Page 91 September 11, 200819:53

Peterson Whaling 91

New legal doctrines refined during UNCLOS III also affected IWC work.
Most important was the general agreement on the exclusive economic zone
(EEZ), which extended coastal state jurisdiction to all resource activities in
waters lying 200 miles or less from shore. On the one hand, this change
complicated IWC efforts by encouraging a larger number of governments to
assert unilateral rights to managing certain whale stocks. The governments
of Chile, Ecuador, and Peru, which had asserted 200-mile jurisdiction claims
since the late 1940s, had stayed out of the IWC lest acceptance of its authority
weaken their claims,48 and now other governments adopted similar attitudes.
The Canadians and the French joined the west coast Latin Americans in
opposing extension of IWC management authority to small cetaceans in
the EEZ and began asserting closer unilateral control of whaling operations
off their coasts. On the other hand, the change facilitated application of
intergovernmental pressures to whaling states by giving others a valuable
asset for issue linkage: control over access to large fish stocks.
UNCLOS III disappointed environmentalist hopes by failing to extend
the common heritage of mankind concept to marine creatures in the high
seas. Doing so would have laid the groundwork for replacing all existing
international fisheries and marine mammal management bodies with one
UN agency having exclusive control over access to resources. Environmen-
talists believed that any such body would be more effective in enforcement
than the existing management bodies and would have a strong international
secretariat likely to be sympathetic to their views. Yet government reluctance
to establish any such global management body was obvious even in the early
stages of negotiations. The common heritage concept was applied explicitly
only to exploitation of deep seabed mineral resources. Marine mammals and
fish outside the EEZ remained shared resources to be managed cooperatively
by interested governments in whatever organizations they established.49 This
decision meant that the IWC was not displaced by a UN agency, though
the common heritage idea was used to inspire and justify nonwhaling state
accessions to the ICRW in the early 1980s.50
Throughout the 1960s and 1970s, the broad community of marine biol-
ogists and resource economists involved in fisheries management continued
to refine their analytic tools. Some of their work indicated that implementing
the new management procedures would require more data and probably
more readiness to err on the side of caution in setting quotas than many cetol-
ogists had thought in the early 1970s. A later stream of research focusing
on multispecies and ecosystems approaches suggested that even the new
management procedures were likely to miss important dynamics because
they relied on single-species models.
This new work reopened the fundamental question of how to define
rational levels of taking. Starting in the late 1960s and gathering momentum
in the early 1970s, researchers severely criticized the reliance on MSY as
the main criterion for determining quotas. A series of studies showed that
the models used to calculate MSY failed on the biological side to account
for all the population responses to changed levels of taking and failed on
Chapter 53 Page 92 September 11, 200819:53

92 International Environmental Issues and Debates

the economic side to properly discount the value of future earnings.51 The
idea of discounting subsequently came under environmentalist criticism as
encouraging depletion of stocks.52
In early 1975, two major conferences of leading fisheries biologists and
cetologists produced recommendations that MSY be replaced by a four-
fold definition of rational taking that would encompass the following: the
maintenance of the ecosystem in a state that would maximize current and
long-term options and minimize the risk of long-term or irreversible adverse
effects of taking; the stipulation of a margin of safety to compensate for
uncertainty of data and species reactions to environmental change; the assur-
ance that conservation of one species would not lead to wasteful uses of
other resources; and the assurance that adequate assessment and monitoring
would take place before and during serious exploitation of any species.53 In
the late 1970s, marine biologists began developing multispecies models that
more accurately portrayed interactions among various stocks of marine life.
Even though these newer models could not be applied immediately because
of the lack of sufficiently powerful computers, preliminary results indicated
that the older single-species models tended to err on the optimistic side in
the calculation of biological MSYs.54
The cetologists were aware of and absorbing these developments into
their own canons of reasoning and causal beliefs. Yet the speed and intensity
of debate surrounding these innovations fractured the cetologist commu-
nity and the wider marine biologist community as well. Elements of the
cetologists shared belief system were shaken as new ideas cast doubt on
established observations and as equally qualified scientists contended over
which models to use in interpreting old and new data. In 1963, there had
been consensus that immediately limiting catches to 4,000 sei and 12,000
fin whales a year while continuing the ban on taking blue and humpback
whales would permit Southern Ocean stocks to recover in thirteen years.55
By 1974, the state of the debate among scientists was such that cetologists
moving toward the environmentalist coalition criticized the IWCs decisions
to adopt the new management procedures and to set even more restrictive
catch limits as woefully insufficient for protecting whale populations. All
eight species of great whales were widely regarded as endangered,56 and this
helped the environmentalist coalition build strong public support.
Arguments among whalers, governments, and cetologists appeared
initially to center around the interpretation of uncertainty. The environ-
mentalists holistic modes of thinking encouraged interpreting uncertainty
as a mandate for severe limits on whaling.57 While cetologists and regu-
lators working with the new management procedures tried to distinguish
among more and less threatened whale stocks, the environmentalists used
uncertainty to support arguments that all stocks should be treated as equally
endangered and left alone.58 Later, arguments appeared to spill over into
the realm of principles as well. A few cetologists, most notably Sidney Holt,
adopted rather preservationist attitudes that entailed attacking not only the
data and models but also the very idea that resource management is the
Chapter 53 Page 93 September 11, 200819:53

Peterson Whaling 93

proper framework for thinking about whales. The preservationists approach

is well illustrated in Patricia Forkans 1980 argument that securing better
resource management is not the only reason to support a moratorium: We
think it is an ethical question and whales should not be killed because it is
unethical to kill them and that it is a growing concern among many people
in the world that mans attitude toward wildlife in general should change,
that [whales] arent just there as a product to be used but that they have
other purposes; we can learn from them.59
The divisions became so deep that cetologists did not operate as a unified
group in the 197482 period. Internal disagreements about which model to
use and how to interpret the data made it difficult to give the unified advice
necessary to counter the influence of either the industry-oriented members
of the IWC Scientific Committee or the environmentalists. The split on prin-
ciple meant that qualified scientists were ready to offer evidence in favor
of highly contrasting policy preferences when conservationist and preser-
vationist stances led in different directions. The range of disagreement was
limited because industry managers realized that highly consumptionist pref-
erences could not be sustained. However, they still preferred more taking
than conservationists regarded as wise. Even so, the arguments among cetol-
ogists meant that they could not frame the policy choice by expert applica-
tion of agreed canons of validity to agreed models and data on the basis of
shared principles and policy preferences. The field was wide open for highly
political and often highly public contention.
This contention stemmed from the intense competition between the group
of industry managers and the group of environmentalists for influence on
government officials who defined and enforced policy. The environmental-
ists used a combined national and transnational strategy from the start. The
industry managers found themselves compelled to explore more transna-
tional strategies because their traditional focus on national ones quickly pro-
ved insufficient. The ultimate shape of their efforts is illustrated in Figure 3.
The competition for influence can be understood best by dividing IWC
member states into four rough categories: (1) states with strong commercial
whaling industries; (2) the United States; (3) states subject to cross-pressures
from environmentalists on one side and small-scale whalers (commercial,
subsistence, or both) on the other; and (4) states open to strong environ-
mentalist influence for lack of competing lobbies. The United States, though
subject to domestic cross-pressures on certain whaling issues, merits a cate-
gory of its own because the U.S. governments willingness to use unilateral
economic measures against other governments on whaling issues has made
it so central to outcomes that both rival coalitions have sought to influence
its policy.
Japanese whalers have been the pivotal members of the industry coalition
because of their influence on several governments. At home, the whaling
industry has relied on strong ties to the Liberal Democratic party, which
has formed the government continuously since 1951. Neither local nor
foreign cetologists or environmentalists have been able to affect Japanese
Chapter 53 Page 94 September 11, 200819:53

94 International Environmental Issues and Debates

government policy to any significant degree.60 The strength of the industry

position was most vividly signaled by the appointment of an industry leader
rather than a government fisheries regulator as Japans IWC commissioner
and by the dispatch of industry representatives rather than cetologists to the
IWC expert working group on improving the new management procedures
in 197980.61
Foreign environmentalists have lacked leverage with the Japanese govern-
ment. Popular Western arguments against treating whales as resources have
had little resonance in Japan, where traditions combine reverence for animals
with acceptance of their taking in a way that makes the Western preserva-
tionists goals almost incomprehensible.62 The conservationists notions have
had broad resonance, and the cetologist community has included Japanese
members, though they have seldom been part of the Japanese delegation
attending IWC meetings. Western environmentalists efforts to encourage
similar groups in Japan have suffered from a number of obstacles, including
lack of resources for the effort, language barriers, local opposition, and a
long history of mutual cultural misunderstanding. In 1980, for example,
one Westerner describing Japan wrote that there is an almost total lack
of concern for international environmental problems, even those created by
Japanese economic activity.63 Similarly, many Japanese have viewed the
Westerners attitudes on whaling as ethnocentric or downright racist and
regarded their actions as unreasonable.64
The Japanese whaling industry has influenced foreign governments in
one of two ways. First, the industry developed a network of connections with
Brazilian, Chilean, Philippine, South Korean, Soviet, and Taiwanese opera-
tions supplying the Japanese market for whale meat. These suppliers initially
included shadowy operations that were active in the 1970s under various
non-IWC state flags of convenience. They were later dropped by Japan when
the extent of their contacts became public knowledge as a result of envi-
ronmentalist efforts to trace and harass them out of existence. Second, the
Japanese government was persuaded to attempt to counter the environmen-
talist influence in a number of third states through promises of additional aid
or threats to cut existing aid.65 Yet the Japanese whalers were not the only
opponents of greater restrictions. Icelandic, Norwegian, South Korean, and
Soviet whalers also sought to continue their existing activity. Each whaling
group was able to secure the support of its own government. Soviet whalers
did so most directly, since they operated a state-owned enterprise reporting
to the Ministry of Merchant Marine and provided a source of edible oil
that supplemented the ever-unreliable domestic supplies without expending
scarce foreign currency. The others relied on traditional forms of economic
interest group lobbying. Whenever environmentalist policy stances appeared
to threaten subsistence whaling as well, the Danish government could be
persuaded to protect Greenland Inuit whalers.
Even before 1982, whalers and environmentalists both agreed that the
U.S. government was the single most important actor on the whaling scene.
Though entitled to only one vote in the IWC, it had what the IWC and any
Chapter 53 Page 95 September 11, 200819:53

Peterson Whaling 95

other single government or group of governments lacked: the ability and

the will to enforce restrictions against others by invoking the trade sanction
provisions of various domestic laws. Sanctions began with closing the U.S.
market, which then accounted for 25 percent of world demand, to whale
oil and other whale products under the Endangered Species Act in 1970.66
These sanctions lost significance as international trade in whale products
decreased. In any case, they were never very effective against Soviet and
Japanese whalers, whose products were all used at home. The ability to
influence trade in Japan and the Soviet Union was acquired with passage of
two amendments to the U.S. Fisheries laws. The Pelly amendment in 1973
allowed for an embargo on all fish and wildlife products from any country
pursuing policies that weakened international fish and wildlife conservation
measures.67 The Packwood-Magnuson amendment in 1979 supplemented
the Pelly amendment by requiring that once a country was certified as dimin-
ishing the effectiveness of the ICRW, it must lose at least 50 percent of
its fishing quota in the U.S. EEZ.68 This provision went beyond Pelly by
attempting to make sanctions automatic and by cutting off access to supplies
of fish rather than markets for already-caught fish.
Environmentalists were quick to see the potential of these amend-
ments and organized to maximize their influence with the U.S. govern-
ment. Initially, they faced few organized competitors and acquired consider-
able influence over policymaking. Later, the counterefforts of other groups
eroded their influence. In 1971, the overall responsibility for U.S. whaling
policy passed from the hands of a government-employed cetologist, who
served as IWC commissioner and was assisted by whatever office of the
State Department was handling ocean-related issues, to the National Oceans
and Atmospheric Administration (NOAA).69 NOAAs head became commis-
sioner and made policy with the assistance of an interagency group consisting
of members from the State, Commerce, and Interior Departments.70 As
interest in whaling issues rose through the 1970s, the interagency group
expanded to include members of Congress and selected private citizens. Both
the environmentalists and their opponents, particularly the Alaskan Eskimos
seeking to continue their traditional hunting of bowhead whales, secured
places in the interagency group and in the U.S. delegations to the IWC meet-
ings.71 Environmentalists had stronger influence in the House Committee
on Foreign Affairs, while Eskimos and Pacific Coast fishers had stronger
influence in the other key congressional committee, the House Committee on
Merchant Marine and fisheries. The shape of the competition for influence
is diagrammed in Figure 5.
U.S. policy followed a generally environmentalist lead in the early and mid
1970s. The U.S. government supported the new management procedures of
the IWC but was also an early proponent of a moratorium. It used threats of
trade sanctions to deter several formal objections to lower IWC quotas on
specific whale species in the late 1970s, and it led Peru to withdraw its
objection to zero quotas on several species and to the overall zero quota
Chapter 53 Page 96 September 11, 200819:53

96 International Environmental Issues and Debates

Figure 5: Channels of Influence within the United States, 1980s

US delegation
to IWC

House Committee on
Foreign Affairs
Department of Inter
Commerce agency House Committee on
group Merchant Marine
and Fisheries
Department of
Department of
the Interior
Marine Mammals
Commission Governments of
whaling states

Environ- Pacific Coast

mentalists fishers
Government agencies or committees
Private groups
Foreign governments

decision in 1983.72 By similar threats, the U.S. government also helped induce
a number of nonmember whaling states to join the IWC in 197880.73
Yet environmentalists soon found themselves engaged in sharper compe-
tition for influence. Continuing criticism of environmentalist stances by
cetologists outside the environmentalist coalition demonstrated that there
were good scientific grounds for doubting that all species of whale
were equally endangered. More important, the Eskimos and then the
Pacific Northwest fishers began opposing certain environmentalist policy
Controversies between environmentalists and Eskimos first became acute
in 197677. The environmentalist concern that Eskimo hunts would lead to
extinction of the bowhead, by then one of the rarest of whales, was fueled by
evidence that Eskimos were not confining themselves to traditional methods
of taking and were striking a considerably larger number of whales than they
were landing.74 Eskimos were able to muster enough support in the Interior
Department and Congress to ensure that domestic law permitting the hunt
Chapter 53 Page 97 September 11, 200819:53

Peterson Whaling 97

was not changed and that U.S. delegations to the IWC were instructed to
protect Eskimo whaling rights. The Eskimos attempted to force the U.S.
government to file a formal objection to the IWCs 1977 decision to impose
a zero quota on bowheads through the courts.75 The lawsuit failed, but the
U.S. government did seek a change through bilateral discussions. Whaling
states were able to exploit the resulting contradictions of U.S. policy until
1982, when the IWC formally adopted a distinction between commercial
and subsistence whaling.76 This allowed the U.S. government to pursue a
moratorium on commercial whaling with renewed vigor.
The first phase of environmentalist effort to secure a moratorium focused
on raising a three-fourths vote from among the existing IWC membership.
Once the United States was committed to the policy in 1974, the environmen-
talists focused their efforts on nonwhaling or cross-pressured states, such as
Australia, Canada, the Netherlands, New Zealand, and Panama.
The environmentalists enjoyed a number of successes. The Dutch govern-
ment, a formerly active IWC participant with a multimember delegation
to the organizations meetings and active participation in the Scientific
Committee, had reduced its participation to a single representative to the
meetings in 1966 after its last whaling operations wound up.77 Environ-
mentalists persuaded the Dutch to become active again in 1976 and to
support the moratorium in 1977. In Australia, antiwhaling campaigns by
supporters of Project Jonah won the backing of 70 percent of the popu-
lation and of the National Liberal party. When the party resumed power
after the 1977 election, Australian policy swung in a promoratorium direc-
tion.78 In Panama and Canada, the political competition was closer. Never-
theless, during the late 1970s, the environmentalists appeared to be making
headway. The Panamanian government was persuaded to include an envi-
ronmental activist, Jean-Paul Fortrun-Gouin, in its delegation to the IWC,
though it disavowed his moratorium proposals.79 The Canadian government
had banned any renewal of commercial whaling in 1972,80 and the way
looked clear for environmentalist influence to prevail.
The environmentalists were also heartened by the 1977 reorganiza-
tion of the IWC Scientific Committee. Advisers from UNEP, FAO, and
IUCN were added to the nationally appointed members, and meetings
were opened to scientist observers from nongovernmental organizations.
Although the changes were intended to increase the weight of indepen-
dent scientific advice, they transformed the Scientific Committee in other
ways as well. In the early 1950s, many IWC member states had failed
to appoint experts to the committee, but appointments had subsequently
rose to the point that most states participated and the total attendance
hovered around twenty-four.81 By 1982, Scientific Committee meetings were
attracting over one hundred participants as more IWC members sent repre-
sentatives, some increased the number of representatives they sent, and scien-
tists from nongovernmental organizations were invited to observe.82 This
expansion of committee memberships pleased environmentalists, since it
gave them the opportunity to participate and present their arguments and
Chapter 53 Page 98 September 11, 200819:53

98 International Environmental Issues and Debates

alternate stock assessments; it did not please others concerned that the size
made meetings unwieldy.
The environmentalist successes at recruitment in the late 1970s were,
however, more than matched by whaling state accessions to the IWC. This
change, which brought in Chile, Peru, South Korea, and Spain for the first
time, cannot be credited solely to the industry coalition. Japanese whalers
did seek the accessions of these states, but so did the U.S. and other restric-
tionist governments on the grounds that the new management procedures
would not be effective if they remained outside.83 Given that whaling states
still had a slight edge over nonwhaling states in the IWC membership in
1979 (see Figure 6), the environmentalists intensified their effort to interest
Third World governments in joining the organization. The governments they
approached had no direct interest in whaling but considerable interest in
environmental issues or in expanding the common heritage principle to all
high seas resources. The recruitment effort yielded considerable success. The
Seychelles government even became an active participant in the transna-
tional efforts of the environmentalists. It allowed them to use its Washington
embassy as a base for lobbying operations84 and appointed the preservation-
minded Sidney Holt to its delegation after his retirement from FAO. Most of
the governments recruited were more passive; some even needed the incentive
of environmentalist funding of their delegations.85
Yet before additional recruits made passage of a modified morato-
rium possible in 1982, the environmentalists suffered some setbacks with
governments subject to cross-pressures. Panamanian policy was wobbling
in 197879 as both environmentalist and whaler pressures were brought
to bear.86 The Panamanian government decided to deal with the problem
by withdrawing from the IWC in 1980. In Canada, the environmentalists
success at securing a ban on sealing off Newfoundland rebounded against
them. A coalition of Newfoundland fishers and Atlantic Coast Inuit, two
groups that used sealing to supplement their meager incomes, reacted bitterly
to the prospect of another ban. The coalition exerted strong pressures on
the Department of fisheries and through Parliament. The situation became
so uncomfortable that the Trudeau government chose to withdraw from
IWC in 1981 rather than deal with another year of lobbying.87 While the
Canadian government was also concerned about the effect that extending
IWC management to smaller cetaceans would have on its claims to juris-
diction in the EEZ,88 the intensity of the domestic cross-pressures was the
immediate trigger of the 1981 decision.
Even so, the combination of additional members and timely threats of
U.S. sanctions against recalcitrant governments resulted in the 1982 adop-
tion of zero quotas for commercial whaling.89 The decision fell short of
many environmentalists desires. It was never formally designated a mora-
torium, though it had that effect and is generally called that. A four-year
delay to permit the winding down of existing enterprises meant it did not
take full effect until 1986. Zero quotas were to be effective for ten years
at most, with the possibility of reconsideration arising any time after the
Figure 6: IWC Membership, 194989
Chapter 53 Page 99

Number of



Nonwhaling states
Whaling states






1949 59 69 79 89

Sources: Patricia Birnie, The International Regulation of Whaling (New York: Oceana, 1985); G. P. Donovan, Thirty-Eighth Annual Meeting of the International

Whaling Commission, Polar Record 23 (January 1987), p. 437; Donovan, Thirty-Ninth Annual Meeting, Polar Record 24 (October 1988), p. 66; Donovan,
Fortieth Annual Meeting, Polar Record 26 (October 1989), p. 250; and Donovan, Forty-first Annual Meeting, Polar Record 26 (January 1990), p. 233.
September 11, 200819:53
Chapter 53 Page 100 September 11, 200819:53

100 International Environmental Issues and Debates

Scientific Committee provided the new stock assessments due in 1990. It was
also hedged by exceptions for subsistence whaling and taking under special
scientific permit.
Nonetheless, a clear decision had been made. The whaling issue then
became a matter of enforcing the zero quotas while awaiting the better
scientific advice promised by the program of comprehensive stock assess-
ments. This program has taken longer than expected, but the exercise has
helped cetologists improve their models, determine what information is
necessary for adequate management, and do some of the surveys needed to
establish species life cycles and stock sizes.90 The work has helped the conser-
vationist cetologists regroup and refine the manner in which they present
scientific advice. These changes, together with the more open proceedings
of Scientific Committee meetings and the changes in presentation of advice
to the IWC discussed earlier, have placed cetologists in a better position to
influence the policy process.
The first enforcement difficulties involved slow Japanese and Soviet ending
of operations. These were soon supplemented by several countries evasion
of the zero quotas by wider issuance of permits to take whales for scientific
research. Controversies about whether zero quotas should continue to
cover all whales arose in the late 1980s because of reasonably strong evidence
that certain stocks, especially North Atlantic and North Pacific minke and
fin whales, were sufficiently large to permit some taking. Two major changes
in the political arena another shift in the composition of the IWC and the
weakening of U.S. influence over other states made strict maintenance of
zero quotas more difficult than expected.
The voting balance in the IWC changed again in the mid-1980s. Anti-
whaler voting strength was reduced as a number of the newly recruited
governments became inactive. Dominica and Jamaica stopped sending dele-
gates, while more countries failed to pay their assessments on time. Thirteen
members were in arrears by June 1986, with nine all developing states
ineligible to vote because they were more than twenty-one months in arrears.
The IWC, unlike most intergovernmental organizations with similar rules,
began applying the penalty rather strictly.91 While environmentalists have
complained because strict application erodes their influence over decisions,92
most member governments, even the conservation-minded, insist on the prac-
tice. The lower participation by the newer nonwhaling members has brought
the number of reliable antiwhaling votes below the three-fourths level needed
to modify the zero quotas or determine other rules for whaling. So far,
however, environmentalist fears that the changed membership would lead
to rapid abandonment of the moratorium have not materialized; the IWC
not only retained the moratorium after the 1990 review but also rejected the
Scientific Committee recommendation that Iceland and Japan be allowed
quotas of minke whales.93
Environmentalist influence over implementation waned because of
changes in both the effectiveness of the U.S. governments enforcement
weapons and the intensity of U.S. policy in the 1980s. The Soviet Union was
Chapter 53 Page 101 September 11, 200819:53

Peterson Whaling 101

largely unaffected by U.S. regulations during most of the decade. Soviet fleets
were banned from the U.S. EEZ as a sanction against Soviet intervention in
Afghanistan from early 1980 until 1984, making the Packwood-Magnuson
amendment irrelevant temporarily.94 The low level of fish imports from the
Soviet Union made the Pelly amendment irrelevant. The 1982 amendments
to the Fisheries Conservation and Management Act, which provided for
Americanization of the fisheries, meant that the Packwood-Magnuson
provision would become irrelevant to almost all countries in six years
because the only foreigners permitted to fish in the U.S. EEZ would be part-
ners in joint ventures with U.S. firms. The Pelly amendment became less
effective against some countries, most notably Japan, because of changes in
fish product trade patterns. The Japanese were aware of this in the early
1980s.95 By the middle of the decade, the value of U.S. fish exports to Japan,
either directly or through joint ventures, was almost twice that of U.S. fish
imports from Japan.96
The political balance within the United States was also changing. Envi-
ronmentalists faced stiffer challenges both from Pacific Coast fishers and
from foreign whalers. Pacific Coast fishers were heavily involved in joint
ventures with the Japanese and came into the policy fray when proposed
trade sanctions against Japan threatened to disrupt their livelihood. They
formed a fairly powerful economic group and had influential connections in
the NOAA and the House Committee on Merchant Marine and Fisheries.
Foreign whalers began organizing direct counterefforts, either by hiring
lobbyists in the United States or by getting help from their own govern-
ments. The Japanese whalers were the first to do so and remain the most
active. They recruited high-powered lobbyists, such as former NOAA admin-
istrator Richard Frank; found sympathizers in Congress, such as Representa-
tive Mervyn Dymally of California; and exerted diplomatic pressure through
their embassy and sympathetic officials in the Far Eastern Bureau of the
U.S. State Department. The Norwegians also hired lobbyists.97 Since neither
they nor the Icelanders could threaten retaliation against U.S. exports, they
invoked other linkages, particularly participation in the activities of the
North Atlantic Treaty Organization.98
The new policy constellation soon demonstrated its effectiveness. The
environmentalists failed in their effort to secure a congressional resolution
urging application of the Packwood-Magnuson provision against Japan,
Norway, and the Soviet Union, the three states still objecting to the mora-
torium in 1985.99 Although U.S. government pressure helped persuade the
Japanese to withdraw their formal objection to the moratorium in 1985
and encouraged the Japanese and Soviets to end commercial whaling in
198788,100 the pace of change was too slow and the interim concessions too
many in environmentalist eyes. When the environmentalists tried to force the
administration into stronger action through litigation in sympathetic federal
courts, they encountered only occasional success.101
The most extended post-1986 discussions centered on the efforts of
whaling states to protect certain operations by using the cover of scientific
Chapter 53 Page 102 September 11, 200819:53

102 International Environmental Issues and Debates

permits, by seeking reclassification of some small-scale coastal opera-

tions as aboriginal/subsistence whaling, or by proposing that the IWC
adopt a separate category for small-scale commercial taking. Japanese
whalers in particular muddied the issue by claiming that certain scientific
programs required data that could only come from dissections of whales, a
claim resisted by environmentalists and contested by cetologists including
some from Japan and Iceland.102 Only with great difficulty were the
Icelandic, Japanese, and Norwegian governments persuaded to reduce or
end these programs. Although governments were supposed to consult the
IWC Scientific Committee before issuing the permits, the committee lacked
authority to deny or limit issuance.103 Consultations with the committee had
so little effect that the U.S. government revived its proposal of requiring
prior Scientific Committee approval. The proposal failed, and the whaling
coalition was able to maintain the existing practice of review and comment
by the full IWC body.104 This does not guarantee success, but it does leave
decisions in an intergovernmental forum where the full range of political
considerations can be taken into account and become transparent. This has
not yet worked to the whalers advantage. The zero quotas were all extended
in 1990. Despite the Scientific Committees view that stocks of minke whales
off Iceland were robust enough to permit such taking, Icelands request for
a quota of two hundred a year was also turned down. Icelandic delegates
charged that the IWCs decision was politically driven, and Icelanders talked
of withdrawing from the organization.105


Whaling and the institutions established for its management would seem to
be an issue and a policy arena in which an epistemic community of rele-
vant experts enjoys great influence over policy. Whaling involves activity
that is highly amenable to analysis using technical forms of reasoning. The
activity can be measured and observed as it proceeds, divided into discrete
acts at particular times, and adjusted upward or downward in intensity as
circumstances merit. In addition, the institutions established for regulating
the activity made formal provision for securing expert advice. Both the nature
of the problem and the institutional structures established for dealing with
it thus appeared favorable to promoting epistemic community influence.
Yet the epistemic community of conservation-minded cetologists only
briefly enjoyed predominant influence over policy. Most of the time, the
influence of cetologists was outweighed by that of other groups, the industry
managers until the mid-1960s and the environmentalists after the mid-1970s.
The explanation for this surprising situation can be found partly in the insti-
tutions, partly in the political process that grew up through and around them,
and partly in the dynamics of the epistemic community itself.
The regulatory regime created by the ICRW included institutions too
weak to gain preponderant influence over outcomes. The IWC itself was not
Chapter 53 Page 103 September 11, 200819:53

Peterson Whaling 103

a supranational body; it consisted of representatives sent by each partici-

pating state. Although most states sent government fisheries regulators, there
was no rule banning the appointment of whaling industry managers, cetol-
ogists, environmentalists, or anyone else. The decision-making procedures
intruded little on national autonomy. Unanimous consent was required to
amend the main provisions of the ICRW, and a three-fourths majority was
needed to change the specific regulations concerning quotas, seasons, and
takings outlined in the schedule. The burden of argument fell squarely on
those who desired change. Since the initial rules included few restrictions,
this placed those seeking greater restrictions at a disadvantage.
The provision establishing a Scientific Committee created the possibility
but not the guarantee that scientific advice would play a major role in shaping
IWC decisions. While the IWC is supposed to take Scientific Committee
conclusions into account, it may also consider other factors. In addition,
there is no guarantee that members of the cetologist epistemic commu-
nity will dominate the Scientific Committee. Participating governments are
supposed to send qualified experts, but these experts can be drawn from the
ranks of whaling industry employees or environmentalists rather than from
any scientific epistemic community.
Policy was pushed toward the lowest common denominator by the weak
enforcement provisions of the ICRW. They gave opponents of a particular
decision ample opportunity to weaken or nullify it by denouncing the ICRW,
filing a formal objection to the decision, or failing to enforce it against their
own whalers. No government both willing and able to champion IWC deci-
sions against all comers appeared until the early 1970s. Until that time, then,
decisions were heavily constrained by the lack of enforcement.
Both the need to muster a three-fourths vote in favor of greater restric-
tions and the need to find ways to ensure that national fisheries agencies
would enforce the adopted restrictions meant that the politics of whaling was
focused mainly in the national capitals. Transnational activity involved iden-
tifying, recruiting, and encouraging like-minded people to enter the policy
process in their own countries. As long as whaling remained a low-salience
issue followed only by fairly restricted circles of government officials, cetol-
ogists, and whalers, the industry managers were able to guide the policy
of more participating governments than were the cetologists. Since most of
the participating governments had active domestic industries, whalers could
count on having a comfortable majority in the IWC.
Whaling issues reached a broader public in many countries after 1970.
The new and growing consciousness of environmental issues included con-
cern for wiser management of all resources. This created a situation favor-
able for adoption of greater restrictions because it was clear by then that
whale populations were being severely depleted. Initially, this rise of concern
helped the cetologists; conservationist arguments that had been made for a
decade or more won more attention from governments and whalers alike.
The process was uneven, however. Some governments and whalers resisted
change. Yet other governments came around to conservationist views while
Chapter 53 Page 104 September 11, 200819:53

104 International Environmental Issues and Debates

other whalers reduced or ended their activities because they read the combi-
nation of economic signals from declining whale oil prices and biological
signals from decreasing catches as warnings to find alternative activities.
Although this conservationist mood was being challenged in the early 1970s,
it was still strong enough that most IWC members were willing to attempt
the more refined regulatory scheme outlined in the new management proce-
This greater conservationism was soon supplanted by a preservationist
opposition to treating whales as resources whatever the quality of manage-
ment procedures. Preservationist attitudes became prominent in large seg-
ments of the environmental movement in several countries by the mid-1970s.
The endangered status of some whale species, research (later challenged)
indicating that whales and other cetaceans possessed intelligence compa-
rable to that of humans,106 and the spread of animal rights movements
all created widespread public opposition to whaling. Save the Whales
became a powerful rallying call that attracted ever-wider public support in
many Western states and brought the issue to the attention of officials in
several Third World countries. The environmentalists quickly did what some
cetologists had been thinking of doing: arousing members of the public to
exert pressure on their own government to close down whaling operations,
support the proposed moratorium on commercial whaling, and exert pres-
sure on other governments to do the same.
Once many vocal environmentalists adopted preservationist attitudes, the
stage was set for a far more bitter policy struggle. The gap between consump-
tionist and conservationist views was significant, but it could be bridged
because both accepted the legitimacy of treating whales as resources. The
gap between consumptionist and preservationist views was unbridgeable,
since preservationists completely rejected the legitimacy of treating whales as
resources. Although conservationists and preservationists could often agree
on the same policy, such as placing zero quotas on endangered species, the
fact that they could not agree on the same principle made it difficult for them
to work together comfortably in the long term.
While some environmentalists attempted to influence outcomes directly
by attacking or harassing whaling operations and by sponsoring consumer
boycotts of products from whaling states, most realized that they needed the
resources commanded by governments to make their preferences the effective
outcome. While they could gain preponderant influence over policy in many
states, the governments of major whaling states remained relatively immune
to their influence. The environmentalists did recruit enough additional like-
minded governments to the IWC to ensure a three-fourths majority of anti-
whaling governments. This did not entirely solve the problem, however,
because even an antiwhaling IWC was unable to enforce its decisions.
Capturing the preponderant power was thus crucial to environmentalist
success in the late 1970s and early 1980s.107 Threats or impositions of U.S.
trade sanctions were the main weapon available against those governments
impervious to local environmentalist influence. As environmentalist control
Chapter 53 Page 105 September 11, 200819:53

Peterson Whaling 105

over U.S. policy was more strongly challenged after 1980 and as the effec-
tiveness of U.S. trade sanctions waned, environmentalists found steering the
politics of whaling more difficult.
Environmentalists also benefited from the internal fracturing of the epis-
temic community of cetologists. This community had the ill fortune of
becoming deeply divided on major scientific questions at the very time envi-
ronmentalists were campaigning for the moratorium. Cetologists retained
their principles and policy preferences but engaged in serious disputes about
which models to apply and how to handle uncertainty in the data.108 The data
concerning whale populations had steadily increased since 1949 but were
still inadequate to satisfy the requirements of the newer multispecies and
ecosystems models. Thus, even those who agreed on the same model argued
about what to assume when observational data were missing. Arguments
concerning the models, which involved questions about canons of validity
and causal beliefs, and disputes concerning the data created a situation in
which it was impossible to offer consensus expert advice.
The cetologists never fully controlled the IWC Scientific Committee.
Whaling state members, particularly the Japanese, were usually industry
employees rather than the academics and government researchers who were
part of the epistemic community. The 1977 decision to open the Scientific
Committee not only to observers from intergovernmental organizations but
also to those from nongovernmental organizations gave environmentalists
more direct lines into the committee. The new Third World members sending
people to the committee sometimes chose environmentalists as well. It is
likely that even a united cetologist community would have been unable to
use the Scientific Committee as a mechanism for framing policy choices in
the 1970s and early 1980s. The cetologists internal divisions ensured that
this became the case.
If the management of whaling under the ICRW is viewed statically, one
decision at a time, the result is a picture of international cooperation only
briefly guided by an epistemic community. If viewed dynamically, the picture
changes somewhat: even when the cetologists did not determine policy, they
did set certain limits to policy change. In the 1940s and 1950s, cetologist
influence prevented policy from swinging to the most intensely consump-
tionist formulations. In the early 1960s, it helped encourage the greater
emphasis on conservation accepted by virtually all involved. In the 1980s, it
acted as a counterweight to the most extreme preservationist views. Although
preservationists were very influential within the environmentalist coalition,
others ready to accept limited resource use were more influential in the
broader policy process, if not always within the environmentalist coalition.
The cetologists continuing efforts mean that if the zero quotas are revised,
policy will not revert to the highly consumptionist bias of the 1950s but,
rather, to the conservationist perspective that was institutionalized in the
new management procedures.
Current IWC policy on commercial whaling can be read as consistent
with either a conservationist or a preservationist approach. Governments
Chapter 53 Page 106 September 11, 200819:53

106 International Environmental Issues and Debates

of whaling states have been brought around to conservationist readings; a

few of the still-active governments that acceded to the IWC in the 1980s
prefer preservationist ones. However, the preferences of a number of major
governments including that of the United States are not clear. Their
policies could go either way. If they move in a conservationist direction,
cetologists will again become more important in the policy process. Whether
they will wield determining influence will depend on whether the cetologists
active in the 1970s or a younger generation of them are able to recreate an
epistemic community sharing conservationist principles, preferences for poli-
cies that protect the truly endangered species while allowing careful taking
of the others, canons of validity, and causal beliefs consistent with both the
data and the policy preferences.


1. United Nations, Treaty Series: Treaties and International Agreements Registered or

Filed and Recorded with the Secretariat of the United Nations, vol. 161 (New York: United
Nations, 1953), pp. 7282.
2. See, for example, James E. Scarff, The International Management of Whales, Dolphins,
and Porpoises: An Interdisciplinary Assessment, Ecology Law Quarterly, vol. 6, 1977,
pp. 38990.
3. The term preponderance is used in this article to avoid the confusion created by
simultaneous Gramscian and non-Gramscian uses of the term hegemony among international
relations scholars. Here, the emphasis is on possessing the far greatest share of relevant material
capability. In the case of whaling, no single country or group ever exerted hegemony in the
Gramscian sense of winning all others over to its preferred categories for defining and dealing
with the problem.
4. Yutaka Hirasawa, The Whaling Industry in Japans Economy, in John R.
Schmidhauser and George O. Totten III, eds., The Whaling Issue in USJapan Relations
(Boulder, Colo.: Westview Press, 1978), pp. 82114.
5. J. N. Tonnessen and A. O. Johnsen, The History of Modern Whaling (Berkeley: Univer-
sity of California Press, 1982), pp. 610 and 631.
6. See Imre Lakatos, Falsification and the Methodology of Scientific Research
Programmes, in Imre Lakatos and Alan Musgrave, eds., Criticism and the Growth of Knowl-
edge (Cambridge: Cambridge University Press, 1970), pp. 91196.
7. See, for example, Michael Polanyi, The Republic of Science, Minerva 1 (Autumn
1962), pp. 5473; and Diana Crane, Invisible Colleges (Chicago: University of Chicago Press,
8. Testimony of Lee Talbot, in U.S. Congress, House Committee on Merchant Marine
and Fisheries, Fish and Wildlife Miscellaneous, Part 4, Import Restrictions: Hearings Before
the Subcommittee on Fisheries and Wildlife Conservation and the Environment, 93d Congress,
2d sess., 1974, p. 21.
9. John A. Gulland, one of the worlds most senior fisheries management experts, notes
that in 1955, virtually no quantitative studies were being made on whale stocks, and the
scientists attending the Commission meetings had little if any quantitative expertise. See
J. A. Gulland, The Management Regime for Living Resources, in Christopher C. Joyner
and Sudhir K. Chopra, eds., The Antarctic Legal Regime (Dordrecht, Netherlands: Martinus
Nijhoff, 1988), p. 224.
10. This private direct action has involved gathering evidence against violators of IWC
rules, organizing consumer boycotts, surrounding whaling fleets, interposing groups in small
boats between harpooners and whales, and even attempting to sink whaling ships flying non-
IWC member state flags. Mandel argues that such methods are sometimes more effective than
Chapter 53 Page 107 September 11, 200819:53

Peterson Whaling 107

government enforcement. See Robert Mandel, Transnational Resource Conflict: The Politics
of Whaling, International Studies Quarterly 24 (March 1984), pp. 99127.
11. For a discussion of secretariat size, see Patricia Birnie, International Regulation of
Whaling (New York: Oceana, 1985), vol. 1, p. 176, and vol. 2, p. 758. Regarding orga-
nizational ideology, see Robert W. Cox, The Executive Head: An Essay on Leadership in
International Organizations, International Organization 23 (Spring 1969), pp. 20530; and
Robert W. Cox and Harold Jacobson, The Anatomy of Inuence: Decision Making in Interna-
tional Organizations (New Haven, Conn.: Yale University Press, 1973).
12. For a discussion of how UNEPs Regional Seas Programme was able to build a transna-
tional network despite having a staff of only six to eight people working on the issue of
Mediterranean pollution, see Peter M. Haas, Saving the Mediterranean (New York: Columbia
University Press, 1990).
13. This is the observation of Roger Payne, who has served as scientific adviser to Antigua
and Barbuda; cited in Bruce J. Stedman, The International Whaling Commission and Nego-
tiation for a Global Moratorium on Whaling, in Lawrence E. Susskind, Esther Siskind,
and J. William Breslin, eds., Nine Case Studies in International Environmental Negotiation
(Cambridge: MIT-Harvard Public Disputes Program, 1990), p. 166. While this means that some
commissioners end up taking cues from other delegations, it is easy to overestimate the effect
of their lack of substantive knowledge on IWC decisions.
14. Birnie, International Regulation of Whaling, vol. 1, p. 180.
15. Ibid., vol. 1, p. 561.
16. This thinking is reflected, for example, in Geoffrey Waughs Fisheries Management:
Theoretical Developments and Contemporary Applications (Boulder, Colo.: Westview Press,
17. A treaty rule may, through long and widespread usage, become binding on nonparties,
particularly if the nonparties have themselves followed the rule. For the most authoritative
discussion of when this might occur, see the International Court of Justices judgment in the
North Sea Continental Shelf Cases, I.C.J. Reports, 1969, p. 4.
18. In 1955, Norway proposed that the IWC appoint its own inspectors, but the protocol
authorizing such action was never ratified because the issue got caught up in disagreements
about national shares of Antarctic quotas. See Scarff, The International Management of
Whales, Dolphins, and Porpoises, p. 365.
19. See Tonnessen and Johnsen, The History of Modern Whaling, p. 621; and Birnie,
International Regulation of Whaling, vol. 1, p. 219.
20. See Michael MGonigle, The Economizing of Ecology: Why Big, Rare Whales Still
Die, Ecology Law Quarterly, vol. 9, 1980, pp. 13839, in which most of the blame is assigned
to the Dutch and Norwegians; Tonnessen and Johnsen, The History of Modern Whaling,
pp. 61027, in which most is assigned to the Japanese and Soviets; and Scarff, The Interna-
tional Management of Whales, Dolphins, and Porpoises, pp. 35666, in which most is assigned
to the Soviets.
21. For a good account of the interfirm agreements, see Tonnessen and Johnsen, The History
of Modern Whaling, chap. 32.
22. The unit of effort is defined as a days operation of a single catcher boat. Most
whaling operations in the 1950s and 1960s consisted of either a landstation or a factory ship
supplied with whale carcasses by several high-speed boats equipped with harpoon guns.
23. The public indifference can be gauged in events noted by environmentalist David Day in
The Whale War (London: Routledge & Kegan Paul, 1987), pp. 56. In the 1950s, the Icelandic
government requested and received U.S. air force assistance in tracking down and killing schools
of orca (killer) whales. The U.S. air force was still using the whales as targets for strafing
practice in the mid-1960s.
24. Tonnessen and Johnsen, The History of Modern Whaling, pp. 4023.
25. Ibid., p. 489.
26. For a discussion of stock size, see J. A. Gulland, The Antarctic Treaty System as a
Resource Management Mechanism, in Gillian D. Triggs, ed., The Antarctic Treaty Regime
(Cambridge: Cambridge University Press, 1987), p. 119.
Chapter 53 Page 108 September 11, 200819:53

108 International Environmental Issues and Debates

27. For an accessible discussion of MSY and later developments, see P.A. Larkin, Objec-
tives of Management, in Robert T. Lackey and Larry A. Nielson, eds., Fisheries Management
(New York: Wiley, 1980), pp. 24562.
28. M. W. Holdgate, Regulated Development and Conservation of Antarctic Resources,
in Triggs, The Antarctic Treaty Regime, p. 131.
29. See Steinar Andresen, Science and Politics in the International Management of
Whales, Marine Policy 13 (April 1989), pp. 1045.
30. See ibid., pp. 1034; and Birnie, International Regulation of Whaling, vol. 1, pp. 228
and 248.
31. Tonnessen and Johnsen, The History of Modern Whaling, pp. 585608.
32. This observation is fully consistent with Ellicksons claim that whalers spontaneously
develop rules about who has the right to capture a particular whale and that these rules not
only maximize the likelihood that a chased whale will be taken but also minimize intership
arguments over priority or shares. As Ellickson notes, such rules are likely to encourage aggre-
gate overexploitation. Whalers, like other users of common pool resources, find it extremely
difficult to overcome the incentives to overexploit. See Ralph Ellickson, A Hypothesis of
Wealth-Maximizing Norms: Evidence from the Whaling Industry, Journal of Law, Economics,
and Organization 5 (Spring 1989), pp. 8397.
33. Each factory ship or landstation was required to report its weekly catch to the Bureau
of International Whaling Statistics in Norway. The bureau closed the season when the quota
was reached.
34. See Gulland, The Management Regime for Living Resources, p. 225.
35. Tonnessen and Johnsen, The History of Modern Whaling, pp. 56970 and 57576.
36. For data on the decline in whale oil prices, see ibid., p. 753, Table 69.
37. See J. L. McHugh, The Role and History of the International Whaling Commission,
in William E. Scheville, ed., The Whale Problem (Cambridge, Mass.: Harvard University Press,
1974), p. 306.
38. The Japanese purchases increased tension between the Soviet and Japanese groups
because it could have meant limiting the Soviets to the 20 percent quota that they had managed
to gain through difficult negotiations with others in 1962. Again, the Soviets were able to secure
concessions. See Michael Taylor and Hugh Ward, Chickens, Whales, and Lumpy Goods,
Political Studies 30 (September 1982), p. 359.
39. For a discussion of the whale meat and oil markets, see G. Small, The Blue Whale
(New York: Columbia University Press, 1971), pp. 3942. Small notes that during the 195063
period, a blue whale used only to produce oil earned its taker an average of $3,675, while a
blue whale used to produce both meat and oil earned an average of $11,250.
40. See Andresen, Science and Politics in the International Management of Whales,
p. 104.
41. See Holdgate, Regulated Development and Conservation of Antarctic Resources,
p. 131; and Gulland, The Antarctic Treaty System as a Resource Management Mechanism,
p. 119. As Gulland notes in The Management Regime for Living Resources, pp. 22425, the
proposed approach did not represent the simple imposition of models developed for another
species. Some of the early work had emerged from studies of whales; see, for example, J. Hjort,
G. Jahn, and P. Ottestad, The Optimum Catch, Hvalradets Skrifter, vol. 7, 1933, pp. 92127.
42. See Andresen, Science and Politics in the International Management of Whales,
p. 107.
43. Scarff, The International Management of Whales, Dolphins, and Porpoises, p. 366.
44. This classification is discussed in ibid., p. 367; and Birnie, International Regulation of
Whaling, vol. 1, pp. 453 and 461.
45. See Scarff, The International Management of Whales, Dolphins, and Porpoises,
pp. 41821.
46. Ibid., p. 634. The decision to set a zero quota for South Pacific Brydes whales in
1976, for example, was highly controversial. Threatened with U.S. sanctions if it filed a formal
objection to this decision, the Japanese government sought to keep its whalers active by issuing
Chapter 53 Page 109 September 11, 200819:53

Peterson Whaling 109

permits for taking as part of scientific research, an early example of a practice that later
became more prominent.
47. Resolution 33, Report on the UN Conference on the Human Environment, UN Doc.
A/CONF.48/14, June 1972, p. 23.
48. See Tonnessen and Johnsen, The History of Modern Whaling, p. 554; and Birnie, Inter-
national Regulation of Whaling, vol. 2, p. 618.
49. Articles 11620 and 13337, Informal Composite Negotiating Text, UNCLOS III, UN
Doc. A/CONF.62/WP.10, 15 July 1977.
50. Andresen, Science and Politics in the International Management of Whales, p. 109.
51. See, for example, Vernon L. Smith, On Models of Commercial Fishing, Journal of
Political Economy 77 (MarchApril 1969), pp. 18198; S. N. S. Cheung, Contractual Arrange-
ments and Resource Allocation in Marine Fisheries, in A. D. Scott, ed., Economics of Fisheries
Management: A Symposium (Vancouver: Institute of Animal Resource Ecology, University of
British Columbia, 1970), pp. 97108; and Colin W. C. Clark, The Economics of Overex-
ploitation, Science 181 (August 1973), pp. 63034.
52. See Scarff, The International Management of Whales, Dolphins, and Porpoises,
pp. 58586; and MGonigle, The Economizing of Ecology, p. 123.
53. The recommendations were made during the Airlie House Workshops, February and
April 1975, and are discussed by Scarff in The International Management of Whales, Dolphins,
and Porpoises, pp. 39597.
54. The importance of computer capacity is noted by Taivo Laevastu and Herbert A. Larkins
in Marine Fisheries Ecosystem (Farnham, U.K.: Fishing News Books, 1981), p. xi. For observa-
tions regarding the older and newer models, see Peter A. Larkin, An Epitaph for the Concept of
Maximum Sustainable Yield, Transactions of the American Fisheries Society, vol. 106, 1977,
p. 8; D. H. Cushing, Fisheries Biology: A Study in Population Dynamics, 2d ed. (Madison:
University of Wisconsin Press, 1981); Waugh, Fisheries Management; and Colin W. C. Clark,
Bioeconomic Modelling in Fisheries Management (New York: Wiley, 1985).
55. Tonnessen and Johnsen, The History of Modern Whaling, p. 619.
56. The belief that the eight species of great whales were endangered was reflected, for
example, in the 1971 decision of the U.S. government to include them on the list of animals
whose harming or taking was banned under the Endangered Species Act (Public Law 93205;
United States Code, Title 16, Sections 153143) and in the 1976 decision of the parties to the
Convention on International Trade in Endangered Species to list them as endangered.
57. See MGonigle, The Economizing of Ecology; and Peter Scott, The End of
Whaling, The Siren, no. 26, December 1984, pp. 24.
58. Compare, for example, the remarks of Craig Van Note and those of J. B. Kelley, in
U.S. Congress, House Committee on Commerce, Science, and Transportation, U.S. Whaling
Policies/International Whaling Commission: Hearings Before the Committee on Commerce,
Science, and Transportation, 97th Congress, 1st sess., 1981, pp. 1923 and 3435.
59. Remarks of Patricia Forkan, in U.S. Congress, House Committee on Foreign Affairs,
Review of the 32d International Whaling Commission Meeting: Hearings Before the Subcom-
mittee on International Organizations, 96th Congress, 2d sess., 1980, p. 67.
60. Hideo Obara, Whaling: The Conservation Side of Japanese Opinion, The Siren,
no. 35, December 1987, pp. 1317.
61. See Tonnessen and Johnsen, The History of Modern Whaling, p. 587; and remarks
of J. B. Kelly, in U.S. Congress, U.S. Whaling Policies/International Whaling Commission:
Hearings, p. 35.
62. For a good introductory discussion of Japanese attitudes toward whaling, see Kathy
Glass and Kirsten Englund, Why the Japanese Are So Stubborn About Whaling, Oceanus 32
(Spring 1989), pp. 4553.
63. See MGonigle, The Economizing of Ecology, p. 185. MGonigle did note, however,
that there was considerable awareness of and political agitation about domestic environmental
problems in Japan.
64. Regarding Western ethnocentrism, see Obara, Whaling. Regarding Western actions,
a Tokyo newspaper warned that if Japan is to take a strong step [opposing IWC decisions],
Chapter 53 Page 110 September 11, 200819:53

110 International Environmental Issues and Debates

it will be dealt all-out attacks by various countries, and restrictions on its fish catch within
the 200-nautical-mile zones will be strengthened. See Yomiuri, 30 June 1977, p. 9; quoted
in MGonigle, The Economizing of Ecology, p. 186. At the 1989 IWC meeting, Japans
commissioner complained that the meat-eating culture was using IWC to dominate and even
destroy the fish-eating culture. See the opening statement of Commissioner K. Shima to the
41st IWC meeting, 12 June 1989; quoted in Stedman, The International Whaling Commission
and Negotiation for a Global Moratorium on Whaling, pp. 15758.
65. Japanese promises and threats to other states became a regular theme of environmen-
talist writings and speeches, some of which appeared to exaggerate. Certain instances, however,
have been documented. See, for example, the U.S. delegations account of the Japanese chal-
lenge to the credentials of St. Lucias delegates, in U.S. Congress, House Committee on Foreign
Affairs, Review of the 34th International Whaling Commission Meeting: Hearings Before the
Subcommittee on International Organizations, 97th Congress, 2d sess., 1982, pp. 1112 and 40.
66. Mandel, Transnational Resource Conflict, p. 106.
67. Public Law 92219; United States Code, Title 22, Section 1978, attached to the
Fishermens Protective Act.
68. Public Law 9661; United States Code, Title 22, Section 1821[e], attached to the
Fisheries Conservation and Management Act.
69. Both Remington Kellogg, who served from 1949 to 1968, and J. L. McHugh, who
served from 1969 to 1970, were trained cetologists.
70. See U.S. Congress, House Committee on Foreign Affairs, International Moratorium
of Ten Years on the Killing of All Species of Whales: Hearings Before the Subcommittee on
International Organizations and Movements, 92d Congress, 1st sess., 1971, pp. 6365.
71. Testimony of NOAA Administrator John V. Byrne, in U.S. Congress, House Committee
on Merchant Marine and Fisheries, U.S. Whaling Policy Oversight: Hearings Before the
Subcommittee on Fish and Wildlife Conservation and the Environment, 98th Congress, 1st
sess., 1983, pp. 2526.
72. See Birnie, International Regulation of Whaling, vol. 1, pp. 195 and 543.
73. See U.S. Congress, U.S. Whaling Policy Oversight: Hearings, p. 11.
74. Mandel, Transnational Resource Conflict, p. 107.
75. For a discussion of the 1977 case, Adams v. Vance, see MGonigle, The Economizing
of Ecology, pp. 15657.
76. Birnie, International Regulation of Whaling vol. 2, p. 620.
77. See IWC, Annual Report, 1965 and Annual Report, 1966 (Cambridge: IWC, 1965 and
1966), Appendix I.
78. Keith D. Suter, The International Politics of Saving the Whale, Australian Outlook
35 (December 1981), pp. 29394.
79. Day, The Whale War, p. 96.
80. Tonnessen and Johnsen, The History of Modern Whaling, p. 648.
81. See Andresen, Science and Politics in the International Management of Whales,
p. 103; and Scarff, The International Management of Whales, Dolphins, and Porpoises,
p. 355.
82. Andresen, Science and Politics in the International Management of Whales, p. 113.
83. For a discussion of the U.S. rationale, see U.S. Congress, House Committee on Merchant
Marine and Fisheries, Fisheries Management and Conservation Act: Hearings Before the
Committee on Merchant Marine and Fisheries, 96th Congress, 1st sess., 1979, pp. 31273.
84. Day, The Whale War, p. 97.
85. See ibid., p. 96; and MGonigle, The Economizing of Ecology, p. 190. Day stops just
short of saying that the environmentalists also offered financial support; however, MGonigle
says that the Seychelles government received some financial assistance from the Threshold
86. For an environmentalist view of the pressures on the Panamanian government, see Day,
The Whale War, p. 97.
87. Statement by Prime Minister Pierre Trudeau, in Canadian House of Commons, Debates,
32d Parliament, 1st sess., vol. 10, pp. 1102527.
Chapter 53 Page 111 September 11, 200819:53

Peterson Whaling 111

88. In The Economizing of Ecology, p. 188, MGonigle regards Canadas concern about
EEZ jurisdiction as the primary reason for its withdrawal from the IWC.
89. For a discussion of U.S. sanctions, see Erin K. Flory, Construing the Pelly and
Packwood-Magnuson Amendments, Washington Law Review 61 (April 1986), p. 655.
90. For brief notices of the programs progress, see the following reports of G. P. Donovan:
Thirty-Eighth Annual Meeting of the International Whaling Commission, Polar Record 23
(January 1987), pp. 43741; Thirty-Ninth Annual Meeting of the International Whaling
Commission, Polar Record 24 (January 1988), p. 66; Fortieth Annual Meeting of the Inter-
national Whaling Commission, Polar Record 26 (October 1989), p. 250; and Forty-First
Annual Meeting of the International Whaling Commission, Polar Record 26 (October 1990),
pp. 23334. For initial results of the program, see G. P. Donovan, ed., The Comprehensive
Assessment of Whale Stocks: The Early Years, special issue no. 11, Reports of the International
Whaling Commission (Cambridge: IWC, 1989).
91. Donovan, Thirty-Ninth Annual Meeting of the International Whaling Commission,
p. 67.
92. See, for example, Sidney Holt, The Whaling Game, The Siren, no. 35, December
1987, pp. 36.
93. G. P. Donovan, Forty-Second Annual Meeting of the International Whaling Commis-
sion, Polar Record 27 (January 1991), pp. 5355.
94. The Packwood-Magnuson amendment was, however, invoked to reduce Soviet quotas
in 1985 and 1986. See David D. Caron, International Sanctions, Ocean Management, and the
Law of the Sea, Ecology Law Quarterly, vol. 16, 1989, pp. 31920.
95. Though confident that the Pelly amendment would not be invoked during the 198283
controversies, the Japanese were ready to counter by charging the United States with unfair
trading practices under the General Agreement on Tariffs and Trade (GATT) as a preliminary to
retaliation. See Robert Friedheim and Tsuneo Akaha, Antarctic Resources and International
Law: Japan, the United States, and the Future of Antarctica, Ecology Law Quarterly, vol. 16,
1989, p. 141.
96. In 1986, Japans fish product exports to the United States were worth about
$310 million a year, while its fish product imports from the United States were worth about
$620 million. See Flory, Construing the Pelly and Packwood-Magnuson Amendments, p. 651.
97. The lobbyists employed by Norway included Terry Leitzel, a former director of the
National Marine Fisheries Service. See remarks of Craig Van Note, in U.S. Congress, House
Committee on Foreign Relations, U.S. Policy with Respect to the International Protection of
Whales: Hearings Before the Committee on Foreign Relations, 99th Congress, 1st sess., 1985,
p. 35.
98. See State Department internal memoranda, published in U.S. Congress, House
Committee on Foreign Affairs, U.S. Policy Toward the International Whaling Commission
and Other Marine Mammal Issues: Hearings Before the Subcommittee on Human Rights and
International Organizations, 101st Congress, 1st sess., 1989, pp. 2627 and 3233.
99. Although the resolution was treated favorably in the House Committee on Foreign
Affairs, it was killed in the House Committee on Merchant Marine and Fisheries, where
Chairman John Breaux argued that its main effect would be to harm U.S. Fishers. See U.S.
Congress, U.S. Policy with Respect to the International Protection of Whales: Hearings,
Appendix 1, letter from John Breaux.
100. See Donovan, Thirty-Ninth Annual Meeting of the International Whaling Commis-
sion, p. 65; and Caron, International Sanctions, Ocean Management, and the Law of the
Sea, p. 320.
101. The Supreme Court overturned earlier district and appellate court decisions requiring
application of the Packwood-Magnuson provision to Japan in July 1986 in American Cetacean
Society v. Baldridge. However, the threat of another suit, Greenpeace USA et al. v. Robert
Mosbacher et al., probably had some influence on the decision to apply the Packwood-
Magnuson provision to Japan in 1988 after the Japanese ignored Scientific Committee criticism
and issued scientific permits for whaling. The threat of the suit also helped keep alive the issue
of trade sanctions against Norway and Iceland for similar, though less extensive, activity.
Chapter 53 Page 112 September 11, 200819:53

112 International Environmental Issues and Debates

102. Sidney Holt, letter to the editor, The Pilot (UNEP-IUCN newsletter for the Marine
Mammal Action Plan), no. 3, April 1989, p. 13.
103. Before 1977, the Scientific Committee lacked even the right to comment on whether
permits should be issued; that right was secured as a compromise between whaler preferences
and the U.S. proposal that the Scientific Committee be required to approve all permits. See
Scarff, The International Management of Whales, Dolphins, and Porpoises, p. 634.
104. Donovan, Thirty-Ninth Annual Meeting of the International Whaling Commission,
p. 66.
105. See Donovan, Forty-Second Annual Meeting of the International Whaling Commis-
sion, pp. 5354. Regarding the Icelandic reaction, see Whaling Could Resume Next Year,
News from Iceland, August 1990, p. 1.
106. See Margaret Klinowska, How Brainy Are Cetaceans? Oceanus 32 (Spring 1989),
pp. 1920.
107. Haas and Ikenberry suggest that capturing the preponderant power is particularly
important when intergovernmental organizations are weak or nonexistent. See the following
articles in this issue of IO: Peter M. Haas, Banning Chlorofluorocarbons: Epistemic Commu-
nity Efforts to Protect Stratospheric Ozone; and G. John Ikenberry, A World Economy
Restored: Expert Consensus and the Anglo-American Postwar Settlement.
108. Other epistemic communities, such as those described by Hopkins and Haas, have
suffered from internal disagreements but maintained their influence because they were less
sharply challenged by other groups. See the following articles in this issue of IO: Raymond
F. Hopkins, Reform in the International Food Aid Regime: The Role of Consensual Knowl-
edge; and Haas, Banning Chlorofluorocarbons.
Chapter 54 Page 113 September 11, 200819:53

Regime Design Matters: Intentional Oil Pollution
and Treaty Compliance
Ronald B. Mitchell

Too many people assume, generally without having given any serious
thought to its character or its history, that international law is and always
has been a sham. Others seem to think that it is a force with inherent
strength of its own.    Whether the cynic or sciolist is the less helpful
is hard to say, but both of them make the same mistake. They both
assume that international law is a subject on which anyone can form his
opinions intuitively, without taking the trouble, as one has to do with
other subjects, to inquire into the relevant facts.
J. L. Brierly

egime design matters.1 International treaties and regimes have value

R if and only if they cause people to do things they would not other-
wise do. Governments spend considerable resources and effort
drafting and refining treaty language with the (at least nominal) aim of
making treaty compliance and effectiveness more likely. This article demon-
strates that whether a treaty elicits compliance or other desired behavioral
changes depends upon identifiable characteristics of the regimes compliance
systems.2 As negotiators incorporate certain rules into a regime and exclude
others, they are making choices that have crucial implications for whether
or not actors will comply.
For decades, nations have negotiated treaties with simultaneous hope
that those treaties would produce better collective outcomes and skepticism
about the ability to influence the way governments or individuals act. Both
lawyers and political scientists have theorized about how international legal
regimes can influence behavior and why they often do not.3 Interest in issues
of compliance and verification has a long history in the field of nuclear arms
control.4 More recently, this interest in empirically evaluating how inter-
national institutions, regimes, and treaties induce compliance and influence
behavior has broadened to include other security areas as well as interna-
tional trade and finance.5 Concern over the fate of the earths environment
recently has prompted a further extension into questions of whether and how

Source: International Organization, 48(3) (1994): 42558.

Chapter 54 Page 114 September 11, 200819:53

114 International Environmental Issues and Debates

environmental treaties can be made more effective at eliciting compliance

and achieving their goals.6
Researchers in all these issue-areas face two critical questions. First, given
that power and interests play important roles in determining behavior at the
international level, is any of the compliance we observe with international
treaties the result of the treatys influence? Second, if treaties and regimes can
alter behavior, what strategies can those who negotiate and design regimes
use to elicit the greatest possible compliance? This article addresses both
these questions by empirically evaluating the international regime controlling
intentional oil pollution. Numerous efforts to increase the regimes initially
low levels of compliance provide data for comparing the different strategies
for eliciting compliance within a common context that holds many impor-
tant explanatory variables constant. The goal of the treaties underlying this
regime has been to reduce intentional discharges of waste oil by tankers after
they deliver their cargoes. Since the late 1970s, these treaties have established
two quite different compliance systems, or subregimes, to accomplish this
goal. One has prohibited tanker operators from discharging oil in excess of
specified limits. The other has required tanker owners to install expensive
pollution-reduction equipment by specified dates. Treaty parties viewed both
subregimes as equally legitimate and equally binding.7 The two subregimes
regulated similar behavior by the same nations and tankers over the same time
period. The absence of differences in power and interests would suggest that
compliance levels with the two subregimes would be quite similar.8 According
to collective action theory, these cases are among the least likely to provide
support for the hypothesis that regime design matters: subregime provisions
required the powerful and concentrated oil industry to incur large pollution
control costs to provide diffuse benefits to the public at large.9 Indeed, the
lower cost of complying with discharge limits would suggest that compliance
would be higher with those limits than with equipment requirements.
Not surprisingly, violations of the limits on discharges have occurred
frequently, attesting to the ongoing incentives to violate the agreement and
confirming the characterization of oil pollution as a difficult collaboration
problem.10 A puzzle arises, however, from the fact that contrary to expec-
tation compliance has been all but universal with requirements to install
expensive equipment that provided no economic benefits. The following
analysis clearly demonstrates that the significant variance across subregimes
can only be explained by specific differences in subregime design. Comparing
the two compliance systems shows that the equipment subregime succeeded
by ensuring that actors with incentives to comply with, monitor, and enforce
the treaty were provided with the practical ability and legal authority to
conduct those key implementation tasks. Specifically, the regime elicited
compliance when it developed integrated compliance systems that succeeded
in increasing transparency, providing for potent and credible sanctions,
reducing implementation costs to governments by building on existing infras-
tructures, and preventing violations rather than merely deterring them.
Chapter 54 Page 115 September 11, 200819:53

Mitchell Regime Design Matters 115

Compliance Theory and Definitions

Explaining the puzzle of greater compliance with a more expensive and

economically inefficient international regulation demands an understanding
of existing theories about the sources of compliance in international affairs.
Realists have inferred a general inability of international regimes to influence
behavior from the fact that the international system is characterized by
anarchy and an inability to organize centralized enforcement. In what has
been the dominant theoretical view, considerations of power rather than
of law determine compliance.11 To explain variance in treaty compliance,
look for variance in the power of those with incentives to violate it or in the
interests of those with the power to violate it. Treaties are epiphenomenal:
they reflect power and interests but do not shape behavior.
This view does not imply that noncompliance is rare in international
affairs. Although nations will violate rules whenever they have both the
incentives and ability to do so, as Hans Morgenthau notes, the great majo-
rity of the rules of international law are generally observed by all nations.12
For the realist, behavior frequently conforms to treaty rules because both
the behavior and the rules reflect the interests of powerful states. More
specifically, compliance is an artifact of one of three situations: (1) a hege-
monic state forces or induces other states to comply; (2) the treaty rules
merely codify the parties existing behavior or expected future behavior; or
(3) the treaty resolves a coordination game in which no party has any incen-
tive to violate the rules once a stable equilibrium has been achieved.13
Treaty rules correlate with but do not cause compliance. Therefore,
efforts to improve treaty rules to increase compliance reflect either the
changed interests of powerful states or are misguided exercises in futility.
The strength of this view has led to considerable attention being paid to
whether rules influence behavior and far less being paid to design features
that explain why one rule influences behavior and another does not.
In contrast, international lawyers and institutionalists contend that the
anarchic international order need not lead inexorably to nations violating
agreements whenever doing so suits them. Other forces such as trans-
parency, reciprocity, accountability, and regime-mindedness allow regimes
to impose significant constraints on international behavior under the right
conditions.14 Implicit in the institutionalist view is the assumption that power
and interests alone cannot explain behavior: a given constellation of power
and interests leaves room for nations to choose among treaty rules that will
elicit significantly different levels of compliance. High compliance levels can
be achieved even in difficult collaboration problems in which incentives to
violate are large and ongoing. Treaties can become more effective over time,
and regimes may even learn.15 Agreeing with Morgenthau that compliance
will be quite common, institutionalists do not exclude the possibility that
the regime, rather than mere considerations of power, causes some of that
Chapter 54 Page 116 September 11, 200819:53

116 International Environmental Issues and Debates

In essence, this debate revolves around whether in a realm of behavior

covered by an international agreement, that behavior is ever any different
than it would have been without the agreement. If we define treaty-induced
compliance as behavior that conforms to a treatys rules because of the
treatys compliance system, institutionalists view treaty-induced compliance
as possible. In contrast, realists see all compliance as coincidental compli-
ance, that is, behavior that would have occurred even without the treaty
The debate between these theories highlights the demands placed on
research that seeks to identify those design characteristics of a regime, if any,
that are responsible for observed levels of compliance. I define compliance,
the dependent variable, as an actors behavior that conforms with an explicit
treaty provision. Speaking of compliance with treaty provisions rather than
with a treaty captures the fact that parties may well comply with some treaty
provisions while violating others. A study of treaty compliance would
aggregate violation of one provision with compliance with another, losing
valuable empirical information.17 Restricting study to the explicit rules in
a treaty-based regime allows the analyst to distinguish compliance from
noncompliance in clear and replicable ways. Obviously, a focus on explicit
rules ignores other potential mechanisms of regime influence, such as norms,
principles, and processes of knowledge creation.18 However, this restrictive
definition has the virtue of bringing the debate to a level at which research
on actual treaties and actual compliance can contribute to the intellectual
and policy debates.
This article evaluates the features of a regime that may determine compli-
ance by differentiating among three parts of any compliance system: a
primary rule system, a compliance information system, and a noncompli-
ance response system. The primary rule system consists of the actors, rules,
and processes related to the behavior that is the substantive target of the
regime. In the choice of who gets regulated and how, the primary rule system
determines the pressures and incentives for compliance and violation. The
compliance information system consists of the actors, rules, and processes
that collect, analyze, and disseminate information on instances of viola-
tions and compliance. Self-reporting, independent monitoring, data analysis,
and publishing comprise the compliance information system that determines
the amount, quality, and uses made of data on compliance and enforce-
ment. The noncompliance response system consists of the actors, rules, and
processes governing the formal and informal responses the inducements
and sanctions employed to induce those in noncompliance to comply.
The noncompliance response system determines the type, likelihood, magni-
tude, and appropriateness of responses to noncompliance. These categories
provide the framework used in the remainder of this article to evaluate the
oil pollution regimes sources of success and failure in its attempt to elicit
Chapter 54 Page 117 September 11, 200819:53

Mitchell Regime Design Matters 117

Two Subregimes for International Oil Pollution Control

For most people, oil pollution conjures up images of tanker accidents such
as that of the Exxon Valdez.19 While oil from such accidents poses a concen-
trated but localized hazard to the marine environment, the waste oil tradi-
tionally generated during normal oil transport has posed a more diffuse
but ubiquitous threat. After a tanker delivers its cargo, a small fraction of
oil remains onboard, adhering to cargo tank walls. Ballasting and tank-
cleaning procedures mixed this oil averaging about 300 tons per voyage
with seawater, creating slops. These in turn were most easily and cheaply
disposed of by discharging them overboard while at sea.20 By the 1970s, the
intentional discharges made on thousands of tanker voyages were putting
an estimated million tons of oil into the oceans annually.21 While scientific
uncertainty remains regarding the extent of damage to marine life caused
by such chronic but low-concentration discharges, their impact and that of
accidents on seabirds and resort beaches have produced regular international
efforts at regulation.22
Intentional oil discharges were one of the first pollutants to become
the subject of an international regulatory regime.23 In the International
Convention for the Prevention of Pollution of the Seas by Oil (OILPOL) of
1954, nations addressed the coastal oil pollution problem by limiting the oil
content of discharges made near shore.24 In what has been a regime largely
focused on regulation,25 numerous revisions were negotiated within diplo-
matic conferences sponsored by the Intergovernmental Maritime Consulta-
tive Organization (IMCO) or within its committees and those of its successor,
the International Maritime Organization (IMO). By the late 1970s, the
regimes major provisions, now contained in the International Convention
for the Prevention of Pollution from Ships (MARPOL), consisted of restric-
tions on both tanker operations and tanker equipment that relied on quite
different compliance systems.26 Although rule-making has remained consis-
tently international, governments and nonstate actors have played crucial
roles in the implementation and enforcement of the regime: tanker owners
and operators have been the targets of the regulations while maritime author-
ities, classification societies, insurers, and shipbuilders have monitored and
enforced the regulations.

The Discharge Subregime

The discharge subregime of the last fifteen years evolved from the initial
regulations of 1954. That agreement constituted a compromise between
the United Kingdom which wielded strong power in oil markets but had
strong environmental nongovernmental organizations pushing it to reduce
coastal pollution and Germany, the Netherlands, the United States, and
other major states that viewed any regulation as either environmentally
unnecessary or as harmful to their own shipping interests. Although the
Chapter 54 Page 118 September 11, 200819:53

118 International Environmental Issues and Debates

United Kingdom had sought to restrict tanker discharges throughout the

ocean, the final agreement limited the oil content of discharges made within
fifty miles of any coastline to 100 parts oil per million parts water (100 ppm).
In 1962, the British pushed through an amendment applying this 100 ppm
standard to discharges made by new tankers regardless of their distance
from shore.
The principle underlying the 1962 amendment that crude oil could float
far enough that discharge zones would not effectively protect coastlines had
gained sufficient support by 1969 that nations agreed to limit discharges by
all tankers throughout the ocean. The pressure to amend the 1954/62 agree-
ment came from two different sources. On one side, the thirty-five million
gallons of oil spilled by the grounding of the Torrey Canyon off Britain and
France on 18 March 1967 and growing environmentalism, especially in the
United States, supported a push for stronger regulations.27 The previously
resistant United States replaced the United Kingdom as the leading activist
state and especially sought to ensure that amendments would address the
growing evidence of enforcement problems with existing regulations.
On the other side, oil companies rightly interpreted the 1962 amendments
as a wake-up call that discharge standards would soon be replaced by expen-
sive equipment requirements. In response, Shell Marine International devel-
oped and promoted an operational means by which tankers could reduce oil
discharges without any new equipment.28 The load-on-top procedure (LOT)
involved consolidating ballast and cleaning slops in a single tank, letting
gravity separate out the water so it could be decanted from beneath the oil,
and loading the next cargo on top of the remaining slops. The beauty of LOT
was that it ensured that less cargo was wasted, thereby advancing both the
environmental goal of reducing intentional oil pollution and the economic
goal of reducing the amount of valuable oil discharged overboard. LOT even
improved on the regimes existing standards, since its use reduced rather than
merely redistributed intentional discharges. The problem was that normal
operation of LOT produced discharges that exceeded the 100 ppm standard.
If this criterion had remained in effect, tankers would have had to install
expensive new equipment to comply with OILPOL, defeating LOTs major
economic virtue. With the support of France, the Netherlands, Norway, and
the now less-activist United Kingdom, oil and shipping companies therefore
also sought to amend the treaty. Oil companies considered LOT so effective
that they wanted diplomats to scrap the 1954/62 zonal approach altogether.
The pressures for greater environmental protection, however, led them to
support the more limited objective of redefining the limits on discharges
from the 100 ppm content criterion to one that could be monitored using
existing on-board equipment.29
In a unanimously accepted compromise in 1969, more stringent and
enforceable regulations were framed in terms that averted equipment require-
ments. Within the fifty-mile near-shore zones, discharges could now only
involve clean ballast that left no visible trace; outside the fifty-mile zones,
discharges could not exceed 60 liters of oil per mile (60 1/m). Proponents
Chapter 54 Page 119 September 11, 200819:53

Mitchell Regime Design Matters 119

argued that the clean ballast provision would improve enforcement by

transforming any sighting of a discharge into evidence of a violation.30 The
more crucial change involved a new limit that total discharges not exceed
one fifteen-thousandth of a tankers capacity.31 Although compliance with
this standard required a tanker to reduce its average discharges by almost
98 percent, Shells J. H. Kirby claimed that any responsibly run ship, no
matter how big, could operate within these standards if it used LOT.32 The
low total discharge limit also allowed port authorities to assume that any
tanker with completely clean tanks had blatantly violated the agreement.33
These standards took effect in 1978 and remain in force today through their
incorporation into the 1973 MARPOL agreement.

The Equipment Subregime

By the early 1970s, public concern was pushing environmental issues onto
the international political scene with increasing frequency. The United
Nations Conference on the Human Environment and negotiation of the
London Dumping Convention in 1972 set the stage for a major overhaul
of the OILPOL agreement. IMCO hosted a major conference in 1973 to
negotiate the MARPOL treaty. Its goal was the replacement of OILPOLs
rules with rules that would cover all major types of vessel-source marine
The U.S. government had become increasingly concerned that the ease
with which tanker crews could violate discharge standards and the massive
resources and diligence needed to detect violations were preventing effec-
tive mitigation of the growing oil pollution problem.34 By 1972, Congress
had adopted legislation that threatened to require all American tankers as
well as all tankers entering U.S. ports to install expensive pollution-reducing
equipment. The legislation included a proposal to require all large tankers to
install double hulls to address accidental spills and segregated ballast tanks
(SBT) to address intentional discharges. The SBT system involved arranging
ballast tanks and associated piping such that ballast water could not come
into contact with oil being carried as cargo. The system was expensive
both in terms of capital and the reduction to cargo-carrying capacity. The
United States sought international agreement to require SBT but threatened
to require it unilaterally if necessary. Discharge requirements clearly were
cheaper, more economically efficient, and in theory    a good idea.35
However, environmental pressures and growing evidence that LOT was
neither as widespread nor as effective as had been hoped led the United
States and the United Kingdom to support rules that offered easier and more
effective enforcement.
The largely U.S.-based oil companies initially opposed SBT requirements
but eventually supported them as preferable to threatened U.S. unilateral
rules. Many shipping states also reluctantly supported SBT requirements.
They believed such requirements would avert an even more costly double
bottom requirement. It was also fiscally acceptable: the combination of
Chapter 54 Page 120 September 11, 200819:53

120 International Environmental Issues and Debates

a recent building boom and the proposed language of the requirements

meant that tanker owners would only have to incur the additional costs of
SBT many years in the future and then only for large tankers. However,
governments representing shipbuilding interests (France and Japan) and
those representing independent tanker owners (Denmark, Germany, Greece,
Norway, and Sweden) opposed the requirement.36 By a vote of thirty to
seven, the conference adopted a requirement for tankers over 70,000 tons
built in 1980 and later to install SBT.
By 1977, a spate of accidents in the United States and continuing enforce-
ment concerns led President Jimmy Carter to propose that SBT require-
ments be applied to all tankers, not just large new tankers.37 Given (1) that
the United States was again explicitly threatening unilateral action and
(2) that the 1973 MARPOL agreement still had been ratified by only three
states, IMCO called a second major conference in 1978.38 State positions
reflected the fact that retrofitting existing tankers with SBT would reduce
each tankers (and the fleets) cargo capacity by some 15 percent.39 Greece,
Norway, and Sweden saw this as a means to put scores of their laid up inde-
pendent tankers back to work. However, most states saw SBT retrofitting
as extremely expensive.40 Just as the 1962 amendments had prompted LOT
development, the 1973 MARPOL agreement prompted oil companies to
perfect a technique known as crude oil washing (COW), which entailed
spraying down cargo tanks with the cargo itself rather than with seawater.
Operating COW equipment during cargo delivery transformed oil that other-
wise would have been discharged as slops into usable delivered cargo, simul-
taneously reducing oil pollution and increasing cargo owner revenues. The
industry proposal for COW as an alternative to SBT produced a compro-
mise in which tankers built after 1982 had to install both SBT and COW,
while existing tankers had to be retrofitted with either SBT or COW by
1985. The 1978 Protocol Relating to the International Convention for the
Prevention of Pollution from ships was made an integral part of the 1973
MARPOL agreement. While MARPOL and its protocol, known collectively
as MARPOL 73/78, did not enter into force until 1983, their standards regu-
lated all new construction after 1979.

Observed Compliance Levels

Available evidence demonstrates a wide divergence in levels of compli-

ance under these two subregimes. During the same time period in which
almost every tanker owner was retrofitting existing tankers and buying new
tankers to conform with MARPOLs requirements for SBT and COW, large
numbers of tanker operators continued to discharge oil well in excess of
legal limits. The variance between the observed compliance rates with the
two subregimes is quite marked.
Violations of the clean ballast, 60 1/m, and total discharge standards in
place since 1978 have been common. Oil company surveys from the 1970s
Chapter 54 Page 121 September 11, 200819:53

Mitchell Regime Design Matters 121

Figure 1: Average tanker discharges, 197277

Fraction of cargo (fifteen-thousandths)

40 Independent tankers

5 Oil company tankers
Legal limit as of October 1978
1972 1973 1974 1975 1976 1977

Source: U.S. Congress, House Committee on Government Operations, Oil Tanker Pollution:
Hearings Before the Subcommittee on Government Activities and Transportation, 18 and
19 July 1978, 95th Congress, 2d sess., p. 322.

show that neither oil company nor independent tankers reduced average
discharge levels to the one fifteen-thousandth limit in any year between 1972
and 1977 (see Figure 1). Although oil company tankers dramatically reduced
average discharges in the early 1970s, discharges remained at three times the
legal limit. The two-thirds of the fleet operated by independent oil trans-
porters did far worse, with discharges that were thirty times the legal limit
and that were not much below levels that a tanker practicing no pollution
control would have produced.41 The trends in these discharges suggest that
few tankers complied with the limit after it took legal effect in 1978.42
Other evidence confirms the frequency of discharge violations. A 1981
National Academy of Sciences estimate of oil pollution relied on an assump-
tion that 50 percent of the worlds tanker fleet was violating the total
discharge limit.43 A 1989 revision of that study assumed 1520 percent of
tankers were still violating this limit, although it provided no evidence to
support the dramatic improvement.44 Representatives of independent trans-
porters admit that tankers often violate discharge limits to comply with their
contracts: some charter arrangements require ships to arrive with clean tanks
but many ports lack facilities to receive the slops they generate by cleaning.45
Studies of detected oil slicks and dead seabirds as well as violation reports
provided to IMO confirm that many tankers continue to discharge their
slops at sea.46
The variety of sources pointing to violation of the discharge standards
contrasts sharply with the uniformity of evidence that compliance with the
equipment standards has been exceptionally high. By 1981, one shipping
research firm already had evidence that new tankers were being built with
Chapter 54 Page 122 September 11, 200819:53

122 International Environmental Issues and Debates

Table 1: Percentage of crude oil tankers weighing over 70,000 deadweight tons with
segregated ballast tanks (SBT) and/or crude oil washing equipment (COW) onboarda

Tanker Construction Date (and MARPOL

1979 and Earlier 198082 Post 1982
Equipment Onboard (SBT or COW) (SBT only) (SBT and COW)

SBT and COW 32% 94% 98%

SBT or COW 94% 99% 100%
Total SBT (alone and with COW) 36% 98% 99%
Total COW (alone and with SBT) 89% 95% 99%
SBT alone 4% 4% 1%
COW alone 58% 1% 1%
Neither SBT nor COW 6% 1% 0%
MARPOL compliance level 94% 98% 98%
Data reflect tankers in the fleet as of 31 December 1991.
MARPOL = International Convention for the Prevention of Pollution from Ships.
Source: Electronic version of Clarkson Research Studies, Ltd., The Tanker Register (London:
Clarkson Research Studies, Ltd., 1991).

SBT and existing tankers were being retrofitted with SBT and/or COW.47
Recent national and international studies as well as industry experts reveal a
common assumption that all tankers comply with the equipment standards
although none provides empirical support for this assumption.48
Analysis of previously unavailable data on equipment installed on large
tankers supports these perceptions (see Table 1).49 Among large tankers in
the fleet at the end of 1991, 94 percent of tankers built in 1979 or earlier had
installed SBT or COW as required, 98 percent of those built between 1980
and 1982 had installed SBT as required, and 98 percent of those built after
1982 had installed both SBT and COW as required. The gures not only
confirm remarkably high compliance rates but also document that tankers
of all nations, not merely those that supported the equipment requirements
during negotiation, have complied.
The variance between the subregimes is more remarkable when one
considers that both international politics and private economics would lead
us to expect higher compliance with the discharge standards, not the equip-
ment standards. The discharge standards had been adopted unanimously.
In contrast, several powerful nations opposed the equipment standards in
both 1973 and 1978. Tankers seeking the economic benefits of conserving
oil could have done so most cheaply by using the equipment-free option of
LOT, not by installing COW or the even more expensive SBT. Indeed, in
1978, one academic analyst, Charles Okidi, predicted that the enormous
costs of SBT would make compliance negligible.50
In short, the empirical evidence of higher compliance levels with the
equipment subregime runs contrary to predictions based on a simple anal-
ysis of exogenous power and interests. How do we explain what appears
Chapter 54 Page 123 September 11, 200819:53

Mitchell Regime Design Matters 123

to be a significant divergence between theory and observed outcomes? Was

any of the observed compliance treaty-induced? If so, what elements of the
equipment standards compliance system explain its greater success at elic-
iting compliance? The rest of this article answers these questions.

Was Compliance Treaty-Induced?

Before we can explain how one subregime produced such dramatically higher
compliance levels than another within the same issue-area, we need to assure
ourselves that we can accurately attribute this variance to features of the
regime. Taking realist analysis seriously requires that we avoid attributing
causation where only spurious correlation exists. Factors other than vari-
ation in the compliance systems of the two subregimes may explain the
observed behaviors. Did tanker owners and operators act any differently
than they would have in the absence of international regulations? The
following accounting of incentives to comply with regulations from both
within and outside of the regime strongly suggests (1) that increased use of
LOT owes more to economics than to international law, (2) that increased
installation of COW equipment owes much to economics but also reflects
the MARPOL regimes influences, and (3) that increased installation of SBT
largely is due to MARPOL influences.


Several pieces of evidence indicate that the 1969 rules had little to do with
the observed increase in the use of LOT by tanker operators. A large share
of tankers simply did not use LOT or comply with the discharge standards.
The continuing noncompliance with discharge standards did not result from
an inability to use LOT a noncomplex procedure that required no new
equipment but from insufficient incentives to use it.
The subregime itself produced few effective mechanisms for inducing
operators to adopt LOT. While I discuss these failures more fully below,
the discharge subregimes compliance system failed to induce the monitoring
and enforcement necessary to deter violations. The subregimes failure effec-
tively to detect, identify, prosecute, and penalize violators left tanker opera-
tors incentives to comply with it largely uninfluenced. As the official IMO
newsletter put it, Little has changed in the three decades since [1962]. The
problem is detecting a violation in the first place (which is difficult) and then
collecting sufficient evidence to prove the case in court (which has all too
often proved to be impossible).51
Given the absence of these pathways for regime influence, it is not
surprising to find that economic influences readily explain the pattern of
LOT usage. A tanker operators first-order incentives to use LOT depended
on the costs of recovering waste oil, the value of that oil, and the ownership
of the oil being transported. This last factor meant that oil companies had
Chapter 54 Page 124 September 11, 200819:53

124 International Environmental Issues and Debates

far greater incentives to adopt LOT than did independent transporters. The
latter carry oil on charter to cargo owners and are paid for the amount of
oil initially loaded, known as the bill-of-lading weight, not for the amount
delivered. Therefore, discharging waste oil at sea costs the independent trans-
porter nothing. Indeed, using LOT reduces the bill-of-lading weight in subse-
quent cargo by the amount of remaining slops, thereby reducing the payment
that the independent transporter receives. In contrast, operators that own
their cargoes, as oil companies usually do, can offset a LOT tankers slightly
smaller cargo capacity with the benefit of having all the oil it paid for deliv-
ered. At 1976 prices, the lower bill-of-lading weight cost the tanker owner
some $700, while the value of oil recovered benefited the cargo owner some
The decrease in average discharges of oil company tankers in the 1970s
and the absence of a similar decrease in discharges of independent tankers
correlate more with these divergent incentives and with rising oil prices than
with any treaty proscription. Oil companies greater incentives to conserve
oil explain why their average discharges were lower than those of indepen-
dent tankers in 1972 and why they decreased discharges more rapidly after
the 1973 oil price hikes (see Figure 1). If the regime, rather than economics,
were influencing oil company behavior, these decreases should have occurred
only after the total discharge limits took legal effect in 1978, not after 1973.
The far smaller decrease in average discharge among independents reflects
the fact that conserved oil had little value to them.
Nevertheless, the OILPOL regime does appear to have been responsible
for the timing of LOT development in the early 1960s and to have at least
contributed to some adoption of LOT. Oil company representatives noted at
the time that they had developed LOT in response to the increasing pressures
for equipment requirements that were evident at the 1962 conference. The
facts that (1) oil prices remained constant throughout the 1960s and (2) LOT
involved a procedural not a technological breakthrough support this
more limited claim of regime influence. Declines in discharges by both oil
company and independent tankers before the oil price increases of 1973 and
declines in independent tankers discharges after 1973 also prove difficult
to attribute exclusively to economic factors. Having said this, however, it
remains clear that economic factors rather than the features of the subregime
were the dominant factors influencing tanker operators behavior.


The almost universal installation of COW equipment initially tempts one

to conclude that compliance was treaty-induced. The contrast in rates of
use of LOT and COW suggest that differences in the designs of the corre-
sponding subregimes may be responsible, given that both methods allowed
a tanker operator to reduce waste oil. However, closer evaluation reveals
that here, too, economic factors played an important role, although not an
exclusive one.
Chapter 54 Page 125 September 11, 200819:53

Mitchell Regime Design Matters 125

Like LOT, COW has economic as well as environmental benefits. COWs

costs include those for the washing machines and the additional time and
labor needed to wash tanks in port during delivery rather than during the
ballast voyage.53 As with LOT, the offsetting benefit of more delivered cargo
accrues to the cargo owner. However, the tanker operator also benefits: the
decrease in oil left on board increases the tankers effective cargo capacity
and reduces sludge buildup, which can lead to large repair and maintenance
costs. Compared with a tanker that was not practicing pollution control,
using COW produced a net savings per voyage of $9,000.
These economic incentives to adopt COW are borne out by the evidence
of the timing of its adoption. In many instances, tankers adopted COW
before required to do so by MARPOL. Recall that negotiators only incorpo-
rated COW requirements into MARPOL in 1978 and only made them appli-
cable to tankers built after 1982. Yet by the mid-1970s, many oil companies
had already incorporated COW as a standard operational procedure.54 This
timing does not correspond with the development of COW technology in
the late 1960s or with the deadline set by MARPOL. Instead, like LOT, it
corresponds with the rising oil prices of the 1970s.
The contrast to the SBT requirements also confirms the role of economics.
The higher capital costs of SBT and the significant reduction to cargo-
carrying capacity that SBT involved imposed a net cost per voyage on a
tanker with SBT of $1,500 relative to a tanker with no pollution-control
equipment. A new tanker installing both COW and SBT, as required by
MARPOL, faced costs of almost $8,000 per voyage. Owners of large tankers
built before 1980, who were allowed to choose between SBT and COW,
installed COW equipment on 89 percent of their tankers and SBT on
only 36 percent (see Table 1). Owners also installed COW equipment on
95 percent of large tankers built between 1980 and 1982, even though
MARPOL only required them to install SBT. COWs economic benefits
certainly appear to be a major influence on COW installations.
Several details suggest that economics were not the sole influence on
behavior, however. If they were, we should expect companies to achieve
the economic goal of conserving oil by the cheapest and most cost-effective
means possible, that is, by LOT, not COW. We should also expect to see
the same divergence between the behavior of independent carriers and oil
companies as we observed in the LOT case. Yet the 99 percent compliance
rate attests to the fact that all tanker owners were installing COW. The adop-
tion of COW more frequently than SBT does not imply that the subregime
was ineffective, only that when the subregime left owners with alternatives,
their choices were driven by costs. In contrast to clear flaws in the compli-
ance system supporting discharge standards, as I detail below, the design of
the compliance system supporting equipment requirements provided several
means of successfully reducing both the incentives and ability of tanker
owners to violate COW requirements. Thus, an interplay among economics
and subregime characteristics appears to have been the source of widespread
COW adoption.
Chapter 54 Page 126 September 11, 200819:53

126 International Environmental Issues and Debates


Adoption of the SBT standard provides an unambiguous example of

subregime influence on behavior. Unlike COW or LOT, tanker owners had
no economic incentives to install this technology. SBTs additional piping and
equipment added several million dollars to the cost of a new tanker, repre-
senting almost 5 percent of total cost.55 Installing SBT also reduced cargo
capacity, especially when installed on an existing tanker. Yet these costs
provided no offsetting benefits in the form of reduced cargo wastage. Even
those governments that had supported the 1978 proposal that all tankers be
retrofitted with SBT admitted that SBT would increase the cost of carrying
oil by 15 percent; some oil company estimates ran up to 50 percent.56 As
late as 1991, oil and shipping interests opposed mandatory SBT retrofitting
as being too expensive.57
The pattern of observed SBT installation follows that which one would
predict for behavior driven by effective treaty rules rather than economics.
Among tankers currently in the fleet, more than 98 percent of those required
to install SBT have done so despite the significant costs involved. Compliance
has been elicited even among those required to install both SBT and COW.
Rates of SBT installation among older tankers bolster the argument: among
tankers built before 1980, which MARPOL allowed to choose between SBT
and COW, only 36 percent have installed SBT. Indeed, owners installed
SBT alone on only 4 percent of older tankers but installed COW alone on
58 percent, suggesting that owners installed SBT only when a tanker was
already in dock to be retrofitted with COW. Figure 2 graphs the percentages

Figure 2: Percentage of tankers with segregated ballast tanks (SBT) and/or crude
oil washing equipment (COW) on board in 1991, by year of tanker construction


Percentage of tankers




Pre-1970 1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991
SBT tankers COW tankers
Source: Electronic version of Clarkson Research Studies, Ltd., The Tanker Register (London:
Clarkson Research Studies, Ltd., 1991), provided to the author.
Chapter 54 Page 127 September 11, 200819:53

Mitchell Regime Design Matters 127

of current tankers using SBT and COW by year of construction. The timing
of the increase in the number of tankers installing SBT seen in the gure
reinforces the conclusion that owners installed SBT only under the regula-
tory threat posed by the subregimes compliance system. In short, owners
have installed SBT only when MARPOL required them to do so. As one
analyst noted, If there were not a regulatory requirement, there would not
be SBT.58 Within several years, the subregime had caused a radical change
in tanker owner behavior.
One alternative explanation of SBT adoption deserves special attention.
At least one analyst has claimed that hegemonic pressures exerted by the
United States explain the success of MARPOL.59 Certainly the negotia-
tion history demonstrates that the SBT requirements of 1973 and 1978
resulted directly from threats of unilateral U.S. regulation. Indeed, the
United States is the only state that adopted national legislation containing
a faster schedule and broader application of equipment requirements than
MARPOL.60 Admitting that MARPOLs rules resulted from hegemonic pres-
sures, however, does not imply that subsequent behaviors result from that
same pressure. As international diplomats are all too well aware, resources
adequate to elicit votes for a resolution during a conference may prove inad-
equate to cause corresponding changes in behavior. The relevant question
is, Could the United States, through unilateral measures, have induced so
many tanker owners to install SBT? Available evidence suggests not.
While the United States wields tremendous diplomatic leverage, it wields
nothing near hegemonic power in oil transportation markets. Since the
United States became concerned about oil pollution in the late 1960s, it
has been responsible for less than 5 percent of new tankers built, less than
7 percent of tanker registrations, and less than 20 percent of world oil
imports.61 Given SBTs high costs, oil transportation companies would have
been more likely to respond to unilateral U.S. equipment requirements by
installing SBT on a sufficient number of tankers to service the U.S. market
than by installing it on all tankers. Nor has the United States sought to link
MARPOL enforcement with other issues through threats of sanctions or
through side-payments.62 Indeed, in terms of power to control oil tankers,
Japan which opposed SBT requirements in both 1973 and 1978 consis-
tently has controlled larger shares of tanker construction, tanker registration,
and oil imports than the United States. Thus, while the nation most strongly
supporting universal installation of SBT could not have imposed its prefer-
ences on its own, the nation most capable of doing so consistently opposed
such measures.

Mechanisms of Influence

Compliance with discharge standards via the use of LOT was largely
an artifact of economic factors. Compliance with requirements for SBT and
COW has been both higher and more clearly the result of the treaty. Rival
Chapter 54 Page 128 September 11, 200819:53

128 International Environmental Issues and Debates

explanations of economic influences and international political hegemony

prove incapable of adequately explaining the observed outcomes and behav-
iors. The equipment subregime succeeded at inducing reluctant tanker
owners to spend considerable money on additional equipment that provided
them with no economic benefit. The following assessment of the exact means
by which it did so simultaneously reinforces the conclusion that the regime
caused the change in behavior and identifies design features that might be
used to improve the effectiveness of other regimes.
Which of the many differences between the two subregimes best explain
the different levels of observed compliance? In what ways did the equipment
subregime get it right where the discharge standards failed? In subse-
quent portions of this article, I shall show how the design of the equipment
regime induced compliance by (1) eliciting monitoring and enforcement and
(2) reducing opportunities for violation.

Enhancing Transparency

The equipment subregime had one major advantage over the discharge
subregime in its significantly higher transparency level. Violations of the SBT
and COW requirements simply were far easier to observe than violations of
any discharge standard.
Consider the two compliance information systems. Both OILPOL and
MARPOL required tanker captains to note discharges in record books and to
make those books available to port authorities for inspection. This obvious
reliance on self-incrimination made naval and aerial surveillance programs
the more common means of detecting illegal discharges. The total discharge
standard of one fifteen-thousandth of cargo capacity improved on this
system by providing a criterion that could be monitored by tank inspections
in port without relying on information supplied by the tanker captain. Prac-
tically speaking, these inspections were restricted to ports in oil-exporting
states, since discharges occurred after delivery, on a tankers return to port
to load more cargo.
In contrast, the compliance information system for equipment standards
relied on the fact that buying or retrofitting a tanker requires the knowledge
and consent of at least three other actors: a builder, a classification society,
and an insurance company. Agents in each of these industries would know
of a violation even before it was committed. MARPOL also required flag
state governments, or classification societies nominated by them, to survey
all tankers to ensure compliance before issuing the required International
Oil Pollution Prevention (IOPP) certificate and to conduct periodic inspec-
tions thereafter.63 As part of the process of evaluating tankers to provide
insurers with the information needed to set rates, classification societies
regularly monitor compliance with international construction requirements
through representatives stationed in shipyards.64 Finally, MARPOL gave all
port states the legal authority to inspect a tankers IOPP certificate and its
equipment to ensure compliance with the equipment requirements.
Chapter 54 Page 129 September 11, 200819:53

Mitchell Regime Design Matters 129

The equipment standards subregime made violations more transparent

than violations in the discharge standards subregime in several ways. To
begin with, regulating the tanker buildertanker buyer transaction yielded a
drastically reduced number of events to be monitored. While several thou-
sand tankers ply the worlds oceans, they are owned, built, and classified by
only a few owners, shipyards, and classification societies. A tanker making
ten trips per year could violate the total discharge standard three hundred
times in its thirty-year life but could only violate the equipment requirements
Equipment standards also required authorities to monitor far fewer loca-
tions to detect violations. The discharge process standards 100 ppm, clean
ballast, and 60 1/m required patrols of wide areas of ocean to detect slicks
that often could not be linked with the responsible tanker. As early as the
1920s, experts had noted the difficulties of such a monitoring system.65 The
addition of total discharge limits allowed detection of violations while a
tanker was in an oil port, a procedure involving far fewer resources. Unfor-
tunately, most oil-exporting states had exhibited little interest in preventing
marine pollution; many such states were not even parties to MARPOL.66
Inspections to verify compliance with equipment standards could occur
in developed oil-importing states, which had shown far more interest in
enforcement. The shift from the 100 ppm and 60 1/m limits to total discharge
limits improved dramatically the practical ability to detect violations. The
shift from total discharge limits to equipment standards improved the regime
further by increasing incentives for monitoring among those who already
had the practical ability to monitor.
Equipment standards dramatically eased the problem of obtaining evi-
dence needed to sanction a violator. The standards eliminated any reliance
on self-incrimination by the perpetrator of a violation. Detecting an equip-
ment violation and identifying its perpetrator also were not time-sensitive.
Successful detection and identification of a violation had to occur within
hours for violations of the initial standards and within days for total
discharge violations but could wait for years for equipment violations.
Authorities also faced several difficulties in transforming detection of a
discharge at sea into a case worthy of prosecution. In what can be called
passive voice violations, often a tanker could not be identified as respon-
sible for a detected slick: authorities could only say a violation had been
committed. Even if a responsible tanker could be identified, determining
whether the 100 ppm or 60 1/m criterion had been exceeded generally was
difficult. The total discharge standard could have eliminated this problem,
but oil-exporting states never established inspection programs. These flaws
in the design of the discharge standards compliance system were not neces-
sarily inherent or insurmountable. For example, some analysts proposed
placing observers on all tankers to verify compliance with discharge stan-
dards.67 Even without such a system, governments could have established
enough ocean patrol and in-port inspection programs to make detec-
tion of discharge violations likely. However, such programs would have
Chapter 54 Page 130 September 11, 200819:53

130 International Environmental Issues and Debates

involved huge expenditures of resources to produce only a low probability

of successful deterrence.
In the first years after OILPOL was signed, evidence quickly demon-
strated that only the Federal Republic of Germany and the United Kingdom
were making any significant efforts to monitor compliance with discharge
standards.68 By the late 1970s, the Americans, British, Dutch, and French had
instituted aerial surveillance programs.69 Many other countries used aerial
surveillance during the 1980s.70 However, these programs were most often
small and nonsystematic. The Dutch program flew more surveillance flights
per year in the late 1970s than at any time in the 1980s, and the United
States discontinued its program in the 1980s due to budgetary pressures.71
Reports to IMO from 1983 to 1990 show that only one-quarter of the sixty-
seven MARPOL signatories had any programs to detect discharges at sea.72
British and Dutch data confirm the problems of identifying perpetrators: the
British could link detected spills to tankers in only 22 percent of cases and
the Dutch, in only 14 percent.73
The entry into force of total discharge standards in 1978 allowed inspec-
tors in oil-loading ports to assume that any incoming tanker with all tanks
free of slops had violated the very low limit placed on total discharges.
However, even those oil-exporting states that were party to MARPOL had
strong disincentives to inspect ships in their ports: ports that were conducting
inspections were less attractive loading sites than neighboring ports that
were not conducting inspections. Not surprisingly, most governments did
not alter their enforcement strategies in response to the greater potential
for enforcement provided by the promulgation of total discharge stan-
dards. In contrast, considerable evidence confirms that the equipment regime
significantly changed the ways in which nations and classification societies
conducted tanker inspections. Many of the states that originally had opposed
the 1973 and 1978 U.S. proposals for equipment regulations subsequently
have conducted the in-port inspections needed to detect violations. In 1982,
the maritime authorities of fourteen European states signed a Memorandum
of Understanding on Port State Control, committing themselves annually to
inspect 25 percent of ships entering their ports for violations of maritime
treaties, including MARPOL.74 Notably, until 1992, the memorandum of
understanding explicitly excluded inspections for discharge violations from
its mandate, limiting cooperation to inspection for equipment violations.
Even though several member states had voted against SBT, all fourteen have
included checks of IOPP certificates in the thousands of inspections they
conduct each year. In reports to the IMO secretariat, five additional coun-
tries and the United States have reported finding discrepancies in tankers oil
pollution certificates. Canada, Japan, Poland, and Russia have major port
inspection programs, and ten Latin American states have recently signed
an agreement similar to the European memorandum.75 While these coun-
tries undoubtedly vary widely in how frequently and carefully they conduct
inspections, all have made inspections for MARPOL-required equipment a
standard element of their inspection programs.
Chapter 54 Page 131 September 11, 200819:53

Mitchell Regime Design Matters 131

The effectiveness of these governmental inspections depends at least

in part on the initial issue of accurate IOPP certificates by flag states or
classification societies designated by them. Reports to IMO for 1984 to 1990
show that missing and inaccurate pollution certificates declined steadily
from 9 percent to 1 percent; the memorandum of understanding secretariat
reports similar declines from 11 percent to 3 percent.76 These trends suggest
that after an initial period of learning how to issue and inspect certificates,
classification societies and governments both now issue thorough and accu-
rate certifications. Like port state governments, flag states and classification
societies appear to have altered their behavior to become active participants
in the equipment subregimes compliance information system. It would seem
unlikely that classification societies and flag states would have responded in
the same fashion to U.S.-only legislation.
The greater transparency of violations of equipment requirements served
perhaps most importantly to reassure other tanker owners that their own
compliance would not place them at a competitive disadvantage in the
marketplace. An environmentally concerned tanker operator inclined to
comply with the discharge standards could not escape the knowledge that
others probably would not comply. The economic incentives to discharge
oil at sea, the absence of transparency about who was and who was not
complying, and the attendant inability of enforcement efforts to effectively
deter discharges precluded any assumption other than that many competi-
tors would violate the discharge standards to reduce their costs. The greater
transparency of equipment requirements assured a tanker owner installing
SBT and COW that all other owners also were doing so. Each company
could rest assured that its competitors also would have to incur equipment
costs or be sanctioned for not doing so.
The equipment standards provided the foundation for a compliance infor-
mation system far more transparent than was possible under the discharge
subregime. In response, even governments that had opposed the adoption
of the requirements conducted inspections for compliance. The subregimes
compliance information system channeled the behavior of both govern-
ments and classification societies into monitoring activities that supported
the regime. It did so by ensuring that those actors with incentives to monitor
compliance also had the practical ability and legal authority to do so. The
transparency of the system improved the ability to deter violations and simul-
taneously reassured tanker owners that their own compliance would not
place them at a competitive disadvantage with respect to other owners.

Facilitating Potent but Low-Cost Sanctions

Greater transparency translated into higher levels of compliance with equip-

ment standards only because the compliance system also induced likely
and potent sanctions. The noncompliance response system of the discharge
subregime failed to do the same. Even after a violation was detected, tanker
operators were unlikely to be successfully prosecuted and equally unlikely
Chapter 54 Page 132 September 11, 200819:53

132 International Environmental Issues and Debates

to receive a stiff penalty. In contrast, the equipment subregime authorized

governments to use the administrative sanction of detention, which made
both the likelihood and the cost of being penalized far higher for the equip-
ment standards than for discharge standards. The incentives and abilities of
governments to prosecute and to impose large penalties for violation were far
lower under the discharge standards than under the equipment standards.
Detected discharge violations frequently remained unprosecuted because
the subregime relied on customary international law with its deference to
enforcement by flag states. Both OILPOL and MARPOL required a govern-
ment that detected a discharge violation at sea to forward all evidence to
the flag state for prosecution. Only if a tanker discharged illegally within a
states twelve-mile territorial sea and then entered a port of that state could
that state prosecute a tanker registered elsewhere. Flag states have generally
been less than aggressive in following up on evidence referred to them.77 Flag
states often lack the ability to prosecute, since tankers flying their flag may
rarely enter their ports. They also have few incentives to prosecute because
vigorous enforcement on their part would induce owners to take their regis-
trations, and the large associated fees, to a less scrupulous state.78 The fact
that pollution occurred off another states coastline and that many devel-
oping flag states lack vocal environmental constituencies only reinforced
these disincentives to prosecute. In short, the flag states with the authority to
prosecute lacked incentives to do so, and the coastal states with the incentives
to prosecute lacked the authority to do so.
Under the discharge standards, even states sincerely seeking to prosecute
and convict a violator faced major obstacles to success. As already noted,
evidence of a violation often failed to produce a violator, and otherwise
convincing evidence often failed to meet the legal standards of proof needed
for conviction. Evidentiary hurdles should have decreased with the prohibi-
tion of discharges that produced visible traces. However, even with aerial
photographs of discharges, tankers frequently avoid conviction.79 Between
1983 and 1990, port and coastal states discarded for lack of evidence an
average of 36 percent of cases occurring in territorial seas and successfully
convicted and fined less than 33 percent of all detected violators.80 An addi-
tional 20 percent of high-seas cases referred to flag states were not prose-
cuted for the same reason, and less than 15 percent of all referrals resulted
in fines being imposed.81 Indeed, according to Paul Dempsey, from 1975
through 1982 ninety-two percent of all fines were imposed through port
state enforcement.82 Many experts had hoped that the clearer evidence from
inspections for total discharge violations would overcome these problems,
but, according to E. J. M. Ball, there is no record of a single case where the
one fifteen-thousandth rule was used for prosecution.83
When conviction was successful, governments rarely imposed penalties
adequate to deter future discharge violations as required by MARPOL.84
Although governments have the ability and legal authority to impose high
fines, the conflicting goals of the judiciary often inhibit them from doing
so. Most states courts are reluctant to impose fines disproportionate to
Chapter 54 Page 133 September 11, 200819:53

Mitchell Regime Design Matters 133

the offense to compensate for low detection and conviction rates. The prin-
ciple that the punishment should fit the crime places an upper bound on
fines that may be too low to successfully deter violation, if detection and
prosecution is difficult. Since 1975, the average fine imposed by states never
has exceeded $7,000 and actually has decreased over time.85 A Friends of the
Earth International study concluded that fines have remained very low in
comparison to the price the vessel would have to pay for using port reception
facilities.86 Even when a large penalty is assessed, the delays between initial
violation and final sentencing and the reluctance of most states to detain
tankers for minor discharge violations often mean that the responsible tanker
and crew have long since left the states jurisdiction, making fine collection
difficult. Owen Lomas points out that the problem is further exacerbated by
the fact that shipowners and their insurers routinely indemnify the masters
of their ships against fines imposed upon them for oil pollution.87
In place of the discharge subregimes legal system of prosecution, convic-
tion, and fines, the equipment subregime relied on quite different responses
to noncompliance. The most immediate sanctions involved the ability
of classification societies, insurers, and flag state governments to withhold
the classification, insurance, and pollution prevention certificates that a
tanker needed to conduct international trade. As John Foxwell put it, tankers
cannot get insurance without certification, and cant get certification
without compliance.88 These sanctions amounted to preventing any ille-
gally equipped tanker from doing business. Even if an owner could devise
a means to avoid these direct economic effects, a noncompliant tanker that
could not trade to all ports would still bring a far lower price in the large
tanker resale market.89
Besides these market-based sanctions, the equipment subregime obligated
port states either to detain tankers with false pollution prevention certificates
or inadequate equipment or to bar them from port.90 As administrative sanc-
tions, these responses skirted both flag state and port state legal systems
and the associated sensitivities regarding legal sovereignty. Paradoxically,
this strategy made port states more likely to use detention and flag states
more willing to accept it. Detention also had the virtue that even low usage
by a few major oil-importing states forced tanker owners to choose between
risking detention and the more costly option of not trading to those lucrative
markets. Authorizing developed states to detain violating tankers effectively
moved the right to sanction to countries that had far greater domestic polit-
ical pressures to use it.
Coupling the equipment requirements themselves with these adminis-
trative sanctions completely eliminated the legal and evidentiary problems
that make even clear violations of discharge standards difficult to prose-
cute successfully. Detention imposed opportunity costs on a tanker oper-
ator of several thousand dollars per day, and forced retrofitting could cost
millions of dollars far exceeding the fines for discharge violations.91 Deten-
tion had the positive quality that it was not so costly as to be considered a
disproportionate response to the crime but was costly enough to deter other
Chapter 54 Page 134 September 11, 200819:53

134 International Environmental Issues and Debates

violations. In short, detention was simultaneously more likely and more

While many states inspected tankers for compliance with equipment
requirements, most have not detained noncompliant ships frequently. IMO
records from 1984 to 1990 reveal that seven of fifteen states, including Japan,
have detained ships at least once. Only Germany, the United Kingdom, and
the United States have detained ships often. This undoubtedly reflects a reluc-
tance on the parts of some states to detain foreign tankers as well as the fact
that most tankers were equipped appropriately in the first place.
Although few states detained ships, available evidence supports the
conclusion that the subregime altered enforcement behavior. Not one of the
states that detained ships began to do so until after MARPOL took effect
in 1983.92 Even the United States waited until that year ten years after
the detention provision had been accepted. Consider the counterfactual: it
is unlikely that the United States would have detained tankers for breaching
U.S.-only requirements for SBT, even though it had the practical ability to
do so. Without MARPOL, such detentions would have constituted a major
infringement of flag state sovereignty. If the use of the more costly detention
sanction had reflected an exogenous increase in the interests of states in envi-
ronmental enforcement, fines for discharge violations should have increased
at the same time. Yet, as states began to use detention, fines did not increase
dramatically.93 Finally, public goods theory predicts that actors will tend not
to enforce rules that supply benefits to other parties.94 Contrary to theory,
however, European states and the United States spend far more on enforcing
equipment standards a public good that improves the global ocean envi-
ronment than on enforcing discharge standards off their own coastlines
the benefits of which would be more private.
The equipment subregime operated not by convincing reluctant actors to
enforce rules with which they disagreed but by removing the legal barriers
that inhibited effective enforcement by those states and nonstate actors
willing to enforce them. Classification societies had interests in ensuring that
the tankers they classified were able to trade without fear of detention. The
incorporation of equipment requirements into their classification criteria
provided the foundation for insurers to penalize noncompliant tankers.
The willingness of a few environmentally concerned oil-importing states to
inhibit tankers that lacked the required equipment from trading freely posed
an extremely potent threat to a tanker owner. However, the ability and will-
ingness of these states to threaten this sanction depended on removing inter-
national legal barriers to its use. Once these barriers were removed, imposing
sanctions involved few costs to those imposing them, whether classification
societies, insurers, or port state authorities. It thereby made detention more
likely, even though it created no new incentives for states to impose sanc-
tions. In a case of nothing succeeds like success, the various threats of the
equipment subregimes noncompliance system led to initial compliance by
almost all tankers, making it rare that sanctions ever needed to be imposed.
Chapter 54 Page 135 September 11, 200819:53

Mitchell Regime Design Matters 135

Building on Existing Institutions

The oil pollution control regime induced implementation of those provisions

that involved few direct costs to governments. Monitoring and enforcement
proved especially likely when their costs were pushed off-budget by depu-
tizing private, nonstate actors to issue certificates and conduct inspections.
Piggy backing monitoring and enforcement efforts onto existing government
programs also has been successful in accomplishing the regimes objectives
with only minor program modifications and minimal cost. Governments
have tended to ignore or put little effort into those stipulations that require
significant new expenditure of government resources.
MARPOLs equipment subregime fostered monitoring by allowing
governments to delegate responsibility for surveys to classification soci-
eties. This in turn increased the likelihood that tankers would be initially
surveyed and subsequently inspected; additionally, the quality of inspec-
tions increased. Delegation also helps developing flag states, many of which
lack the resources, the practical ability, and the incentives to conduct such
inspections. MARPOL allowed such states to fulfill their treaty commitments
by assigning classification and inspection responsibilities to actors who
often had greater access to and more resources with which to conduct such
inspections. Classification societies also had strong incentives to conduct
accurate surveys as a means of protecting their business reputations and
avoiding problems with insurance companies. The strategy thus simultane-
ously removed these tasks and the resources they required from the hands
of governments and placed them in the hands of actors who could more
easily accomplish them. Classification societies already had infrastructures
to monitor tanker purchases for safety, financing, and insurance purposes.
Adding pollution control to their long inspection checklists required only
marginal changes to existing procedures.
The many inspection programs operated by developed port states parallel
this pattern. Like classification societies, the maritime authorities of the
European memorandum of understanding states, the United States, and
other states interested in enforcing the equipment requirements could make
simple, low-cost alterations to port state inspections already being conducted
for safety, customs, and other purposes. The recent establishment of a
Latin American memorandum and current negotiations for an Asian-Pacific
memorandum suggest that the equipment subregime has provided states with
a low-cost means to implement their international commitments as their
interests in enforcement increase. In contrast, where states have had to incur
significant new costs to implement treaty provisions, they have proved highly
unlikely to do so. Detection of discharge violations required development
of completely new surveillance programs. Most developed states have not
established large, ongoing surveillance programs. Even in environmentally
concerned states, aerial surveillance programs have tended to be relatively
small and subject to the vagaries of domestic budget battles.
Chapter 54 Page 136 September 11, 200819:53

136 International Environmental Issues and Debates

In the realm of compliance, the tendency for governments to push imple-

mentation costs onto nonstate actors is obvious. Compliance with the equip-
ment standards has involved significant costs to tanker owners and no direct
costs to governments. Yet, the treaty also required member states to ensure
that their ports had facilities to receive the slops that tankers traditionally
had discharged overboard. Although developed states have built more recep-
tion facilities, ports in the oil-loading states where they are most needed
still largely lack any facilities. IMO participants consistently have failed to
adopt proposals for developed states to fund reception facilities in developing
states. Even in many developed states, facilities are sorely inadequate relative
to the demands of their tanker traffic.95 Additionally, the task of determining
which ports have adequate facilities and which do not largely has fallen on
the shoulders of nonstate actors.96 National governments consistently have
argued that providing reception facilities is the responsibility of either the
ports themselves or of the oil industry.

Coercing Compliance rather than Deterring Violation

The compliance systems of the two subregimes differ most strikingly in

the fundamental model underlying their regulatory strategies. The equip-
ment standards subregime relied on a coerced compliance strategy, which
sought to monitor behavior to prevent violations from occurring in the
first place. The discharge standards subregime was deterrence-oriented,
attempting to detect, prosecute, and sanction violations after they occurred
to deter future violations.97 This basic difference in orientation made the
compliance task facing the equipment standards subregime more manageable
than that facing the discharge standards subregime. The underlying strategy
choice had important consequences for the level of compliance achieved:
inhibiting the ability to violate treaty provisions proved far more effective
than increasing the disincentives for violating them.
MARPOLs equipment standards created a remarkably effective system
for detecting and sanctioning violations. Even if this compliance system
had relied exclusively on the threat of oil-importing states detecting and
detaining noncompliant tankers, most tankers would have installed COW
and SBT. However, the equipment subregimes strength really came from
the fact that it rarely had to use the more potent sanctions it made possible.
Involving shipbuilders, classification societies, and insurers in the regula-
tory process could well have produced the same outcome even without the
additional threat of detention. The subregime relied on surveying behavior
and preventing violations rather than detecting and investigating them after-
wards.98 By regulating the business transaction of a tanker purchase rather
than the autonomous action of a discharge, the equipment rules allowed
identification of potential violators and made it harder to actually commit
a violation. Tanker captains faced many regular autonomous decisions
about whether to violate discharge standards. In contrast, tanker owners
only had to decide once between violating or complying with equipment
Chapter 54 Page 137 September 11, 200819:53

Mitchell Regime Design Matters 137

standards, and their decision required cooperation from other actors and
involved major economic consequences. Even before construction began,
classification societies and insurance companies were pressing for and moni-
toring compliance with international standards, helping avert violations
before they occurred. Classification societies, insurance companies, and flag
state inspectors could withhold the papers necessary to conduct business in
international oil markets, thereby frustrating any tanker owners attempt to
reap the benefits of sidestepping these standards.
Experience with the discharge standards had shown that many states
would not enforce pollution standards; indeed, even detention was used regu-
larly by only a few states. Given the costs of SBT, if deterrence had been the
major source of compliance, one would expect some tankers initially to have
violated the equipment standards in an attempt to identify which and how
many states actually would enforce the rules. Especially in light of their votes
against the requirements, owners might well have assumed less than rigorous
enforcement in places like France and Japan. Yet, compliance levels did not
follow a pattern of initial noncompliance followed by stiff sanctions and
subsequent compliance. The compliance system of the equipment subregime
succeeded by effectively restricting the opportunities to violate it rather than
making the choice of violation less attractive. The very low noncompli-
ance levels suggest that in most cases an owner simply decided it would
be impossible to convince a tanker builder, a classification society, and an
insurer to allow the purchase of a tanker without COW and SBT. Likewise,
tankers coming in for repairs and maintenance undoubtedly would have
found it difficult to explain why they were not planning on installing SBT
or COW, as required. The low levels of detected violations of the equipment
standards reveal that obstacles to committing a violation played a major
role in preventing such violations. New tankers have been built initially to
MARPOL standards, not retrofitted later in response to deterrence threats.
Even before MARPOLs equipment deadlines passed, owners were building
new and retrofitting older tankers to meet the requirements.
The equipment subregime may have been as successful as it was precisely
because it produced a redundant regulatory system. It established compli-
ance information and noncompliance response systems that prevented most
violations but could successfully deter any actors who might otherwise
have considered violating it. As the experience with discharge standards
clarifies, deterrence-based strategies often require the successful completion
of a complex chain of actions to be effective. The initial discharge stan-
dards subregime faced problems at almost every step of the process: detecting
violations, identifying violators, prosecuting violators, and imposing potent
sanctions. The shift to total discharge standards eliminated or mitigated
some of these problems, but the problems remaining left overall deter-
rent levels essentially unchanged. A tanker captain evaluating the expected
costs of violating OILPOLs or MARPOLs discharge standards could only
conclude that the magnitude and likelihood of a penalty were quite small.
Successful deterrence strategies must ensure that the whole legal chain
Chapter 54 Page 138 September 11, 200819:53

138 International Environmental Issues and Debates

operates smoothly, since the breakdown of any link can significantly impair
its effectiveness.


Nations can design regime rules to improve compliance. This article has
demonstrated that, even within a single issue-area, reference to design
features of compliance systems surrounding particular provisions is neces-
sary to explain observed variance in compliance. In the regime regulating
intentional oil pollution, the same governments and corporations with the
same interests during the same time period complied far more frequently
with rules requiring installation of expensive equipment than they did with
rules limiting total discharges of oil. Where theories of hegemonic power and
economic interests fail to explain this variance, differences in the subregimes
compliance systems readily explain why the former subregime led powerful
actors to comply with it while the latter did not.
The equipment standards elicited significantly higher compliance because
they selected a point for regulatory intervention that allowed for greater
transparency, increased the likelihood of forceful responses to detected viola-
tions, built on existing institutions, and coerced compliance by preventing
actors from violating them rather than merely deterring actors from doing so.
In any regime, the distribution of state power and interests and the traits of
relevant economic sectors constrain, but fail to fully explain, the regulations
to which states will agree and the degree of possible compliance. By acknowl-
edging these limits and realizing that the same goal often may be achieved by
regulating quite different activities, policymakers can improve compliance
by regulating those sectors more vulnerable to pressures for compliance and
by facilitating the efforts of those governments and nonstate actors more
likely to implement and enforce such regulations. This matching of regula-
tory burdens to expected behavior places the careful choice of the regimes
primary rules at the center of any effective compliance system.
Once such primary rules have been established, careful crafting of the
compliance information system and the noncompliance response system
can further increase the likelihood of compliance. Oil pollution regula-
tions succeeded by facilitating the goals of, placing responsibilities on,
and removing the legal and practical barriers limiting those governments
and private actors predisposed to monitor and enforce agreements, not by
imposing obligations on recalcitrant actors. Inducing compliance required an
integrated system of rules and processes that placed actors within a strategic
triangle of compliance so that they had the political and economic incentives,
practical ability, and legal authority to perform the tasks necessary to imple-
ment the treaty.99 When such efforts succeeded, governments and private
actors acted differently than they would have in the absence of the regime.
When such efforts failed, opportunities to increase compliance were missed.
We should not expect treaties to achieve perfect compliance.100 Nevertheless,
Chapter 54 Page 139 September 11, 200819:53

Mitchell Regime Design Matters 139

negotiators can and should design and redesign them to maximize compli-
ance within the constraints that power and interests impose.
Eliciting compliance is only one of the criteria on which we would want
to judge a regimes rules. Indeed, the value of compliance itself rests on
the assumption that more compliance makes the treaty itself more effective.
In the oil pollution case, compliance with the equipment rules involved at
least as great a reduction in intentional discharges as did compliance with
the discharge standards. Thus, we can safely infer that the higher compli-
ance levels under the former rules also led to increased treaty effectiveness,
a fact confirmed by a consensus among most experts that intentional oil
discharges have declined since MARPOL took effect.101 However, I am not
arguing here for command and control regulations but for considering
compliance levels along with efficiency, cost, and equity as an important
evaluative criteria in regime design. The cheaper, more flexible, and more
efficient discharge standards simply failed to induce the level of compliance
needed to achieve a socially desired outcome; yet the costs of the equipment
standards may have exceeded the benefits of that outcome. In cases in which
more efficient solutions elicit compliance sufficient to achieve a policy goal,
they are clearly preferable. If expected compliance with such solutions appears
low, effective regime design requires evaluating whether the benefits of higher
compliance outweigh the expense and inefficiency of alternative solutions.
Can we apply the findings developed from studying these two oil pollu-
tion cases to other issue-areas? Initial selection of a difficult collaborative
problem with characteristics common to many international collabora-
tion problems provides some confidence that we can do so. Other treaties
provide anecdotal support for some of the findings reported herein. Nuclear
powers consistently have sought to increase the transparency of arms control
treaties through progressively tighter provisions for data exchange and on-
site inspections. Although the experience with Iraq provides a dramatic
example of failure, the nuclear materials and missile technology control
regimes rely primarily on coerced compliance models of regulation, seeking
to prevent countries from violation in the first instance. Human rights
regimes frequently have used information from nongovernmental organi-
zations to monitor compliance with their provisions. These design features
seem likely to be the source of some regime compliance, but confirming that
conclusion requires considerably more research. The solutions adopted in
the oil pollution regime also undoubtedly cannot be applied to all regimes
or even to all environmental regimes. Wildlife and habitat protection, for
example, can rarely be achieved through technological solutions or quan-
titative requirements that can be easily monitored. In other instances, the
solutions to new problems will not be able to build on existing infrastruc-
tures and institutions. The strategies available to international regulators will
depend at least in part on features unique to the problem being addressed.
Analysts have already shown how regimes influence behavior in realms
involving security.102 How the impacts of similar compliance systems vary
Chapter 54 Page 140 September 11, 200819:53

140 International Environmental Issues and Debates

across security, economic, human rights, or environmental regimes remains

one of many important future questions.
Whether the nations of the world can collaborate to resolve the many
international problems, both environmental and otherwise, that face them
will depend not on merely negotiating agreements requiring new behaviors
but on ensuring that those agreements succeed in inducing governments,
industry, and individuals to adopt those new behaviors. We can hope and
work for a day when all nations and their citizens are sufficiently concerned
about peace, economic well-being, human rights, and the environment that
we will not need international law to criminalize the behavior threatening
those values and to dictate more benign behaviors. Until then, however,
careful crafting and recrafting of international treaties provides one valuable
means of managing the various problems facing the nations of the world.


. The epigraph is from J. L. Brierly, The Outlook for International Law (Oxford:
Clarendon Press, 1944), pp. 12.
1. This article summarizes the arguments made in Ronald B. Mitchell, Intentional Oil
Pollution at Sea: Environmental Policy and Treaty Compliance (Cambridge, Mass.: MIT Press,
2. The term compliance system comes from Oran Young, Compliance and Public
Authority: A Theory with International Applications (Baltimore, Md.: Johns Hopkins Univer-
sity Press, 1979), p. 3.
3. See, for example, Abram Chayes and Antonia Handler Chayes, On Compliance,
International Organization 47 (Spring 1993), pp. 175205; Young, Compliance and Public
Authority; Roger Fisher, Improving Compliance with International Law (Charlottesville:
University Press of Virginia, 1981); and W.E. Butler, ed., Control over Compliance with Inter-
national Law (Boston: Kluwer Academic Publishers, 1991).
4. See, for example, Abram Chayes, An Inquiry into the Workings of Arms Control
Agreements, Harvard Law Review 85 (March 1972), pp. 90569; Coit D. Blacker and Gloria
Duffy, eds., International Arms Control: Issues and Agreements, 2d ed. (Stanford, Calif.:
Stanford University Press, 1984); and Antonia Handler Chayes and Paul Doty, Defending
Deterrence: Managing the ABM Treaty into the Twenty-f irst Century (Washington, D.C.:
Pergamon-Brasseys International Defense Publishers, 1989).
5. See, for example, John S. Duffield, International Regimes and Alliance Behavior:
Explaining NATO Conventional Force Levels, International Organization 46 (Autumn 1992),
pp. 81955; Ethan Kapstein, Governing the Global Economy: International Finance and the
State (Cambridge, Mass.: Harvard University Press, 1994); and Joseph M. Grieco, Coopera-
tion Among Nations: Europe, America, and Non-tariff Barriers to Trade (Ithaca, N.Y.: Cornell
University Press, 1990).
6. For example, see Peter Haas, Robert Keohane, and Marc Levy, eds., Institutions for
the Earth: Sources of Effective International Environmental Protection (Cambridge, Mass.:
MIT Press, 1993); Peter H. Sand, Lessons Learned in Global Environmental Governance
(Washington, D.C.: World Resources Institute, 1990); and Peter M. Haas, Do Regimes Matter?
Epistemic Communities and Mediterranean Pollution Control, International Organization 43
(Summer 1989), pp. 377403. Current projects that deal with questions of regime compliance
and effectiveness (and their principal investigators) include those being conducted at, or with
funding from, Dartmouth College (Oran Young and Marc Levy); the European Science Founda-
tion (Kenneth Hanf and Arild Underdal); the Foundation for International Environmental Law
and Diplomacy (James Cameron); the Fridtjof Nansen Institute (Steinar Andresen); Harvard
Chapter 54 Page 141 September 11, 200819:53

Mitchell Regime Design Matters 141

University (Abram Chayes and Antonia Chayes); Harvard University (William Clark, Robert
Keohane, and Marc Levy); the International Institute for Applied Systems Analysis (David
Victor and Eugene Skolnikoff); and the Social Science Research Council (Edith Brown Weiss
and Harold Jacobson).
7. Thomas M. Franck, The Power of Legitimacy Among Nations (New York: Oxford
University Press, 1990).
8. Case selection that holds these other factors constant avoids the notorious difficulties of
measuring power and interests and allows us to attribute variance in collective outcomes to the
impact of institutional arrangements with some degree of confidence; see Oran Young, Inter-
national Cooperation: Building Regimes for Natural Resources and the Environment (Ithaca,
N.Y.: Cornell University Press, 1989), p. 208. On difficulties in measuring power, see David
A. Baldwin, Power Analysis and World Politics: New Trends Versus Old Tendencies, World
Politics 31 (January 1979), pp. 16193.
9. Michael McGinnis and Elinor Ostrom, Design Principles for Local and Global
Commons, Workshop in Political Theory and Policy Analysis, Bloomington, Ind., March
1992, p. 21. Olsons argument that small groups supply public goods more often than large
groups assumes that group members benefit from providing the good, which is not true in the
oil pollution case; see Mancur Olson, The Logic of Collective Action: Public Goods and the
Theory of Groups (Cambridge, Mass.: Harvard University Press, 1965), p. 34.
10. See Arthur A. Stein, Why Nations Cooperate: Circumstance and Choice in Interna-
tional Relations (Ithaca, N.Y.: Cornell University Press, 1990); and Robert Axelrod and Robert
O. Keohane, Achieving Cooperation Under Anarchy: Strategies and Institutions, in Kenneth
Oye, ed., Cooperation Under Anarchy (Princeton, N.J.: Princeton University Press, 1986).
11. The quotation is from Hans Joachim Morgenthau, Politics Among Nations: The
Struggle for Power and Peace, 5th ed. (New York: Alfred A. Knopf, 1978), p. 299. See also
Kenneth Waltz, Theory of International Politics (Reading, Mass.: Addison-Wesley Publishing
Co., 1979), p. 204; and Susan Strange, Cave! Hic Dragones: A Critique of Regime Analysis, in
Stephen D. Krasner, ed., International Regimes (Ithaca, N.Y.: Cornell University Press, 1983),
pp. 33754 at p. 338. For a contrasting view, see Young, International Cooperation, p. 62.
12. Morgenthau, Politics Among Nations, p. 267.
13. On this distinction, see Stein, Why Nations Cooperate.
14. See, for example, Abram Chayes and Antonia Chayes, Compliance Without Enforce-
ment: State Behavior Under Regulatory Treaties, Negotiation Journal 7 (July 1991), pp. 31130;
Young, International Cooperation; Robert O. Keohane, Reciprocity in International Rela-
tions, International Organization 40 (Winter 1986), pp. 127; and Krasner, International
15. Joseph S. Nye, Jr., Nuclear Learning and U.S.Soviet Security Regimes, International
Organization 41 (Summer 1987), pp. 371402.
16. See, for example, Louis Henkin, How Nations Behave: Law and Foreign Policy (New
York: Columbia University Press, 1979), p. 47; Young, International Cooperation, p. 62; and
Chayes and Chayes, Compliance Without Enforcement, p. 31.
17. At the extreme, if all parties violated treaty provision A and complied with treaty provi-
sion B, they could all be classified as in partial compliance, ignoring the important variance in
compliance rates.
18. See Haas, Keohane, and Levy, Institutions for the Earth; George W. Downs and
David M. Rocke, Tacit Bargaining, Arms Races, and Arms Control (Ann Arbor: University of
Michigan Press, 1990); Charles Lipson, Why Are Some International Agreements Informal?
International Organization 45 (Autumn 1991), pp. 495538; and Chayes and Chayes, On
Compliance, pp. 18892.
19. The Exxon Valdez wrecked in Prince William Sound, Alaska, on 24 March 1989.
20. For comparison, the Exxon Valdez spilled thirty-five thousand tons.
21. National Academy of Sciences, Petroleum in the Marine Environment (Washington,
D.C.: National Academy of Sciences, 1975). See also National Academy of Sciences and
National Research Council, Oil in the Sea: Inputs, Fates, and Effects (Washington, D.C.:
National Academy Press, 1985).
Chapter 54 Page 142 September 11, 200819:53

142 International Environmental Issues and Debates

22. See, for example, National Academy of Sciences and National Research Council, Oil in
the Sea; and Joint Group of Experts on the Scientific Aspects of Marine Pollution (GESAMP),
The State of the Marine Environment, Reports and Studies no. 39 (New York: United Nations,
23. For the history of oil pollution control from the 1920s through the 1970s, see Sonia
Zaide Pritchard, Oil Pollution Control (London: Croom Helm, 1987); for a history from the
1950s through the 1970s, see R. Michael MGonigle and Mark W. Zacher, Pollution, Politics,
and International Law: Tankers at Sea (Berkeley: University of California Press, 1979).
24. International Convention for the Prevention of Pollution of the Sea by Oil, 12 May
1954, Treaties and Other International Agreements Series (TIAS), no. 4900 (Washington, D.C.:
U.S. Department of State, 1954).
25. For an excellent description of a regime more focused on developing scientific under-
standing of an environmental problem, see Levys description of the regime on European acid
precipitation in Marc Levy, European Acid Rain: The Power of Tote-board Diplomacy, in
Haas, Keohane, and Levy, Institutions for the Earth, pp. 75132.
26. See International Convention for the Prevention of Pollution from Ships (MARPOL),
2 November 1973, reprinted in International Legal Materials (ILM), vol. 12 (Washington,
D.C.: American Society of International Law, 1973), p. 1319 (hereafter cited by abbreviation,
volume, and year); and Protocol of 1978 Relating to the International Convention for the
Prevention of Pollution from Ships, 17 February 1978, reprinted in ILM, vol. 17, 1978, p. 1546
(hereafter cited together as MARPOL 73/78).
27. MGonigle and Zacher, Pollution, Politics, and International Law, p. 100.
28. J.H. Kirby, The Clean Seas Code: A Practical Cure of Operation Pollution, in Third
International Conference on Oil Pollution of the Sea: Report of Proceedings, Rome 79 October
1968 (Winchester, England: Warren and Son, 1968), pp. 20119.
29. Kirby, The Clean Seas Code, p. 206.
30. Assembly resolution 391, IMCO/IMO doc. resolution A.391(X), 1 December 1977,
Annex, par. 5. All document citations herein refer to IMCO/IMO documents housed in the
IMO Secretariat library. They are numbered similarly: according to issuing committee (abbre-
viated), meeting number, agenda item, and document number. Information documents are
designated by Inf. prior to the document number. Circulars are designated by Circ., issuing
committee, and circular number only. Resolutions are designated by adopting body, resolution
number, and meeting number. Conference documents are cited by abbreviated conference title,
preparatory meeting number, agenda item, and document number. Hence the above resolu-
tion citation would be interpreted as the 391st resolution adopted by the 10th meeting of the
(IMCO) assembly.
31. 1969 Amendments to the International Convention for the Prevention of Pollution of
the Sea by Oil, 21 October 1969, reprinted in Bernd Ruster and Bruno Simma, eds., Interna-
tional Protection of the Environment: Treaties and Related Documents (Dobbs Ferry, N.Y.:
Oceana Publications, 1975).
32. Kirby, The Clean Seas Code, p. 208.
33. See Kirby, The Clean Seas Code, pp. 200 and 209; and William T. Burke, Richard
Legatski, and William W. Woodhead, National and International Law Enforcement in the
Ocean (Seattle: University of Washington Press, 1975), p. 129.
34. MGonigle and Zacher, Pollution, Politics, and International Law, p. 108.
35. See statements submitted by the U.S. delegation to the 13th Preparatory Session for an
International Conference on Marine Pollution in 1973: IMCO/IMO doc. MP XIII/2(c)/5, 23
May 1972. (Using note 30 as a guide, this would be the 5th document issued relating to agenda
item 2[c]). See also doc. MP XIII/2(a)/5, 1 June 1972; G. Victory, Avoidance of Accidental and
Deliberate Pollution, in Coastal Water Pollution: Pollution of the Sea by Oil Spills (Brussels:
North Atlantic Treaty Organization [NATO], 26 November 1970), p. 2.3.
36. MGonigle and Zacher, Pollution, Politics, and International Law, p. 114.
37. Jacob W. Ulvila, Decisions with Multiple Objectives in Integrative Bargaining, Ph.D.
diss., Harvard University, 1979, appendix A1.1.
38. MGonigle and Zacher, Pollution, Politics, and International Law, pp. 122 and 130.
Chapter 54 Page 143 September 11, 200819:53

Mitchell Regime Design Matters 143

39. See Sonia Z. Pritchard, Load on Top: From the Sublime to the Absurd, Journal of
Maritime Law and Commerce 9 (April 1978), pp. 185224 at p. 194.
40. For an excellent discussion of state positions during both the 1973 and 1978 confer-
ences, see MGonigle and Zacher, Pollution, Politics, and International Law, pp. 10742.
41. See, for example, the estimate of 0.3 percent in James E. Moss, Character and Control
of Sea Pollution by Oil (Washington, D.C.: American Petroleum Institute, 1963), p. 47, and
the estimate of 0.4 percent in IMCO/IMO doc. OP I/21, 15 January 1965, of the Oil Pollution
subcommittee. (Using note 30 as a guide, this indicates the only document issued relating to
agenda item 21 at the 1st meeting of the subcommittee.)
42. Unfortunately, oil companies discontinued the surveys after 1977. Personal communi-
cation from Arthur McKenzie, Tanker Advisory Center, New York, 1992.
43. Informational document of the Marine Environment Protection Committee:
IMCO/IMO doc. MEPC XVI/Inf.2, 4 November 1981.
44. IMCO/IMO doc. MEPC 30/Inf.13, 19 September 1990, p. 15.
45. For example, IMO, Tanker Owners Urge Increase in Facilities Accepting Oily Wastes,
International Environment Reporter, 8 March 1989, p. 130.
46. See, for example, C.J. Camphuysen, Beached Bird Surveys in the Netherlands 1915
1988: Seabird Mortality in the Southern North Sea Since the Early Days of Oil Pollution
(Amsterdam: Werkgroep Noordzee, 1989); United States Coast Guard, Polluting Incidents
In and Around U.S. Waters (Washington, D.C.: U.S. Department of Commerce, 1973 and
197586); N. Smit-Kroes, Harmonisatie Noordzeebeleid: Brief van de Minister van Verkeer
en Waterstaat (Tweede Kamer der Staten-Generaal: 17408) (Harmonization of North Sea
policy: Letter from the Minister of Transport and Waterways; Lower House of Parliament)
(The Hague: Government Printing Office of the Netherlands, 1988); IMCO/IMO doc. MEPC
21/Inf. 8, 21 March 1985; and Second International Conference on the Protection of the North
Sea, Quality Status of the North Sea: A Report by the Scientif ic and Technical Working Group
(London: Her Majestys Stationery Office, 1987), p. 14.
47. Drewry Shipping Consultants, Ltd., The Impact of New Tanker Regulations, Drewry
publication no. 94 (London: Drewry Shipping Consultants, Ltd., 1981), p. 25.
48. See IMCO/IMO doc. MEPC 30/Inf.13, 19 September 1990, p. 8; Second International
Conference on the Protection of the North Sea, Quality Status of the North Sea, p. 57; Pieter
Bergmeijer, The International Convention for the Prevention of Pollution from Ships, paper
presented at the 17th Pacem in Maribus conference, Rotterdam, August 1990, p. 12; and
personal interview with E.J.M. Ball, Oil Companies International Marine Forum, London, 26
June 1991;
49. The detailed statistics in Table 1 and Figure 2 were developed from an electronic version
of Clarkson Research Studies, Ltd., The Tanker Register (London: Clarkson Research Studies,
Ltd., 1991) generously provided by Clarkson Research Studies, Ltd.
50. Charles Odidi Okidi, Regional Control of Ocean Pollution: Legal and Institutional
Problems and Prospects (Alphen aan den Rijn, The Netherlands: Sijthoff and Noordhoff, 1978),
p. 34.
51. See p. 9 of Cleaner Oceans: The Role of the IMO in the 1990s, IMO News, no. 3,
1990, pp. 612.
52. The following discussion of the costs of LOT, COW, and SBT draws heavily on William
G. Waters, Trevor D. Heaver, and T. Verrier, Oil Pollution from Tanker Operations: Causes,
Costs, Controls (Vancouver, B.C.: Center for Transportation Studies, 1980).
53. Drewry Shipping Consultants, Ltd., Tanker Regulations: Enforcement and Effect,
Drewry publication no. 135 (London: Drewry Shipping Consultants, Ltd., 1985), p. 25.
54. See M.G. Osborne and J.M. Ferguson, Technology, MARPOL, and Tankers:
Successes and Failures, IMAS 90: Maritime Technology and the Environment (London:
Institute of Marine Engineers, 1990), p. 62; Testimony of William Gray, in U.S. Congress,
House Committee on Government Operations, Oil Tanker Pollution: Hearings Before
the Subcommittee on Government Activities and Transportation, 18 and 19 July 1978,
95th Congress, 2d sess., 1978, p. 92; and IMCO/IMO doc. MEPC V/Inf.A, 27 April 1976.
Chapter 54 Page 144 September 11, 200819:53

144 International Environmental Issues and Debates

55. See Philip A. Cummins, Dennis E. Logue, Robert D. Tollison, and Thomas D. Willett,
Oil Tanker Pollution Control: Design Criteria Versus Effective Liability Assessment, Journal
of Maritime Law and Commerce 7 (October 1975), pp. 18182; and Charles S. Pearson,
International Marine Environmental Policy: The Economic Dimension (Baltimore, Md.: The
Johns Hopkins University Press, 1975), p. 98.
56. See IMCO/IMO doc. MEPC V/Inf. 4, 8 March 1976, p. A18; and MGonigle and
Zacher, Pollution, Politics, and International Law, p. 134.
57. See IMCO/IMO doc. MEPC 31/8/5, 4 April 1991; and Osborne and Ferguson, Tech-
nology, MARPOL, and Tankers, p. 62.
58. Personal interview with Sean Connaughton, marine transportation analyst, American
Petroleum Institute, Washington, D.C., 8 April 1992.
59. Jesper Grolin, Environmental Hegemony, Maritime Community, and the Problem of
Oil Tanker Pollution, in Michael A. Morris, ed., NorthSouth Perspectives on Marine Policy
(Boulder, Colo.: Westview Press, 1988).
60. Drewry Shipping Consultants, Ltd., Tanker Regulations, p. 11.
61. See Lloyds Register of Shipping, Annual Summary of Merchant Ships Completed
(London: Lloyds Register of Shipping, various years); Lloyds Register of Shipping, Statistical
Tables (London: Lloyds Register of Shipping, various years); and United Nations, Statistical
Yearbook (New York: United Nations, various years).
62. For examples of U.S. sanctions to enforce the International Convention for the Regu-
lation of Whaling, see Gene S. Martin, Jr., and James W. Brennan, Enforcing the Interna-
tional Convention for the Regulation of Whaling: The Pelly and PackwoodMagnuson Amend-
ments,Denver Journal of International Law and Policy 17 (Winter 1989), pp. 27192.
63. MARPOL 73/78, Annex I, Regulations 4 and 5.
64. Personal interview with John Foxwell, Shell International Marine, London, 27 June
65. Charles Hipwood, United Kingdom Marine Department, cited in Pritchard, Oil Pollu-
tion Control, p. 23.
66. While Iran and Iraq never have signed either agreement, Kuwait, Libya, Nigeria, Qatar,
Saudi Arabia, the United Arab Emirates, and Venezuela have never signed MARPOL.
67. Cummins et al., Oil Tanker Pollution Control, p. 171.
68. IMCO/IMO doc. OP/CONF/2, 1 September 1961.
69. See James Cowley, IMO and National Administrations, IMO News, no. 4, 1988,
pp. 611; Smit-Kroes, Harmonisatie Noordzeebeleid; and IMCO/IMO doc. MEPC 21/Inf.9,
25 March 1985.
70. James McLoughlin and M.J. Forster, The Law and Practice Relating to Pollution
Control in the Member States of the European Communities: A Comparative Survey (London:
Graham and Trotman, 1982).
71. Personal interview with Daniel Sheehan, U.S. Coast Guard, Washington, D.C., 9 April
72. Gerard Peet, Operational Discharges from Ships: An Evaluation of the Application of
the Discharge Provisions of the MARPOL Convention by Its Contracting Parties (Amsterdam:
AIDEnvironment, 1992), annexes 5 and 10.
73. See United Kingdom Royal Commission on Environmental Pollution, Eighth Report:
Oil Pollution of the Sea (London: Her Majestys Stationery Office, 1981), p. 195; and Smit-
Kroes, Harmonisatie Noordzeebeleid.
74. Memorandum of Understanding on Port State Control, reprinted in ILM, vol. 21,
1982, p. 1.
75. Acuerdo de Via del Mar: Acuerdo Latinoamericano Sobre Control de Buques por el
Estado Rector Del Puerto (Via del Mar Accord: Latin American Accord on Port State Control
of Vessels), 5 November 1992. The text of the agreement is almost identical to the text of
the Memorandum of Understanding on Port State Control, cited above. Reference to the
agreement can be found in Secretariat of the Memorandum of Understanding on Port State
Control, Annual Report (The Hague: The Netherlands Government Printing Office, 1992).
Chapter 54 Page 145 September 11, 200819:53

Mitchell Regime Design Matters 145

76. Secretariat of the Memorandum of Understanding on Port State Control, Annual Report
(The Hague: The Netherlands Government Printing Office, various years).
77. See Organization for Economic Cooperation and Development (OECD), OECD
Study on Flags of Convenience, Journal of Maritime Law and Commerce 4 (January 1973),
pp. 23154.
78. Paul Stephen Dempsey, Compliance and Enforcement in International LawOil Pollu-
tion of the Marine Environment by Ocean Vessels, Northwestern Journal of International
Law and Business 6 (Summer 1984), pp. 459561 and p. 576 in particular.
79. See ibid., p. 526; and personal interview with Ronald Carly, Ministry of Transportation,
Brussels, 10 June 1991.
80. Peet, Operational Discharges from Ships, pp. 1718, Tables 11 and 12; and Marie-Jose
Stoop, Olieverontreiniging door schepen op de noordzee over de periode 19821987: opsporing
en vervolging (Oil pollution by ships on the North Sea 19821987: Investigations and prose-
cution) (Amsterdam: Werkgroep Noordzee, July 1989).
81. Ronald Bruce Mitchell, From Paper to Practice: Improving Environmental Treaty
Compliance, Ph.D. diss., Harvard University, Cambridge, Mass., 1992, Table 51.
82. Dempsey, Compliance and Enforcement in International Law, p. 537.
83. Personal interview with E.J.M. Ball.
84. MARPOL 73/78, Article 4(4).
85. Mitchell, From Paper to Practice, Table 45.
86. IMCO/IMO doc. MEPC 29/10/3, 15 January 1990.
87. Owen Lomas, The Prosecution of Marine Oil Pollution Offences and the Practice of
Insuring Against Fines, Journal of Environmental Law, vol. 1, no. 1, 1989, p. 54. See also
IMCO/IMO doc. MEPC 32/14/3, 17 January 1992.
88. Personal interview with John Foxwell, Shell International Marine, London, 27 June
89. Bergmeijer, The International Convention for the Prevention of Pollution from Ships,
p. 12.
90. MARPOL 73/78, Articles 5(2) and 5(3).
91. Personal interviews with John Foxwell; and with Richard Schiferli, Memorandum of
Understanding Secretariat, Rijswijk, The Netherlands, 17 July 1991.
92. Personal interview with Daniel Sheehan.
93. See Peet, Operational Discharges from Ships, annex 15; and Dempsey, Compliance
and Enforcement in International Law.
94. Axelrod and Keohane, Achieving Cooperation Under Anarchy.
95. IMCO/IMO doc. MEPC 30/Inf.32, 12 October 1990.
96. See IMCO/IMO docs. MEPC 19/5/2, 21 October 1983; MEPC 22/8/2, 8 October 1985;
and MEPC 30/Inf.30, 15 October 1990.
97. Neither strategy was incentive-based, as was the funding of compliance under the
Montreal Protocol and Framework Convention on Climate Change. For development of the
distinction between these three strategies, see Albert J. Reiss, Jr., Consequences of Compli-
ance and Deterrence Models of Law Enforcement for the Exercise of Police Discretion, Law
and Contemporary Problems 47 (Fall 1984), pp. 83122; and Keith Hawkins, Environment
and Enforcement: Regulation and the Social Def inition of Pollution (Oxford: Clarendon Press,
98. Reiss, Consequences of Compliance and Deterrence Models of Law Enforcement for
the Exercise of Police Discretion.
99. I am indebted to Robert O. Keohane for the notion of a strategic triangle of compliance.
100. Chayes and Chayes, On Compliance.
101. See Ronald B. Mitchell, Intentional Oil Pollution of the Oceans, in Haas, Keohane,
and Levy, Institutions for the Earth, pp. 183248.
102. See Robert Jervis, Security Regimes, in Krasner, International Regimes, pp. 17394;
and Duffield, International Regimes and Alliance Behavior.
Chapter 55 Page 146 September 11, 200819:54

Assessing the Effectiveness of International
Environmental Agreements: The Case
of the 1985 Helsinki Protocol
Evan J. Ringquist and Tatiana Kostadinova

he last few decades have been marked by the emergence of a number
of environmental protection regimes in the international arena. Early
research examined the conditions under which international environ-
mental agreements (IEAs) were most likely to arise (Keohane, Haas, and
Levy 1993; Sprinz and Vaahtoranta 1994; Young 1994). A second era saw
scholars focusing upon the implementation of and compliance with these
agreements (Barrett 1994; Jnicke and Weidner 1997; Weiss and Jacobson
1998). Currently, scholars have begun to investigate questions of effec-
tiveness: i.e., do international environmental agreements actually improve
environmental quality (see Helm and Sprinz 2000)?1 In a major review of
research on international environmental policy, Zurn concludes that regime
effectiveness has become a driving force in the analysis of international
relations (1998, 649; see also Young 1999).
When assessing effectiveness, researchers face the difficult puzzle of sepa-
rating the signal of regime effects from the noise arising from the impacts
of a wide range of other forces (Young 2001, 100). To be sure, evalu-
ating the effectiveness of international agreements faces several formidable
obstacles. First, many of these agreements are so recent that estimating
their effects is premature, while consistent and reliable data for estimating
the effects of older agreements are difficult to obtain. Second, indicators
of environmental quality are affected by a host of factors independent of
these agreements. Finally, the overwhelming majority of IEAs are charac-
terized by voluntary participation, which means that any evaluation must
generate measures of program effectiveness from a self-selected sample
a notoriously difficult proposition.2 In this study, we assess the effects of
the 1985 Helsinki Protocol for reducing sulfur dioxide emissions in Europe,
paying particularly close attention to overcoming the three obstacles noted
above. Specifically, we investigate the effectiveness of the Helsinki Protocol
by estimating models that (a) employ comparable data on sulfur dioxide
emissions from 19 countries in Europe over a 15-year period (19801994);
(b) control for non-Protocol factors affecting emissions; and (c) control for

Source: American Journal of Political Science, 49(1) (2005): 86102.

Chapter 55 Page 147 September 11, 200819:54

Ringquist and Kostadinova Evaluating Environmental Agreements 147

the nonrandom process by which nations choose to ratify the Protocol. We

conclude the manuscript by discussing the implications of our results for
studies of international environmental cooperation.

Environmental Protection in the International System

Increasingly, scientists and policy practitioners are recognizing that many

environmental problems have regional or global consequences and thus must
be addressed through regional or global policies and institutions (Vig and
Axelrod 1999). By one count, more than 90 international environmental
agreements have been negotiated since the 1972 U.N. Conference on the
Human Environment in Stockholm, Sweden (Chasek 2001). The main objec-
tive of these agreements is to guarantee compliance with established rules,
norms, standards, and policies in the absence of world government.
While international agreements are becoming increasingly common tools
for addressing environmental problems, designing these agreements and their
associated implementing institutions is no easy task. Policy makers who
would tackle environmental problems in this manner face a familiar
dilemma: while the global environment would benefit from compliance with
IEAs (i.e., there are collective gains from cooperation), there is an even
greater incentive for nations to defect, gambling that they might obtain the
benefits of the agreement through free riding. The most commonly used tools
for combating free riding sanctions are rarely available to the architects
of international environmental agreements and are even seen as counterpro-
ductive by some scholars (Levy 1993). Indeed, the overwhelming majority
of IEAs are voluntary, with few sanctions for noncompliance or free riding
other than expressions of dissatisfaction within the international community.
Under these conditions, it is reasonable to ask how environmental agree-
ments are supposed to solve international environmental problems; i.e., how
they can motivate nations to exceed pollution control activities that they
would take in the absence of the agreements, thereby obtaining gains from
collective action.3

Why International Agreements May Be Effective

Scholars have offered a variety of reasons why voluntary IEAs might induce
states to pursue environmental protection more aggressively. At the national
level, effectively addressing environmental problems requires (a) increased
governmental concern regarding the issue; (b) a hospitable contractual envi-
ronment emphasizing monitoring and credible commitments; and (c) the
governing capacity necessary to change policy. International environmental
agreements can affect all of these elements through the generation and
dissemination of scientific research, the creation of monitoring networks,
and the generation of public pressure (Helm 1998). Nations might also
act more aggressively in pursuing environmental protection if such actions
could be tied to other foreign policy issues like trade, and international
Chapter 55 Page 148 September 11, 200819:54

148 International Environmental Issues and Debates

agreements provide an avenue for these types of cross-policy linkages (Haas

1993). Finally, Keohane, Haas, and Levy (1993) argue that these agree-
ments provide a channel for the productive expression of domestic political
demands for environmental protection.
Downs (2000) provides an excellent summary of the various perspec-
tives on how voluntary international agreements might generate collective
environmental benefits. Downs argues that these agreements might be able
to reshape national preferences through the processes of legitimation, role
redefinition, and reflection. Through legitimation, national leaders who
regard the rules and provisions of international regimes as legitimate will
follow the regimes rules without a full cost-benefit calculation (i.e., they
will be less likely to free-ride). Regimes become more legitimate by maxi-
mizing participation, through the use of consensus-based decision rules,
and as they age (older regimes are seen as more legitimate). Through role
redefinition, international environmental regimes define new roles for partic-
ipating nations, and these new roles cause nations to redefine their iden-
tities (e.g., as trustworthy environmental stewards) and attempt to live
up to these identities. Finally, the reflectivist appraisal process gener-
ates cooperation through repeated interactions with other nations. Such
repeated interaction brings a better understanding of other nations, builds
trust among participants, and fosters positive feedback as successful coop-
eration engenders more cooperation. In order to capture these coopera-
tive benefits, constructivist international environmental regimes should
be inclusive (to maximize legitimation and role redefinition effects), have
weak rules (to maximize participation and discourage abstentions), employ
consensual decision-making processes, and eschew sanctions and other coer-
cive enforcement mechanisms (as these interfere with legitimation and the
positive feedback aspects of reflectivism; Downs 2000; see also Levy 1993).
While international environmental agreements may be effective, not all
agreements are equal in this regard. One of the best developed frameworks
focusing on this question is that of Arild Underdal. In brief, Underdal (2002)
posits that the effectiveness of international environmental agreements is a
product of the interaction between problem malignancy and the institutional
setting created to address the problem. The core element of problem malig-
nancy is incongruity, or the extent to which rational action on the part of
individual nations produces a suboptimal outcome for all nations. The key
element of the institutional setting is capacity, or the ability of these insti-
tutions to generate resources, establish rules and goals without unanimity
among participants, and enforce the terms of the agreement (a very noncon-
structivist position). Regimes that address malign problems with low levels
of institutional capacity are less likely to be effective, and vice versa.

Why International Agreements May Be Ineffective

The discussion above demonstrates that there are plausible causal arguments
for the effectiveness of IEAs. Counterpoints to these arguments come from
Chapter 55 Page 149 September 11, 200819:54

Ringquist and Kostadinova Evaluating Environmental Agreements 149

four directions. From a broadly theoretical standpoint, questions regarding

the effectiveness of international environmental agreements invoke debates
about international cooperation in general. The prospects for the emergence,
maintenance, and success of such regimes form the core of a theoretical
discourse dominated by the realist and the neoliberal schools of political
thought. According to the realist explanation of world politics, coopera-
tion is almost impossible to achieve and even more difficult to sustain in
a world of egoistic nation-states that act rationally under circumstances of
prevailing anarchy. National leaders main concerns are how to increase
absolute gains when interacting with others in order to keep powerful posi-
tions and guarantee national security. In this world of uncertainty, states are
not attracted to the idea of cooperation because of the potential for relative
gains by competitors-cum-partners who may use these gains to their advan-
tage (Grieco 1993). Maintaining cooperation would be even more difficult
because of the high probability for cheating and noncompliance with the
established rules. Applying this logic to environmental protection, commit-
ment to a pollution control regime may decrease capacities for economic
growth and competitiveness that exceed the uncertain benefits of improving
international air quality. At the same time, other countries may decide to
free ride and increase their political and economic capabilities at the expense
of ratifying nations. Realists, therefore, are skeptical of the effectiveness of
international environmental agreements (see Mearsheimer 1995).
Formal theorists have also expressed reservations about the effective-
ness of IEAs. For these scholars, the essential feature of voluntary agree-
ments is that they must be self-enforcing. With no sanctions, international
governing bodies must make joining the agreement attractive to all relevant
nations and provide incentives for these nations to abide by the terms of
the agreement. Barrett (2003) concludes that international environmental
regimes may achieve a high degree of cooperation among nations, but only
if the difference in net benefits between the regime and no-regime situations
is small. When agreements pursue improvements in environmental quality
far beyond what would be possible in a noncooperative world, cooperation
can not be sustained among a large number of countries. Downs (2000)
and Downs, Rocke, and Barsoom (1996) concur, noting that international
environmental regimes are almost always shallow and rarely require states
to do more than they would have done without the agreement. Thus, self-
enforcing international agreements may not be able to improve much upon
noncooperative outcomes in environmental protection.
Third, some scholars question the effectiveness of policy tools in general,
particularly those developed by international organizations. A leading propo-
nent of this perspective sums up the argument this way:    the material,
institutional and socio-cultural capacity of a country are more relevant in
determining the outcome of environmental policies than its choice of policy
instruments (Jnicke 1992, 47).4 Stated differently, it matters little what
environmental policies nations adopt; success in protecting the environment
is a function of economic development, the political system, social values,
Chapter 55 Page 150 September 11, 200819:54

150 International Environmental Issues and Debates

and the cultural framework of the country. Together, these contribute to a

capacity for modernization that countries must posses to be able to deal
effectively with modern problems like environmental protection.5 According
to this argument, international environmental agreements that are initiated
by supranational organizations and implemented downward are often inef-
fective mechanisms for protecting and improving environmental quality (for
a supporting view, see Ostrom 1990). Such an approach to environmental
protection hardly works because the higher the international level, the
more the environmental policy is reduced to verbal declarations (Jnicke
1992, 57).
Finally, international environmental agreements embody few of the char-
acteristics typically associated with effective public policies (see Mazmanian
and Sabatier 1989).6 When studying air pollution control in the United
States, for example, Ringquist (1993) finds that a basic formula for impro-
ving air quality is to enact stringent regulations, fund them adequately, and
enforce them. Magat and Viscusi (1990) concur, finding that regulatory
inspection and enforcement are critical to the success of water pollution
control regulation in the United States. For all their shortcomings, prescrip-
tive and coercive regulatory programs have often proven effective at reducing
pollution (Gunningham, Kagan, and Thornon 2003; Knoepfel and Weidner
1982; White 1982). The environmental standards found in voluntary inter-
national agreements are seldom backed up with inspection programs or sanc-
tions for noncompliance, and the institutions charged with overseeing the
implementation of these agreements almost never have resources adequate to
the task. Even advocates of these agreements note that there is good reason
for skepticism regarding their effectiveness, since nations retain control of
policymaking and implementation and the international institutions charged
with overseeing the agreements are weak (Keohane, Haas, and Levy 1993).
Thus, traditional policy scholars also might be skeptical of the effectiveness
of IEAs.

The 1985 Helsinki Protocol on Sulfur Dioxide Emissions

A Brief Description

In the 1970s, the United Nations Economic Commission for Europe (UN
ECE) identified the environment as an area of potential East-West cooper-
ation during the Cold War (see Wettestad 2002). The first product of this
cooperation was the Convention on Long-Range Transboundary Air Pollu-
tion (CLRTAP, or the Convention) signed by 34 governments and the
European Community in November 1979. According to the convention, the
signatories agreed to reduce air pollution through balanced development and
the use of the best available technology (UN ECE 1986, ix).
While the framework agreed to at the convention covered all types of
transboundary air pollutants, priority was given to the abatement of sulfur
Chapter 55 Page 151 September 11, 200819:54

Ringquist and Kostadinova Evaluating Environmental Agreements 151

dioxide pollution with the signing of the Helsinki Protocol in July 1985 (later
protocols covered particulate emissions, nitrous oxide emissions, emissions
of volatile organic compounds, and various other pollutants). The idea for
the protocol was introduced by Sweden, Norway, and Finland in 1983 and
later supported by West Germany. A binding agreement was reached that
called for reducing the sulfur dioxide emissions of each signatory by at least
30% compared to 1980 levels as soon as possible and at the latest by 1993
(UN ECE 1986, ix). The Protocol entered in to force in early 1987, and
20 parties to the convention have ratified the Protocol, with some of them
(e.g., Austria, Canada, France, and Sweden) having already made unilateral
commitments for larger reductions (McCormick 1998). Several convention
signatories refused to sign the Helsinki Protocol, including Greece, Ireland,
Portugal, Spain, the United Kingdom, and the United States.

Predicting the Effectiveness of the Helsinki Protocol

Using Underdals (2002) framework, the problem of reducing sulfur dioxide

is highly malignant; the transboundary character of the pollutant and the
resulting public good nature of pollution control creates a large incentive
for nations to free ride. With respect to institutional capacity, the CLRTAP
and the Helsinki Protocol are designed for weakness, in the constructivist
tradition (Wettestad 2002, 214; for a concurring view, see Barrett 2003). The
convention identified the transboundary air pollution problem and affirmed
the responsibilities of individual nations for air quality management in a
global plan, but imposed no specific obligations or binding commitments to
undertake particular measures. Emphasis was placed on building consensus
on a broader and less-specified basis, avoiding harsh rules and unambiguous
obligations that would have pushed away some of the negotiating parties.
The CLRTAP oversight commission has no independent source of funding,
and the main responsibilities of the five air pollution staff members include
planning the regular meetings of the commission (Levy 1993). The responsi-
bilities of the oversight commission include coordinating various national air
quality research programs, standardizing the measurement and monitoring
of air quality, gathering information on the air pollution control policies
of member nations, and collecting and disseminating scientific and policy
information. The commission engages in no research of its own, relying upon
member nations to conduct air quality research. Moreover, the commission
has no inspection or sanctioning authority, and the implementation of the
terms of the Helsinki Protocol (and the other protocols) relies wholly upon
domestic policy changes enacted by ratifying governments.

Evidence Regarding the Effectiveness of the Helsinki Protocol

Many international relations scholars are skeptical of the effectiveness of

IEAs (see above), and some of the best empirical work focusing on the
Helsinki Protocol supports this skepticism. For example, when examining
Chapter 55 Page 152 September 11, 200819:54

152 International Environmental Issues and Debates

national goals under the Protocol, Murdoch and Sandler conclude that most
treaty participants formulated mandates that they would have abided by,
even in the absence of a treaty (Murdoch and Sandler 1997, 142). More-
over, when simultaneously evaluating the decisions of nations to ratify the
Protocol and the sulfur emissions of nations under the Protocol, Murdoch,
Sandler, and Vijverberg (2003) found that though nations responded posi-
tively to relatively small emission reductions in neighboring states, they
tended to free ride if other nations reduce their emissions by large amounts.
While neither of these studies evaluated the effectiveness of the Protocol per
se, their conclusions are consistent with the proposition that international
agreements are unable to produce improvements in environmental quality
beyond what would be accomplished in the absence of cooperation.
Research explicitly evaluating the effectiveness of the Helsinki Protocol
comes to more optimistic conclusions. Based upon detailed case studies,
Wettestad (2002) concludes that CLRTAP in general and the Helsinki Pro-
tocol in particular have met with some success. Ratification of the protocol
was followed by a large amount of policy activity in ratifying states, and
compliance with the agreement is high. In addition, the regime clearly
contributed to a reduction in emissions, though other factors have helped
reduce emissions as well. In an early quantitative assessment, Levy (1993)
comes to an even more positive conclusion, finding that the Protocol was
effective in improving air quality in Europe. More technical analyses have
produced similar results. Using a spatial autoregressive model, Murdoch,
Sandler, and Sargent (1997) find that the Helsinki Protocol reduced emis-
sions beyond what would have been expected in the absence of the Protocol
(more pessimistic conclusions from their other work notwithstanding), but
that this effect was very small. Most recently, Helm and Sprinz (2000) sought
to assess both the relative and absolute effectiveness of the Protocol. Helm
and Sprinz take the unusual approach of establishing their no-regime and
perfect regime counterfactuals by obtaining the best judgments from experts
on what emission levels would have been in the absence of the Protocol and
what they would be under a perfectly effective Protocol. They then compared
these predictions to actual emission levels and aggregated these gures across
nations to come up with an index of program effectiveness. The effectiveness
score for the Helsinki Protocol is .39, indicating that emissions are lower
than they would be in the absence of the agreement (the index lower bound
of 0), but higher than they would be in a perfectly effective regime (the index
upper bound of 1).

Remaining Questions Regarding the Effectiveness

of the Helsinki Protocol

The studies summarized above have begun to answer the key question
of program effectiveness regarding the Helsinki Protocol. The work by
Chapter 55 Page 153 September 11, 200819:54

Ringquist and Kostadinova Evaluating Environmental Agreements 153

Murdoch and his colleagues (1997, 2003) and Helm and Sprinz (2000) is
particularly noteworthy in that it has dramatically advanced both the theory
and method of program evaluation in this area. Still, assessing the effective-
ness of international environmental agreements in general, and the Helsinki
Protocol in particular, could be improved by addressing three significant
shortcomings in the existing research. First, none of the existing studies
makes use of temporal variation when assessing program effectiveness.
Murdoch, Sandler, and Sargent (1997) and Helm and Sprinz (2000) measure
Protocol effectiveness by employing static change scores that assign a single
value for emissions to each nation in the data set. Both pollution emissions
and Protocol implementation are dynamic processes, so static models may
not be adequate for assessing program effectiveness. Moreover, given that
it can take a decade or more for programs to display significant impacts
(Mazmanian and Sabatier 1989), the four-year post-Protocol time period
employed by Murdoch et al., (1997, 2003) may be too short to adequately
assess Protocol effects.
Second, while these studies employ innovative methodologies, the
methods generate concerns regarding the reliability and validity of empir-
ical results. For Murdoch et al. (1997, 2003) these concerns stem from a
small sample size. Since the models these authors use employ a single change
score for each nation, this means that their conclusions arise from sample
sizes of roughly 25 with fewer than 20 degrees of freedom. Statistical results
generated under conditions of micronumerisity are notoriously unstable. For
Helm and Sprinz (2000), concerns stem from the expert judgments used
to generate the counterfactuals. The authors employ only one expert for
each nation in their study, and a different sample of experts might generate
different data. More importantly, the methodology itself may be in need of
refinement, since in their analysis actual emission reductions exceeded what
would be expected under a perfect regime in eight cases (see Young [2001]
for a similar critique).
The greatest shortcoming in the extant research seeking to evaluate the
effectiveness of the Helsinki Protocol is that none of this research recog-
nizes that nations do not ratify the Protocol at random.7 Participation in
the Protocol is voluntary, which means that nations self-select into the
treatment group (ratifying nations) and the control group (nonratifying
nations). Failing to account for the self-selected nature of treatment and
control groups poses grave dangers for estimates of program effectiveness.
Under conditions of nonrandom assignment, we have no way of knowing
if post-Protocol differences between ratifying and nonratifying nations are
due to the Protocol, or due to pre-existing differences between these two
groups of nations. Indeed, the work by Murdoch and Sandler (1997) strongly
suggests the latter in the case of the Helsinki Protocol. A relatively well-
developed technical literature exists regarding the estimation of program
effects from self-selected samples. We introduce this literature, and the asso-
ciated statistical techniques, in the next section.
Chapter 55 Page 154 September 11, 200819:54

154 International Environmental Issues and Debates

Selection Bias, Nonrandom Assignment,

and the Estimation of Policy Effects

A Basic Policy Effect Model

Assuming that the policy intervention takes place during time period k, the
simplest statistical model for estimating the effects of public policies takes
the following form

Yit = a + dPit + eit  where t > k (1)

Yit = a + eit  where t k

and the estimate of the policy or treatment effect is EdP = l, or d. In our
application, Y represents changes in sulfur dioxide emissions, and P is the
ratification of the Helsinki Protocol.

Policy Effects under Nonrandom Assignment

With nonrandom assignment to treatment and control groups, this basic

model is no longer accurate. Defining Yit as above and Yit as the value of Yit
in the absence of the policy, under nonrandom assignment;

EYit P = 1 = EYit P = 0 (2)

which simply means that in the absence of the policy, the expected value of
the dependent variable is different for the treatment and control groups. This
situation is common where entities are allowed to self-select into a program,
whether the policy is a job training program, a drug treatment program, or
an international protocol for reducing pollution emissions. In this situation,
the mean effect of the policy on those subject to the policy (i.e., the effect of
the treatment on the treated) is:

EYit Yit P = 1 (3)

The difference between this estimate of the policy effect and the estimate in
the basic model in Equation (1) becomes clear when we represent the policy
effect in terms of differences in the observed value of the dependent variable.
This policy effect is

EYit P = 1 EYit P = 0 = EdP = 1 + EYit P = 1 EYit P = 0 (4)

where the bracketed term is the effect of nonrandom assignment (Heckman

and Hotz 1989).
Including regressors in the basic policy effect equation offered above gives
us the linear model most commonly estimated by social scientists,

Yit = a + k X kit + dPit + eit (5)
Chapter 55 Page 155 September 11, 200819:54

Ringquist and Kostadinova Evaluating Environmental Agreements 155

The selection bias stemming from nonrandom assignment generates endo-

geneity of the policy intervention variable in this equation, so that
Eeit Pit  Xkit  = 0. It is well known that under these conditions, ordinary least
squares (OLS) will yield inconsistent estimates of both the policy effect (d)
and the other structural parameters
. In practical terms, models contami-
nated by this type of selection bias will almost always overestimate the effect
of the policy.

Model Estimation under Nonrandom Assignment

Various approaches have been developed that allow researchers to obtain

consistent estimates of policy effects under nonrandom assignment. In this
section we discuss the three most common approaches to estimating policy
effects with panel data and a diagnostic tool necessary to select the best
estimation routine from these options.
Control Function Estimators. The first class of estimators, dubbed
control function estimators (see Heckman and Hotz 1989; Heckman and
Robb 1985), assumes that differences between treatment and control groups
with respect to Yit are due to factors that are causally related to the treat-
ment status of units, but not to the outcome of interest. This process is
often termed selection on observables, since the factors leading to system-
atic differences between treatment and control groups can be observed and
modeled. Murdoch and Sandler (1997) suggest that just such process oper-
ated within the context of the Helsinki Protocol. These authors posit that
nations with larger forest regions, more political freedom, etc., were more
likely to ratify the protocol and that these factors lead to fundamental differ-
ences between ratifiers and nonratifiers that make estimating the effects of
the protocol quite problematic. The most appropriate approach under these
circumstances is to include the variables related to treatment status (Z) in
the equation estimating the effect of the policy, so that the full equation now
takes the form
Yit = a + k Xkit + Ck Zkit + dPit + eit (Heckman and Hotz 1989) (6)

Fixed Effects Estimators. This class of estimators assumes that the system-
atic differences between treatment and control groups with respect to Yit is
a function of unobserved but temporally invariant characteristics of units
(e.g., persons, countries), and these characteristics are often assumed to
affect both the probability of treatment and the effect of the treatment.
Thus, controlling for these effects eliminates the endogeneity of the treat-
ment noted above. The approach to generating consistent estimates of policy
effects under these circumstances will be familiar to anyone experienced in
estimating panel models with unit effects. The fixed-effects model can be
estimated by including cross-sectional dummy variables so that
Yit = a + k Xkit + i vi + dPit + eit (7)
Chapter 55 Page 156 September 11, 200819:54

156 International Environmental Issues and Debates

where vi represents the unit effects (Heckman and Hotz 1989;Wooldridge

Random Trend Estimators. This final class of traditional estimators
assumes that the systematic differences between treatment and control
groups with respect to Yit are a function of unit-specific rates of change in
Yit , represented here as wit . This situation is highly likely with respect to the
Helsinki protocol, since nation specific trends in sulfur dioxide emissions
varied dramatically in the pre-protocol period (Levy 1993). The random
trend approach commonly attributes some of the differences between treat-
ment and control groups to temporally invariant unit effects, so that the full
random trend estimation model takes the following form9
Yit = a + k Xkit + i vi + i wit + dPit + eit  (8)

Assessing the Adequacy of Selection Bias Estimators

Heckman and Hotz (1989) offer a number of relatively simple statis-

tical tests for assessing whether the models described above adequately
account for differences between treatment and control groups stemming
from nonrandom selection. With longitudinal data of the type employed
here, these specification tests assign a policy intervention or trend variable
to all members of the treatment group in the pre-policy period, and then
estimate the selection bias models using only pre-policy data. Since no policy
intervention has taken place, the parameters associated with the policy and
trend variables in the pre-policy phase should be nonsignificant. If significant
policy effects are found in the pre-policy phase, any differences between
treatment and control groups in the post-policy period almost certainly
reflect a continuation of these pre-policy phase differences, rather than any
true policy effect.

Modeling the Effects of the Helsinki Protocol

In this section we craft a model for investigating changes in sulfur dioxide

emissions for nations that participated in the 1979 Convention on Long-
Range Transboundary Air Pollution. This model contains three elements:
policy factors, external environmental factors, and national capacity and
development factors. Before discussing the components of the model,
however, it might be useful to reiterate two points made earlier. First, because
not all nations participating in the Convention signed the Helsinki Protocol,
our model compares changes in sulfur dioxide emissions between signa-
tory and nonsignatory nations. Second, since not all signatories ratified the
protocol at the same time, our model compares the relative progress made
in signatory states before and after ratification.
Due to various factors, we can not investigate these two questions for all
of the 33 nations that participated in the Convention. First, during our time
frame, both the Soviet Union and Yugoslavia ceased to exist, while other
Chapter 55 Page 157 September 11, 200819:54

Ringquist and Kostadinova Evaluating Environmental Agreements 157

newly independent nations were created. Many of these nations participated

in the Convention and/or signed the Protocol. Given that the boundaries and
the political, economic, and social character of these states changed dramat-
ically, they are omitted from the analysis. Second, several of the participants
to the Convention did not monitor sulfur dioxide emissions (e.g., Greece,
Romania, Iceland, the Holy See). With no data for the dependent variable, we
had to exclude these nations from the analysis. Finally, given our emphasis
on Europe, the United States and Canada are excluded from the analysis.
After these omissions, our data set includes information for 19 nations over
15 years.

The Dependent Variable

The Helsinki Protocol does not set targets for absolute emission levels.
Rather, the protocol establishes percentage reduction goals for each nation
(see above). Thus, our dependent variable is sulfur dioxide emissions in
each of the 19 nations identified in the appendix, from 1980 to 1994, as a
percentage of emissions in 1980.

Independent Variables

Policy Effects. While previous research has demonstrated that, under certain
conditions, public policy may be able to effectively address certain social
problems, there is significant disagreement over the effectiveness of public
policy in improving environmental quality. This uncertainty is particularly
evident with respect to the effectiveness of international environmental agree-
ments. In order to speak to this debate, our models of changes in air pollution
must include factors representing these agreements. Traditionally, policy
interventions are represented by a dummy variable taking on a value of 1 for
nations in years after signing the protocol (which captures level effects), and
a counter or lagged endogenous variable that takes the value of zero prior
to ratification of the protocol (which captures trend effects; see Berry and
Lewis-Beck 1986; McCleary and Hay 1980). We adopt this convention here,
employing a counter variable that measures the number of months since each
nation ratified the protocol (UN ECE 1994). If ratification and subsequent
efforts to meet the goals of the protocol reduced sulfur dioxide emissions,
the coefficient for both policy variables will be negative.10
External Environmental Factors. If we are to adequately assess the effects
of public policy in improving environmental quality, we need to control for
external environmental factors. For example, nations that consume more
fossil fuels and produce a higher level of industrial output will generate more
pollution, all other things equal. We represent industrial activity as the total
value of industrial output in each nation, in each year, measured in millions
of 1990 US dollars (European Marketing Data and Statistics 1993; Industrial
Development Global Report 1995; UN ECE 1997). Because our dependent
variable is measured as a percentage from a 1980 baseline, industrial activity
Chapter 55 Page 158 September 11, 200819:54

158 International Environmental Issues and Debates

is measured as percentages from a 1980 baseline as well. We expect that this

variable will produce a positive parameter estimate.
While our emissions model controls for industrial production, it does not
control for fossil fuel consumption. The combustion of fossil fuels, partic-
ularly coal, clearly contributes to sulfur dioxide pollution. On the other
hand, reducing coal consumption has been an important tactic employed
by European nations seeking to reduce their emissions of sulfur (Underdal
and Hanf 2000). Thus, changes in coal consumption are probably endoge-
nous to Protocol ratification. Previous research regarding the effectiveness of
air pollution control policies has estimated the direct and indirect effects of
these policies through recursive causal models (Ringquist 1993). We take a
different approach here, estimating a reduced form emissions model in which
the parameter for the ratification of the Helsinki Protocol reflects both direct
and indirect effects, including indirect effects through coal consumption. We
also estimated models that included a coal consumption variable, and our
conclusions regarding the effectiveness of the Helsinki Protocol remained
National Capacity and Development. There are sufficient theoretical
reasons and empirical evidence to believe that national capacity and develop-
ment factors will affect a nations ability to adequately protect environmental
quality. Indeed, Jnicke (1990, 1992), Sprinz (1998), and others make a
compelling case for a strong relationship between national characteristics
and environmental quality. When attempting to operationalize these multi-
faceted relationships, however, two things become clear. First, there are as
yet no good over-time measures for most of the concepts identified with
a nations institutional capacity (e.g., the cooperative nature of the polit-
ical system, the ability to make long-term commitments) or socio-cultural
capacity (e.g., public opinion data on support for environmental protection).
Second, the most critical measure of national capacity is available for each
nation in each year of our study per capita gross domestic product.
There are several reasons why we believe per capita GDP is the most crit-
ical measure of national capacity and development. First, it is the single best
measure of what Jnicke (1992) calls material capacity. Second, given the
interrelated nature of the Jnickes three dimensions of national capacity, per
capita GDP serves as a reasonable proxy for other capacity dimensions (e.g.,
better-off nations find it easier to have open political systems and profes-
sional political institutions (Jnicke 1992); and public concern for environ-
mental issues varies directly with national standards of living in Europe
(Inglehart 1990)). Third, this particular measure of economic development
has proven to be a powerful predictor of pollution levels (Grossman and
Krueger 1995; Seldon and Song 1994).
For these reasons, the fourth and fifth independent variables in the
models that follow are per capita GDP and per capita GDP squared in
each nation, in each year, measured in thousands of U.S. dollars (at 1991
price levels and purchasing power parity) (Goldman 1997; OECD 1997; UN
ECE 1997). If certain environmental advocates are correct and increased
Chapter 55 Page 159 September 11, 200819:54

Ringquist and Kostadinova Evaluating Environmental Agreements 159

economic development leads to decreased environmental quality, coefficients

for both variables should be positive. On the other hand, if Kuznets curve
advocates are correct (e.g., Grossman 1995), the linear coefficient should
be positive, while the squared term should be negative (i.e., pollution emis-
sions decline at higher levels of economic development). Of course, the
expected signs for these coefficients assume that several of the countries in
our sample have not yet reached the threshold level of economic develop-
ment where further development produces less pollution. Given that there
are no clearly developing nations in our sample, this assumption may be
misplaced. Finally, if material capacity is an important predictor of pollu-
tion levels, at a minimum the sign of the linear term should be negative. The
sign of the coefficient on the squared term is less determinate. If the capacity
effect experiences diminishing marginal returns, the squared term will be
positive, but the entire function will never become positive within the bounds
of the observed data (i.e., increased per capita GDP will never result in higher
pollution levels in our sample). On the other hand, if the marginal returns
to pollution reduction from economic development are linear or increasing,
the coefficient for the squared term will be zero or negative.11
Strategic Behavior. Controlling pollution is often viewed as an imperfect
public good in that because of spillovers (i.e., pollution that crosses national
boundaries), some of the value of these goods accrues to those who do not
provide them. In a substantial contribution to our understanding of interna-
tional behavior in environmental protection, Murdoch et al., (1997, 2003)
illustrate the value of explicitly modeling the strategic behavior of nations
when faced with providing these goods. For example, consider the pollution
control decisions of nations i and j where a substantial amount of js pollution
falls on i. If j aggressively reduces pollution, i will benefit and thus has less
incentive to reduce pollution on its own. In short, in a system characterized
by spillovers, nations will attempt to free-ride on the public good provision
of others. Data on the spillovers of sulfur dioxide emissions were obtained
from the sulfur transport matrix developed by the Cooperative Program for
Monitoring and Evaluation of Long-Range Transmission of Air Pollution
in Europe (EMEP 2003). For each nation (i) in our data set we identified
the three largest sources of sulfur dioxide spillovers (j1-j3). We then created
three variables representing the percentage of all sulfur depositions in nation
i coming from nations j1-j3, weighted by emissions as a percentage of the
1980 baseline in nations j1-j3. If nations behave strategically, larger emis-
sions reductions in nations j1-j3 will generate smaller emission reductions in
nation i, which will be represented by negative parameter estimates on these
three strategic response variables.12
German Reunification. Both before and after reunification, Germany was
the largest source of sulfur dioxide emissions in Europe. Prior to reunification,
Germany was also a leader in pushing for reductions in the emissions of this
pollutant. After reunification, however, sulfur dioxide emissions attributable
to Germany increased markedly. To control for this effect, we include a
binary variable coded 1 for Germany in the years after reunification.
Chapter 55 Page 160 September 11, 200819:54

160 International Environmental Issues and Debates

Model Estimation

Given the near equivalence of the spatial and temporal dimensions of our
panel and the high probability of correlated errors across nations in the
sample, we estimate all models using OLS with panel corrected standard
errors (Beck and Katz 1995). All models also include year-specific intercepts.

Analysis and Discussion

Estimating Policy Effects Using a Base Model

The results reported in the first column of Table 1 look reasonably good.
Most important for our purposes, both the estimated level and trend effect
from the ratification of the Helsinki Protocol are negative and significant.
In particular, the results from the base model suggest that ratifying nations
saw an immediate and permanent emission reduction 8.3% greater than
that experienced by nonratifying nations, and that this difference grew by
roughly 3% per year (.247 12). In addition, almost all of the control
variables have parameter estimates in the hypothesized direction (though
few are statistically significant). We see that increases in per capita GDP
are associated with reduced emissions, though the positive coefficient on
the quadratic term signifies possible diminishing marginal returns associ-
ated with this effect. Second, the negative parameter estimates on all three
strategic response variables are consistent with free-riding behavior in emis-
sions reduction (these variables are jointly significant in all models). Third,
emissions attributable to Germany rose dramatically after reunification.
Only the industrial production variable generates a parameter estimate
contrary to expectations, though this estimate is effectively zero.
To test the adequacy of this model, we estimated the model using only pre-
ratification data. This diagnostic demonstrates that any conclusions that the
Helsinki Protocol has been effective are in error. While there is no difference
in levels, the specification test shows that in the pre-protocol period nations
that would eventually ratify the protocol saw their sulfur dioxide emis-
sions drop 2.5% faster each year than did nations that would not ratify the
protocol. This gure is nearly identical to the post-protocol trend reported
above and illustrates that this post-protocol difference is simply a contin-
uation of pre-protocol trends. Clearly, we need to move beyond this base
model if we are to control for selection bias and obtain consistent estimates
of protocol effects.

Estimating Policy Effects Using a Control Function Model

Recall that in attempting to eliminate the dependence between the error

term and the policy treatment, control function models include factors that
affect the assignment to the treatment or control group but are unrelated to
the outcome variable. For our purposes, this means that a control function
Chapter 55 Page 161 September 11, 200819:54

Ringquist and Kostadinova Evaluating Environmental Agreements 161

model should include variables that affect whether a country will ratify the
Helsinki Protocol, but are unrelated to sulfur dioxide emissions. Murdoch
and Sandler (1997) offer a model emphasizing just these types of control
factors. Specifically, nations that are heavily forested and nations whose SO2
pollution falls largely within their own borders will be more likely to ratify
the protocol since such an action will produce relatively large net benefits.
Similarly, governments in nations scoring high on a common index of polit-
ical freedom will be more likely to ratify the protocol to avoid the political
consequences of being seen as an environmental laggard. Finally, nations
where the costs of reducing sulfur dioxide emissions are high will be less
likely to ratify the protocol. Thus, our control function model includes vari-
ables representing the percent of each nations land area covered by forests,
the percentage of SO2 deposition in each nation originating from domestic
sources, the cost of reducing SO2 pollution, and a dummy variable identi-
fying those nations scoring high on political freedom.
The estimated policy effect from our control function model is nega-
tive and strongly significant, indicating a level change of 10.5% for rati-
fying nations and that SO2 emissions in these countries fell 2.6% faster per
year than SO2 emissions in countries that did not ratify the protocol. Other
variables in the model also have coefficients in the hypothesized direction.
Among the control function variables, our results suggest that freer nations
and nations whose pollution falls largely within their own borders reduced
pollution more, while the costs of pollution control are negatively associated
with emission reductions, as expected.
To test the adequacy of this model, we turn to the pre-protocol
specification tests. Once again, the test demonstrates that our conclusions
regarding the effectiveness of the Helsinki Protocol were in error. The
specification test shows that in the pre-protocol period, on average, nations
that would eventually ratify the protocol saw their sulfur dioxide emis-
sions drop 2.4% faster each year than did nations that would not ratify the
protocol. Again, this gure is nearly identical to the estimated post-protocol
trend effect, suggesting that the post-protocol estimate is simply the contin-
uation of the pre-protocol trend. These results also illustrate the inadequacy
of the control function model in mitigating selection bias.

Estimating Policy Effects Using a Fixed Effects Model

The third column of Table 1 presents the results of our fixed effect approach
to estimating policy effects. The estimated policy effect from this model is
also negative and significant, suggesting that SO2 emissions in countries that
ratified the Helsinki Protocol fell immediately by 9.9% and continued to fall
1.8% faster per year when compared with emissions in countries that did
not ratify the protocol. Other variables in the model have coefficients in the
hypothesized direction, though again few are statistically significant.
The fixed effect model suggests that the Helsinki Protocol has reduced
sulfur dioxide emissions in ratifying nations at a modest rate. When
Chapter 55 Page 162 September 11, 200819:54

162 International Environmental Issues and Debates

Table 1: Regression Coefficients for Percent Change in Sulfur Dioxide Emissions,

19801994, All Nations

Control Fixed Random

Base Function Effects Trend
Independent Variables Model Model Model Model

Helsinki Protocol Signatory 8298 10535 9880 3468

Time Since 0247 0217 0154 0140
Industrial Output 0009 0045 0048 0240

GDP per capita 1717 1398 1742 0565

(GDP per capita)2 0010 0009 0011 0006

Strategic Response 1 0160 0208 0192 0140

Strategic Response 2 0126 0472 0046 0017

Strategic Response 3 0277 0264 0260 0301

Reunification 130163 117271 113027 128056

Forest Cover 0019

Own Source SO2 0505

Political Freedom 31315

Cost of Control 2442

Adjusted R2 083 087 089 094
N 285 285 285 285
Selection Specication Tests
Level Difference 3031 2018 na na

Trend Difference 2448 2422 1162 0686


Note: Figures are unstandardized regression coefficients, standard errors in parentheses. The
random trend specification model is estimated using GLS with controls for first-order autocor-
relation and panel heteroskedasticity.

p < 10 p < 05 p < 01 two-tailed tests.

evaluating this conclusion using pre-protocol data in the fixed effect and
random trend frameworks, we can no longer test for pre-program differences
in levels. These tests, then, focus only upon the more important differences in
trends.13 Once again, this diagnostic test demonstrates that initial conclu-
sions regarding the effectiveness of the Helsinki Protocol were in error.
The specification test shows that in the pre-protocol period, on average,
Chapter 55 Page 163 September 11, 200819:54

Ringquist and Kostadinova Evaluating Environmental Agreements 163

nations that would eventually ratify the protocol saw their sulfur dioxide
emissions drop 1.2% faster each year than did nations that would not ratify
the protocol. This suggests that any post-protocol differences are simply
continuations of pre-protocol trends.

Estimating Policy Effects Using a Random Trend Model

The diagnostic tests above demonstrate that neither the control function
model nor the fixed effects model adequately control for differences between
nations that ratified the Helsinki Protocol and those that did not. In their
study of job training programs, Heckman and Hotz (1989) find both the
control function and fixed effects models inadequate to the task. Their
analysis suggests that under conditions of self-selection into treatment and
control groups, a random trend model may be superior to these first two
alternatives. Moreover, the more recent technical literature emphasizes the
more general utility of random trend models (Wooldridge 2003), and the
specific air pollution histories of European nations suggest that random trend
models are more appropriate (Levy 1993; Murdoch and Sandler 1997). The
results of our random trend model are presented in the fourth column of
Table 1.
The results from the random trend model are substantially different
from previous models. The estimated level effect associated with ratification
of the Helsinki Protocol is still negative, but is no longer statistically
significant. Moreover, the estimated trend effect is now positive and also
not significant. In short, the results from the random trend model indicate
that the Helsinki Protocol has had no independent effect on reducing SO2
emissions in ratifying countries. Once again, other variables in the model
have coefficients in the expected direction including, for the first time, the
expected positive and significant coefficient for industrial output. Turning
to the specification test, we see that the coefficient on the pre-protocol trend
variable for ratifying nations is no longer statistically significant. This illus-
trates that introducing unit specific levels and trends effectively controls
for the bias arising from self-selection on the part of nations and gives us
confidence in the results generated by the random trend model.14

Sensitivity Analyses

Before concluding that the Helsinki Protocol has been ineffective at reducing
sulfur dioxide emissions, we should take care to rule out the most common
threats to the validity of this conclusion. We see three such threats: domain
restrictions on the sample, specification error, and alternative measurement
of the dependent variable. In this section, we address each in turn.
Controlling for Eastern Europe. Several of the former Soviet Bloc nations
of Eastern Europe have ratified the Helsinki Protocol. However, these nations
have very different historical experiences with environmental protection, not
Chapter 55 Page 164 September 11, 200819:54

164 International Environmental Issues and Debates

to mention economies, political institutions, and even national cultures that

are in flux. This legacy of inefficient industrial capacity, a political culture
that placed little value on environmental quality, and even the use of coal
for residential space heating means that the process of SO2 production and
policy effects may be very different in these countries. One could plausibly
argue, therefore, that the effect of protocol ratification should differ between
Eastern and Western Europe.
To test for the effects of the Helsinki Protocol in Western Europe only,
we first eliminated all former Soviet bloc nations from our sample and
then reestimated the fixed effects and random trend models from Table 1.
These results, found in the first and third columns of Table 2, are virtu-
ally identical to those produced from the sample of all nations. The fixed
effects model shows negative and significant level and trend effects from
protocol ratification, but the specification test shows that this model does not
adequately control for selection bias. The statistically consistent treatment
effects estimates from the random trend model show no effect on emission
reductions from ratification of the Helsinki Protocol.
Controlling for National Pollution Control Efforts. A different picture
of the Helsinki Protocol might emerge if we could control for the domestic
capacity and/or domestic investments made by each nation in pollution
control. Toward this end, we included a variable measuring governmental
spending for environmental protection in each nation, from 1980 to 1994,
as a percentage of GDP. Expenditures are generally identified as one of
the most valid indicators of the capacity of environmental policymaking
institutions (Jnicke 1992; Ringquist 1993) and are often used to measure
environmental policy efforts (see Lester 1995). Unfortunately, no widely
available sources contain data for pollution control expenditures for all
years and all countries in our study. In fact, the most complete data set
available contains missing expenditure data for approximately 39% of all
country/years (OECD 1997).15 We dealt with this missing data problem by
employing the multiple imputation algorithm of King et al., (1999). Through
the use of King et al.s (1999) AMELIA missing data software, we created
five sets of imputed spending data. We then used these data to reestimate the
fixed effects and random trend models from Table 1, though even with this
approach we could only estimate the models for 11 nations. The averages
from these estimations are presented in columns 2 and 4 of Table 2, and
these results illustrate that including measures of national effort in pollution
control has no effect on the estimated impact of the Helsinki Protocol.16
Estimating Effects on Overall Sulfur Dioxide Emissions. Finally, our
conclusions regarding the ineffectiveness of the Helsinki Protocol might be
a function of the dependent variable. While protocol targets are stated in
terms of percentage reductions from a 1980 base level, perhaps the effective-
ness of the protocol might be more apparent when looking at total sulfur
dioxide emissions. We reestimated the fixed effects and random trend models
from Tables 1 and 2 employing a dependent variable representing the annual
SO2 emissions for each nation from 1980 to 1994, measured in thousands
Chapter 55 Page 165 September 11, 200819:54

Ringquist and Kostadinova Evaluating Environmental Agreements 165

Table 2: Regression Coefficients for Percent Change in Sulfur Dioxide Emissions,

19801994, Western Europe

Independent Variables Fixed Effects Models Random Trend Models

Helsinki Protocol Signatory 11239 7890 3208 2931

Time Since 0151 0063 0043 0096
Industrial Output 0017 0188 0023 0104

GDP per capita 1828 0397 1688 0586

(GDP per capita)2 0003 0008 0002 0003

Strategic Response 1 0250 0425 0149 0274

Strategic Response 2 0397 0911 0221 0396

Strategic Response 3 0442 0647 0424 0683

Reunification 116623 108483 129097 123753

Pollution Control Spending 3010 2001

Adjusted R2 091 092 094 094
N 225 165 225 165
Selection Specication Tests
Trend Difference 1788 2211 2107 2896


Note: Figures are unstandardized regression coefficients, standard errors in parentheses.

Random trend specification test models estimated using GLS with corrections for heteroskedas-
ticity and first-order autocorrelation.

p < 10 p < 05 p < 01 two-tailed tests.

of tons. All independent variables previously measured as percentages or

changes were also transformed to levels. The results from these models (not
shown) showed no significant differences in patterns of sulfur dioxide emis-
sions between nations that ratified the Helsinki Protocol and those that
did not.


We began our investigation of the effectiveness of the 1985 Helsinki Protocol

for reducing sulfur dioxide emissions with the recognition that cross-national
data comparable over time are needed, that factors other than the Protocol
affect pollution levels, and that the decisions of nations to join the regime
are voluntary. Previous research evaluating the effectiveness of the Helsinki
Protocol all of which found significant Protocol effects failed to control
for one or more of these issues, which inevitably biased their assessments of
Chapter 55 Page 166 September 11, 200819:54

166 International Environmental Issues and Debates

policy effectiveness. After controlling for these elements, we obtain results

suggesting that the 1985 Helsinki Protocol has made no difference in nations
success at reducing sulfur dioxide emissions.
The findings obtained using the more traditional models of policy
impact the base, control function, and fixed effects models indicate
that nations ratifying the Helsinki Protocol reduced sulfur dioxide emis-
sions at rates higher than nonratifiers. Not controlling for the nonrandom
nature of ratification, one would conclude that the Helsinki Protocol was
responsible for greater reductions in air pollution in ratifying nations. The
diagnostic tests, however, demonstrate that differences in emission reduc-
tion rates between ratifying and nonratifying nations were virtually iden-
tical before and after ratification. Thus, the better environmental perfor-
mance of ratifying countries can not stem from ratification of the Helsinki
accord. More probably, larger reductions in sulfur dioxide in nations that
ratified the protocol are attributable to domestic policies, determination,
and commitments that would have occurred even in the absence of the
protocol. In addition, our conclusions are consistent with those of Murdoch
and Sandler (1997) in that they suggest nations that were already substan-
tially reducing their sulfur dioxide emissions may have been more likely to
ratify the protocol. These results demonstrate the importance of control-
ling for nonrandom selection when assessing the effectiveness of voluntary
environmental agreements.
While our results do not give us confidence in the environmental effec-
tiveness of the Helsinki Protocol, any conclusions that international regimes
in general are irrelevant for protecting the global environment are premature.
First, overall evaluations of the success of the Helsinki Protocol do not stand
or fall on the effectiveness of the protocol at reducing sulfur dioxide pollu-
tion. Previous research has shown that CLRTAP in general and the Helsinki
Protocol in particular has had the important effects of elevating awareness
of air pollution problems in Europe, establishing a system of air pollution
monitors generating consistent data, reducing the uncertainty regarding the
causes and extent of sulfur dioxide pollution in Europe, and facilitated the
sharing of scientific information between nations (Levy 1993; Underdal and
Hanf 2000; Wettestad 2002). All of these effects must be considered together
when evaluating the success of the Helsinki Protocol.
Second, while the Helsinki Protocol may not have been successful at
reducing sulfur dioxide emissions, a more differentiated approach to setting
regime goals may prove to be more effective. The Helsinki Protocol imposed
a common pollution reduction requirement on all participants, but this
protocol has been superceded by the Oslo Protocol, agreed to in 1994 and
ratified by a majority of UNECE member nations in 1998. The Oslo Protocol
replaces the across-the-board goal of a 30% reduction in sulfur emissions
with a set of country-specific emission reduction targets based upon crit-
ical loads for sulfur deposition in each signatory nation. Thus, the Oslo
Protocol may be more effective.
Chapter 55 Page 167 September 11, 200819:54

Ringquist and Kostadinova Evaluating Environmental Agreements 167

Third, our null results regarding the effectiveness of the Helsinki Protocol
may stem from unmodeled temporal dynamics. On the one hand, replacing
the industrial capital responsible for most sulfur dioxide pollution (e.g.,
utility and industrial boilers, though not the fuel they use) is a long-term
process. Thus, our evaluation of protocol effects may come too soon. On
the other hand, ratifying nations may have changed their pollution emission
behavior in anticipation of ratification of the protocol, so that some of the
effects of the protocol may pre-date ratification of the protocol itself. In
this case, our evaluation of protocol effects comes too late. Either story is
plausible, but we leave it to future research to assess their validity.17
Finally, the Helsinki Protocol is simply one element of one interna-
tional environmental agreement (CLRTAP), and given the weakness of the
CLRTAP as an institution, an argument could be made that the Helsinki
Protocol was not likely to be particularly effective. In addition, the effec-
tiveness of any environmental regime is determined not only by the charac-
teristics of the institution, but by the malignancy of the problem addressed
by the regime (Underdal 2002). The problem of sulfur dioxide pollution is
highly malignant, so if the Helsinki Protocol has been ineffective at reducing
sulfur dioxide pollution, this may say more about the difficulty in control-
ling this pollutant than it does about the overall effectiveness of international
environmental agreements.
While these caveats all present plausible arguments for the effectiveness of
international environmental agreements, conclusions regarding the overall
success of these agreements must wait until we have evaluations of a repre-
sentative sample of the more than 90 international environmental agree-
ments currently in existence. When conducting these evaluations, researchers
should rigorously evaluate these arguments using techniques that recog-
nize the nonrandom nature of participation in international environmental


Signatories to the 1979 Convention on Long-Range

Transboundary Air Pollution

Included of Helsinki
Country Protocol? Date of Ratification In Data Set?

Austria Yes June 4, 1987 Yes

Belarus Yes September 10, 1986 No
Belgium Yes June 9, 1989 Yes
Bulgaria Yes September 26, 1986 Yes
Canada Yes December 4, 1985 No
Czechoslovakia Yes January 1, 1993 Yes
Denmark Yes April 29, 1986 Yes

Chapter 55 Page 168 September 11, 200819:54

168 International Environmental Issues and Debates


Included of Helsinki
Country Protocol? Date of Ratification In Data Set?

Finland Yes June 24, 1986 Yes

France Yes March 13, 1986 Yes
Germany (West) Yes March 3, 1987 Yes
Greece No na No
Holy See No na No
Hungary Yes September 11, 1986 Yes
Iceland No na No
Ireland No na Yes
Italy Yes February 13, 1986 Yes
Lichtenstein Yes February 13, 1986 No
Luxembourg Yes August 24, 1987 No
Netherlands Yes April 30, 1986 Yes
Norway Yes November 4, 1986 Yes
Poland No na Yes
Portugal No na Yes
Romania No na No
Russia Yes September 10, 1986 No
San Marino No na No
Spain No na Yes
Sweden Yes March 31, 1986 Yes
Switzerland Yes September 21, 1987 Yes
Turkey No na No
Ukraine Yes October 2, 1986 No
United Kingdom No na Yes
United States No na No
Yugoslavia No na No

Source: UN ECE. 1993. The State of Transboundary Air Pollution. New York. P. 19.


1. The foci of these three eras correspond closely to the types of dependent variables most
often investigated with respect to public policies: outputs, outcomes, and impacts (see Keohane,
Haas, and Levy 1993; Mazmanian and Sabatier 1989; Underdal 2002).
2. The problems of data availability, the complexity of environmental outcomes, and the
reality of self-selection are all part of what Young (2001) sees as the greatest obstacle to impact
analysis in this area: the inability to construct realistic counterfactuals.
3. For example, Underdal (1982) argues that when international environmental manage-
ment is established voluntarily through agreement of all significant parties, collective action
will be limited to those measures acceptable to the least committed nation.
4. Jnicke (1992) also discounts other obviously political explanations (e.g., political
parties, elite bargaining, constitutional guarantees of a clean environment) for cross-national
variation in success at protecting environmental quality.
5. While this perspective discounts cross-national policy differences as explanations for
variations in pollution levels across countries, national policy efforts are not completely irrele-
vant to desired changes in environmental quality. Clearly institutional capacity affects a nations
ability to adequately implement environmental programs, and thus nation-specific efforts at
Chapter 55 Page 169 September 11, 200819:54

Ringquist and Kostadinova Evaluating Environmental Agreements 169

implementation do matter, especially if the motivation for these efforts percolates up from the
6. For public policies to have a reasonably good chance of improving the social conditions
they aim to remedy, they must share a certain set of characteristics. First, the enabling legal direc-
tive must identify policy objectives that are clear and consistent. Second, the enabling legal
directive must incorporate a sound causal theory (i.e., it must identify the principal factors
producing the problem of interest and identify the causal mechanisms by which policy objec-
tives can be met). Finally, the policy must give implementing officials sufficient authority over
target groups so that the behavior of these groups may be altered enough to attain the goals
specified in statute. While the Helsinki Protocol scores reasonably well on the first, it relies
upon signatory nations to address the remaining two criteria.
7. Murdoch, Sandler, and Vijverberg (2003) do explicitly address this issue, but their study
does not assess program effectiveness. Issues of selection bias have been addressed in the broader
study of international conflict (see Reed 2000; Signorino 1999).
8. The fixed effects estimator is comparable to the traditional approach to addressing
dichotomous endogenous variables in the cross-sectional case introduced by Heckman (1978).
9. Recent research illustrated the utility of fixed effects and random trend approaches to
estimating treatment effects when compared to alternative methods (Heckman et al., 1998;
Moffitt 1991). In the same spirit, Winship and Morgan (1999) offer a more general nonrandom
assignment model in which all exogenous variables are interacted with time and the treatment
10. Employing a counter variable assumes that the treatment effect will be linear over time.
Clearly, nonlinear functions are also possible if, for example, efforts to reduce emissions face a
steep learning curve or diminishing marginal returns. We estimated the implementation effect of
the protocol using a logged counter, a quadratic term, and a second-degree polynomial. None
performed better than the linear counter reported in the results.
11. Clearly, GDP serves as an indicator for various concepts (e.g., economic development,
political professionalism, public support for environmental protection, etc.). While this use of
GDP is consistent with the literature cited above, it does mean that if we find a significant
relationship between GDP and pollution emissions, we can not say for certain which of these
concepts is responsible for this relationship.
12. Limited by cross sectional data, Murdoch et al. (1997, 2003) modeled this strategic
behavior using a single spatial autoregressive function. With panel data, we can estimate indi-
vidual strategic responses with multiple variables. In preliminary analyses, we found no evidence
of different strategic responses conditional upon the size of emission reductions of the sort
reported by Murdoch and his colleagues.
13. In the pre-protocol period, the adoption or level variable is temporally invariant for all
nations. Such variables are always eliminated when controlling for unit effects.
14. For both the fixed effects and random trend models we employed Hausman (1978) and
Wooldridge (2003) tests for any remaining endogeneity with respect to the protocol variable.
All tests were unable to reject the null hypothesis that the protocol variable was exogenous.
15. The degree of missingness in these data varies by country. For some countries, data are
available for all years. For other countries, the degree of missingness varies from 7% to 87%.
We imputed missing values for each nation that had three or more observed values for pollution
control expenditures between 1980 and 1994.
16. We also measured national efforts in pollution control as per capita spending, with
identical results.
17. We would like to thank the reviewers for suggesting these possibilities.


Barrett, Scott. 1994. Self-enforcing International Environmental Agreements. Oxford

Economic Papers 46: 87894.
Chapter 55 Page 170 September 11, 200819:54

170 International Environmental Issues and Debates

Barrett, Scott. 2003. Environment and Statecraft: The Strategy of Environmental Treaty-
Making. Oxford/New York: Oxford University Press.
Beck, Nathaniel, and Jonathan Katz. 1995. What To Do (and Not To Do) with Time-
Series-Cross-Section Data in Comparative Politics. American Political Science Review 89
(September): 63447.
Berry, William D., and Michael S. Lewis-Beck, eds. 1986. New Tools for Social Scientists:
Advances and Applications in Research Methods. Beverly Hills: Sage Publications.
Chasek, Pamela S. 2001. Earth Negotiations: Analyzing Thirty Years of Environmental Diplo-
macy. New York: United Nations University Press.
Downs, George. 2000. Constructing Effective Environmental Regimes. Annual Review of
Political Science 3 (June): 2542.
Downs, George W., David M. Rocke, and Peter N. Barsoom. 1996. Is the Good News About
Compliance Good News About Cooperation? International Organization 50 (Summer):
EMEP. 2003. Cooperative Program for Monitoring and Evaluation of Long-Range Transmis-
sion of Air Pollution in Europe. http://www.emep.int/reports.
European Marketing Data and Statistics. 1993. London: European Research Consultants.
Goldman, Minton. 1997. Revolution and Change in Central and Eastern Europe: Political,
Economic, and Social Cleavages. New York: M.E. Sharpe.
Grieco, Joseph M. 1993. Anarchy and the Limits of Cooperation: A Realist Critique of the
Newest Liberal Institutionalism. In Neorealism and Neoliberalism: The Contemporary
Debate, ed. David A. Baldwin. New York: Columbia University Press, pp. 11640.
Grossman, Gene M. 1995. Pollution and Growth: What Do We Know? In The Economics of
Sustainable Development, ed. I. Goldin and L.A. Winters. Cambridge: Cambridge Univer-
sity Press, pp. 1950.
Grossman, Gene M., and Alan B. Krueger. 1995. Economic Growth and the Environment.
Quarterly Journal of Economics 112 (May): 35378.
Gunningham, Neil, Robert Kagan, and Dorothy Thornton. 2003. Shades of Green: Business,
Regulation, and Environment. Stanford: Stanford University Press.
Haas, Peter. 1993. Protecting the Baltic and North Seas. In Institutions for the Earth, ed.
Robert Keohane, Peter Haas, and Marc Levy. Cambridge: MIT Press, pp. 13382.
Hausman, Jerry. 1978. Specication Tests in Econometrics. Econometrica 46 (November):
Heckman, James. 1978. Dummy Endogenous Variables in a Simultaneous Equation System.
Econometrica 46 (July): 93159.
Heckman, James, and V. Joseph Hotz. 1989. Choosing Among Alternative Nonexperimental
Methods for Estimating the Impact of Social Programs: The Case of Manpower Training.
Journal of the American Statistical Association 84 (408): 86274.
Heckman, James, Hidehiko Ichimura, Jeffrey Smith, and Petra Todd. 1998. Characterizing
Selection Bias Using Experimental Data. Econometrica 66 (September): 101798.
Heckman, James, and Richard Robb, Jr. 1985. Alternative Methods for Evaluating the Impact
of Interventions. In Longitudinal Analysis of Labor Market Data, ed. James Heckman
and Burton Singer. London: Cambridge University Press, pp. 156245.
Helm, Carsten. 1998. International Cooperation Behind the Veil of Uncertainty. Environ-
mental and Resource Economics 12 (September): 185201.
Helm, Carsten, and Detlef Sprinz. 2000. Measuring the Effectiveness of International Envi-
ronmental Regimes. Journal of Conict Resolution 44 (October): 63052.
Industrial Development Global Report. 1995. New York: UN Industrial Development Orga-
Inglehart, Ronald. 1990. Culture Shift in Advanced Industrial Society. Princeton: Princeton
University Press.
Jnicke, Martin. 1990. State Failure: The Impotence of Politics in Industrial Society. Cambridge:
Polity Press.
Jnicke, Martin. 1992. Conditions for Environmental Policy Success: An International
Comparison. The Environmentalist 12 (l): 4758.
Chapter 55 Page 171 September 11, 200819:54

Ringquist and Kostadinova Evaluating Environmental Agreements 171

Jnicke, Martin, and Helmut Weidner, eds. 1997. National Environmental Policies: A Compar-
ative Study of Capacity Building. Berlin: Springer.
Keohane, Robert, Peter Haas, and Marc Levy. 1993. The Effectiveness of International Envi-
ronmental Institutions. In Institutions for the Earth, ed. Robert Keohane, Peter Haas, and
Marc Levy. Cambridge: MIT Press, pp. 324.
King, Gary, James Honacker, Anne Joseph, and Kenneth Scheve. 1999. Listwise Deletion
is Evil: What To Do About Missing Data in Political Science. PolMeth Working Paper
Archive. http://polmeth.calpoly.edu/working99.alpha.html.
Knoepfel, Peter, and Helmut Weidner. 1982. Implementing Air Quality Control Programs
in Europe: Some Results of a Comparative Study. Policy Studies Journal 11 (Spring):
Lester, James, ed. 1995. Environmental Politics and Policy: Theories and Evidence. Durham:
Duke University Press.
Levy, Marc. 1993. European Acid Rain Policy: The Power of Tote-Board Diplomacy. In
Institutions for the Earth, ed. Robert Keohane, Peter Haas, and Marc Levy. Cambridge:
MIT Press, pp. 75132.
Magat, Wesley, and W. Kip Viscusi. 1990. Effectiveness of EPAs Regulatory Enforcement:
The Case of Industrial Efuent Standards. Journal of Law and Economics 33 (October):
Mazmanian, Daniel, and Paul Sabatier. 1989. Implementation and Public Policy. Lanham:
University Press of America.
McCleary, Richard, and Richard A. Hay, Jr. 1980. Applied Time Series Analysis for the Social
Sciences. Beverly Hills: Sage Publications.
McCormick, John. 1998. Acid Pollution. The International Communitys Continuing
Struggle. Environment 40 (April): 1625.
Mearsheimer, John J. 1995. The False Promise of International Institutions. International
Security 19 (Winter): 549.
Moftt, Robert. 1991. Program Evaluation with Nonexperimental Data. Evaluation Review
15 (June): 291314.
Murdoch, James, Todd Sandler, and Wim Vijverberg. 2003. The Participation Decision versus
the Level of Participation in an Environmental Treaty: A Spatial Probit Analysis. Journal
of Public Economics 87 (2): 33762.
Murdoch, James, and Todd Sandler. 1997. Voluntary Cutbacks and Pretreaty Behavior: The
Helsinki Protocol and Sulfur Emissions. Public Finance Review 25 (March): 13962.
Murdoch, James, Todd Sandler, and Keith Sargent. 1997. A Tale of Two Collectives: Sulfur
and Nitrogen Oxides Emission Reduction in Europe. Economica 64 (May): 281301.
OECD. 1997. OECD Environmental Data. Paris, France: Organization for Economic Co-
operation and Development.
Ostrom, Elinor. 1990. Governing the Commons: The Evolution of Institutions for Collective
Action. Cambridge: Cambridge University Press.
Reed, William. 2000. A Unied Statistical Model of Conict Onset and Escalation. American
Journal of Political Science 44 (January): 8493.
Ringquist, Evan. 1993. Environmental Protection at the State Level: Politics and Progress in
Controlling Pollution. Armonk: M.E.Sharpe.
Seldon, Thomas M., and Daqing Song. 1994. Environmental Quality and Development:
Is There a Kuznets Curve for Air Pollution. Journal of Environmental Economics and
Management 27 (September): 14762.
Signorino, Curtis S. 1999. Strategic Interaction and the Statistical Analysis of International
Conict. American Political Science Review 93 (June): 27998.
Sprinz, Detlef. 1998. Domestic Politics and European Acid Rain Regulation. In The Politics
of International Environmental Management, ed. Arild Underdal. Dordrecht, The Nether-
lands: Kluwer Academic Publishers, pp. 4166.
Sprinz, Detlef, and Tapani Vaahtoranta. 1994. The Interest Based Explanation of International
Environmental Policy. International Organization 48 (Winter): 77105.
Chapter 55 Page 172 September 11, 200819:54

172 International Environmental Issues and Debates

Underdal, Arild. 2002. One Question, Two Answers. In Environmental Regime Effective-
ness: Confronting Theory with Evidence, ed. Edward Miles et al. Cambridge: MIT Press,
pp. 345.
Underdal, Arild. 1982. The Politics of International Fisheries Management: The Case of the
Northwest Atlantic. Oslo: Universitetsforlaget.
Underdal, Arild, and Kenneth Hanf, eds. 2000. International Environmental Agreements and
Domestic Politics: The Case of Acid Rain. Aldershot, England: Ashgate.
UN ECE. 1993. The State of Transboundary Air Pollution. Air Pollution Studies, n. 9. New
York: The United Nations.
UN ECE. 1994. Effects and Control of Long-Range Transboundary Air Pollution. Air Pollution
Studies, n. 10. New York: The United Nations.
UN ECE. 1997. Effects of Long-Range Transboundary Air Pollution. Air Pollution Studies,
n. 13. New York: The United Nations.
UN ECE. 1986. Transboundary Air Pollution: Effects and Control. Air Pollution Studies, n. 3.
New York: The United Nations.
Vig, Norman J., and Regina S. Axelrod, eds. 1999. The Global Environment: Institutions, Law,
and Policy. Washington: CQ Press.
Weiss, Edith Brown, and Harold K. Jacobson. 1998. Engaging Countries: Strengthening
Compliance with International Environmental Accords. Cambridge: MIT Press.
Wettestad, Jrgen. 2002. The Convention on Long-Range Transboundary Air Pollution.
In Environmental Regime Effectiveness: Confronting Theory with Evidence, ed. Edward
Miles et al., Cambridge: MIT Press, pp. 197221.
White, Lawrence. 1982. U.S. Mobile Source Emissions Regulation: The Problems of Imple-
mentation. Policy Studies Journal 11 (Spring): 7787.
Winship, Christopher, and Stephen Morgan. 1999. The Estimation of Causal Effects from
Observational Data. Annual Review of Sociology 25 (August): 659707.
Wooldridge, Jeffrey. 2003. Econometric Analysis of Cross Section and Panel Data. Cambridge:
MIT Press.
Young, Oran. 1994. International Governance: Protecting the Environment in a Stateless
Society. Ithaca: Cornell University Press.
Young, Oran. 2001. Inferences and Indices: Evaluating the Effectiveness of International Envi-
ronmental Regimes. Global Environmental Politics 1 (February): 99121.
Young, Oran R., ed. 1999. The Effectiveness of International Environmental Regimes: Causal
Connections and Behavioral Mechanisms. Cambridge: MIT Press.
Zrn, Michael. 1998. The Rise of International Environmental PoliticsA Review of Current
Research. World Politics 50 (July): 61749.
Chapter 56 Page 173 September 11, 200820:0

Toward a World Environment Organization:
Reflections upon a Vital Debate
Steve Charnovitz

1. Introduction

ver a dozen years have elapsed since the run-up to the United

O Nations Conference on Environment and Development (held in Rio

de Janeiro) when the idea of a World Environment Organization
(WEO) began to receive serious attention. Although a spirited and illumi-
nating debate has transpired, no real progress toward a WEO has occurred.
Today, our planet still lacks effective global environmental governance.
During the same period, the world trading system succeeded in rational-
izing and strengthening its institutional foundation. At the time of the 1992
Rio Conference, the General Agreement on Tariffs and Trade was in the
sixth year of the multilateral Uruguay Round of negotiations, and prospects
for a successful conclusion were in doubt. Some plans for establishing a
new international organization were on the drawing board, but whether
governments would agree to such a treaty and be able to ratify it in national
parliaments remained a big question. The idea that the negotiations would
lead to a powerful, rule-based organization with a judicial system that can
expound law, and that membership in the new entity would increase from
110 to 147 countries after nine years, was hardly the consensus scenario at
that time. Clearly, governance for trade has enjoyed a much better recent
run than governance for environment.
Today, the World Trade Organization (WTO) is in the midst of another
negotiating round originally set to conclude in 2005. Efforts are underway
to strengthen and broaden trade rules. Several environmental issues are on
the negotiating table in Geneva, including reducing trade barriers to environ-
mental goods and services and controlling fishery subsidies. Using compara-
tive institutional analysis, commentators have asked why the WTO appears
to be so effective at accomplishing its mission, and at expanding its mission,
while the environment regime seems relatively less capable. One answer is
that the trade regime is centred in an effective international organization

Source: Frank Biermann and Steffen Bauer (eds), A World Environment Organization:
Solution or Threat for Effective International Environmental Governance? (Burlington, VT:
Ashgate, 2005), pp. 87115.
Chapter 56 Page 174 September 11, 200820:0

174 International Environmental Issues and Debates

while the environment regime is not (German Advisory Council on Global

Change 2001, 17677; Ostry 2001, 29093).
Since the early 1990s, many analysts have called for correcting this orga-
nizational dysfunction in the environment regime. The opening salvo in the
contemporary debate1 was the perceptive article by Sir Geoffrey Palmer,
the former Prime Minister of New Zealand, who examined New Ways to
Make International Environmental Law (Palmer 1992). Sir Geoffrey called
for the establishment of an International Environment Organization within
the UN system. In 1994, Daniel Esty began an intellectual campaign for a
Global Environment Organization (GEO) that would develop a comprehen-
sive international response to environmental challenges (Esty 1994, 230).2
Several other analysts have also advocated a new environmental organi-
zation, such as Ford Runge (Runge 1994, 100; 2001), Frank Biermann
(Biermann 2000; 2001), the Zedillo Commission (High-Level Panel on
financing for Development 2001, 26), Gus Speth (Speth 2002b, 2223), the
Shadow G-8 group (Shadow G-8 2003, 29), and the team of John Whalley
and Ben Zissimos (Whalley and Zissimos 2001; 2002). Some thoughtful
analyses of the various WEO proposals have now appeared (Hierlmeier
2002; Lodefalk and Whalley 2002; Marshall 2002; Gaines 2003; Haas 2004;
Oberthr and Gehring, this volume).
Two responses have emerged concerning these ideas for more cohesive
environmental governance. Several environmental analysts have criticized
these proposals for being over-ambitious, centralist, pro-North, unsophisti-
cated, or unnecessary (for example, see Juma 2000; Najam 2002 and this
volume). Often, however, the WEO proposals have been overlooked. For
example, a major new study of environmental governance omits any atten-
tion to the debate regarding the WEO (UNDP et al. 2003).
The concept of a WEO was only briefly mentioned in the preparatory
sessions for the 2002 World Summit on Sustainable Development (WSSD),
and was not discussed at all at that Johannesburg Summit. In 1997, at the
Special Session of the UN General Assembly, four governments had proposed
consideration of a global environmental umbrella organization of the UN,
and there were expectations that such ideas would be advanced over the
following five years. Instead, the UN General Assembly took a side-step
by convening an annual Global Ministerial Environmental Forum (GMEF),
which was set up merely as a periodic Forum rather than a continuing
organization. When the GMEF held its first meeting, at Malm in 2000,
the Forum called for the forthcoming 2002 conference (later termed the
WSSD) to review the requirements for a greatly strengthened institutional
structure for international environmental governance based on an assess-
ment of future needs for an institutional architecture that has the capacity
to effectively address wide-ranging environmental threats in a globalizing
world (Malm Declaration 2000, para. 24). At the same time, the UN
Environment Programme (UNEP) launched a series of meetings on inter-
national environmental governance. This series concluded in Cartagena in
early 2002 with recommendations for a stronger funding base for UNEP,
Chapter 56 Page 175 September 11, 200820:0

Charnovitz Toward a World Environment Organization 175

and for the GMEF and UNEP Governing Council to be utilized more effec-
tively.3 Not surprisingly, the UNEP-sponsored meetings did not recommend
that UNEP be replaced by a WEO. By the time of the Johannesburg Summit,
the construction of a WEO was no longer being actively considered.4
Notwithstanding this reticence at the WSSD about environmental gover-
nance, the delegates were eager to pontificate on trade governance. For
example, the Johannesburg Plan of Implementation (2002, paras. 47, 90, 96)
encourages efforts by international financial and trade institutions to have
more open and transparent decision-making processes, urges WTO members
to facilitate the accession to the WTO by developing countries, and calls for
action at all levels to enhance trade infrastructure and strengthen institutions.
To be sure, the amount of text in the Plan on Implementation regarding trade
is much less than the amount regarding environment. Still, it seems note-
worthy that the governments were willing to delve into the management of
the trading system while not paying much attention to needed improvements
in environmental governance.
Two possible explanations exist for why the results from the WSSD were
so meagre on governance, and for that matter on environmental stewardship
in general. The first is that the environment regime is running smoothly and
the other is that it is so poorly designed that it cannot reform itself incremen-
tally. Explanation No. 1 is obviously wrong: Our planet faces significant
environmental challenges (Turner 2000; Worldwatch 2003; Speth 2004),
and I am not aware of any serious analyst who claims that current governance
of the Earths ecolonomy is sufficient. The second explanation is that the
environment regime has a vested interest in maintaining its compartmental-
ization, and will strongly resist any consolidation. If this second explanation
is right, then the prospects for reform are dire indeed.
On the assumption that environmental governance needs fixing, what is to
be done? Proponents of a WEO should reflect on why so little has been accom-
plished toward that goal. One problem is that the advocacy for a WEO has not
been convincing. Anyone who examines the various proposals would see an air
of first positing reorganization and then searching for a mission. Proponents
of a WEO will need to renew efforts to make a more cogent case for reform.
The frustration with the stalled debate was undoubtedly a reason why
Esty has shifted to advocating a more gradualist Global Environmental
Mechanism that would link together existing institutions and add new struc-
tures when warranted to carry out core functions of environmental gover-
nance. In a recent paper, Esty and his colleague Maria H. Ivanova point to the
need for better data collection, compliance monitoring, scientific assessment,
bargaining, rule-making, civil society participation, financing, technology
transfer, dispute settlement, and implementation coordination mechanisms
(Esty and Ivanova 2002).
The Johannesburg Summit presented an opportunity to address these
gaps within existing governance structures. Unfortunately, the governments
did not do so, and also failed to consider the organizational prerequisites for
environmental policy.
Chapter 56 Page 176 September 11, 200820:0

176 International Environmental Issues and Debates

The newest model for a WEO is the proposal from Whalley and Zissimos
for an organization to help governments and private actors do environmental
bargaining. This initiative, funded by the MacArthur Foundation, is creative
and useful. But at best it could deliver only a partial solution to current
governance problems. Building on Ronald Coases classic analysis of how
polluters and victims could bargain to achieve joint gains so long as the
liability rule is clear and transaction costs are low, Whalley and Zissimos
extend the argument beyond pollution into the use of natural resources. They
suggest that a global mechanism could lower transaction costs and facilitate
negotiated exchanges, and perhaps help to clarify property rights. The ulti-
mate goal is to achieve full internalization of cross-border externalities so
that those who undertake economic activities bear their full environmental
Whalley and Zissimos are right that considerable scope exists for interna-
tional bargaining beyond the quantity of deal-making occurring now. This
gap certainly suggests the need for better mechanisms to assure contrac-
tual performance in international environmental deals (see Stone 1993,
3942). But Whalley and Zissimos go further than that in boldly arguing
that other WEO proposals are not focused on central and substantive envi-
ronmental policy problems, and that their own plan could achieve more than
the current network of issue-specific environmental treaties (Whalley and
Zissimos 2002, 16466).
In my view, the Whalley and Zissimos proposal to enhance markets is
worth trying, but it is hardly a sufficient solution to the challenges of environ-
mental protection. The subtext of their proposal is that the rich countries will
pay poor countries for guaranteed outcomes such as preserving a rainforest.
Although a higher volume of exchanges might be possible, one wonders
how deep the pockets are in rich countries for such monetary deals. While
no one can deny the potential benefit of proper pricing for environmental
resources, Whalley and Zissimos do not offer any reason to believe that
bargaining will occur on a sufficient scale to achieve a significant amount
of cost internalization. In one revealing passage, the authors note that their
proposal focuses on cross-border externalities since within-country exter-
nalities can in principle be dealt with by solely domestic initiatives (Whalley
and Zissimos 2002, 166). Yet their study provides no evidence that robust
domestic markets for such bargaining currently exist and are achieving
significant cost-internalization. If such bargaining does not actually occur
domestically, why imagine that it will occur across borders? Perhaps what
Whalley and Zissimos meant by domestic initiatives is government-imposed
taxes and regulations. Yet if such a regulatory strategy is needed within
each country, then why would one pursue an entirely different strategy for
transborder issues as a substitute for regulation through treaties and, when
justified, extraterritorial application of law? The next stage of the Whalley
and Zissimos project should consider these points.
The purpose of this article is to restate the case for a WEO. The remaining
discussion has three parts: Section 2 will seek to explain why a WEO is
Chapter 56 Page 177 September 11, 200820:0

Charnovitz Toward a World Environment Organization 177

needed by examining, in turn, the W, the E, and the O. Section 3 will suggest
that the paradigm for the WEO should be competition, as well as coopera-
tion, the goal stressed in the pro-WEO literature. In both of these parts, the
article will take note of the WTO, and point out where it is a good model for
a WEO or a poor model. The article ends with a short conclusion (section 4).

2. Why a W-E-O Is Needed

Because ecosystems overlap political units, it stands to reason that inter-

national and/or transgovernmental organizations will be needed to manage
human interaction with the environment. States alone may be able to
perform this function with respect to environmental problems residing
within national borders (e.g., noise pollution), but most of the serious
environmental problems today are transborder and/or global. For those,
solutions will require collective action. Truly effective international envi-
ronmental institutions can improve the quality of the global environment
(Keohane, Haas, and Levy 1993, 7). Institutions help by increasing govern-
mental concern, by building capacity, and by enhancing the ability to make
and keep agreements (Levy, Keohane, and Haas 1993, 398, 424).
Externalities occur when a producer or consumer does not take into
account the adverse effects that it imposes on others. Such market failure
is the core problem of environmental policy. Responses to this problem
include regulation, clarification of property rights, and facilitating bargains.
As Andr Dua and Dan Esty have pointed out, when externalities traverse
national borders, they can be viewed as super externalities because of the
additional hurdle they present of securing cooperation among sovereigns
(Dua and Esty 1997, 59).
Governments began responding to transborder environmental problems
in the 19th century through treaties, and, during the 20th century, drafted
hundreds of new treaties and established scores of international organi-
zations with responsibility for environment, natural resources, and global
public goods. The record shows that governments have been willing to
initiate cooperation on specific problems by establishing conventions and
related institutions. Yet few governments have shown a willingness to meld
these institutions into a holistic entity.
Although UNEP, established in late 1972, has helped to promote interna-
tional environmental law (Tolba 1998; McNeill 2000, 350), the organization
has chronically underperformed. The problem is not quality of leadership.
Over 30 years, UNEPs executive leadership has been better than average
for international organizations, and its current Executive Director, Klaus
Tpfer, is quite capable. The problems of UNEP stem from its low status
within the UN bureaucracy, its disadvantageous and dangerous location in
Nairobi, its inadequate and insecure funding, and its detachment from many
of the multilateral environmental agreements (e.g., climate change).
The trade and environment debate of the 1990s stimulated many out-
siders to examine UNEP, and that Programme looked feeble organizationally
Chapter 56 Page 178 September 11, 200820:0

178 International Environmental Issues and Debates

in comparison to the trading system. UNEPs sorry state, including its weak
presence and staffing in Geneva, triggered recommendations for a world
organization for the environment that could operate in equipoise with the
WTO. For example, Supachai Panitchpakdi, now the Director-General of
the WTO, once stated that the problem is that there is nobody of the
same stature to deal with the WTO because there is no World Environment
Organization (Supachai 2001, 443). The institutional strength of the WTO
reflects an acknowledgement by governments that economic interdepen-
dence is a fact, and that nations will be better off with a robust organization
that can help manage that interdependence. Yet ironically, even though the
global environment is more of an integrated system than the global economy
(Speth 2002a, 13), governments have not drawn a parallel conclusion about
the value of a WEO.

Why a World Organization?

A common complaint about creating a WEO is that such an organization

would be too powerful and intrusive. In that respect, the WTO analogy
has hurt the pro-WEO cause because of the political baggage the WTO
now carries. The developing countries, as a generalization, view the WTOs
rule-based approach as being too coercive to them, and too insensitive to
national development plans. Some groups in civic society view the WTO
as the champion of harmful globalization from above because it promotes
economic integration and elevates decision-making to a level beyond the
influence of the public. Another complaint about the WTO is that even
though each governmental member ostensibly has the same influence, in
practice richer countries have a greater say. Thus, advocates of a WEO now
have the burden of explaining how a WEO will avoid these reputed problems
of the WTO.
To the extent that a WEO would be a centralized, top-down institution,
that seems to rub against the grain of subsidiarity (Newell 2002, 669), a
term from European law positing that authority to make decisions should not
be raised to a higher level (the Community) when a lower level (a Member
State) would be adequate (see Bermann 1994; Vergs 2002).
The term subsidiarity originated in Catholic philosophy. In his 1931
Encyclical Reconstruction of the Social Order, Pope Pius XI explained that
Just as it is gravely wrong to take from individuals what they can accomplish
by their own initiative and industry and give it to the community, so also it
is a grave evil and disturbance of right order to assign to a greater and higher
association what lesser and subordinate organizations can do (Pius XI 1931,
para. 79). The Pope termed this principle the subsidiary function, and called
on those in power to pursue a graduated order. Most of the discussion about
subsidiarity in the Encyclical focuses on the State vis--vis the individual.
The Pope did not discuss this principle with respect to governments in the
League of Nations (or, for that matter, with respect to decision-making in
Chapter 56 Page 179 September 11, 200820:0

Charnovitz Toward a World Environment Organization 179

the Catholic Church). Thus, one cannot assume that the doctrine will always
apply to the relationship between an intergovernmental organization and its
member States.
States may have valid reasons to delegate decision-making upward
to international entities. Doing so may enhance the dignity of the indi-
vidual even though decision-making may be slightly more remote. Although
subsidiarity is sometimes characterized as a principle of non-interference,
this shorthand misses the duality in the Popes discussion which is as much
concerned with helping smaller units as it is with not interfering with them.
As a scholar of subsidiarity explains, each larger grouping is understood to
serve the smaller, and all in the end are understood to serve individual dignity
(Carozza 2003, 43).
In any event, a commitment to subsidiarity does not present a true stum-
bling block for WEO advocates, who point out that a WEO is only needed
for those problems that are not being solved at a lower level. Numerous
global problems exist, such as climate change, ozone depletion, ocean pollu-
tion, and fisheries depletion, and so there would be plenty of issues for which
a WEO might be the right level to assign a lead competence. Ironically, none
of those particular issues is now overseen by UNEP; all have been assigned
to other organizations or treaty entities. Many WEO advocates would also
assign it responsibility for helping governments address transboundary envi-
ronmental problems (such as hazardous waste), and the maintenance of
public goods (such as biodiversity). The subsidiarity rationale for allowing
a WEO to share jurisdiction over such issues is that the lower organizations
(i.e., the national governments) consent because they need coordination from
The case for a world-level response is probably the weakest for the
common challenges that all countries face, such as clean water, waste dis-
posal, etc. Yet it was precisely that genre of issues (rather than, say, climate
change), that were the centrepiece of the WSSD in 2002. To my knowl-
edge, no government raised an objection in Johannesburg to devoting so
much time to issues that are inherently local. That focus at Johannesburg
was similar to the orientation of the UN Millennium Development Goals.
For example, one of the Goals is to reduce by half the proportion of people
without sustainable access to safe drinking water.
Any international environmental entity is accountable with regard both to
its outcomes and its procedures. Some analysts fear that a global organization
cannot possibly appreciate the subtleties of environmental policy appro-
priate for separate communities around the world, and thus a WEOs norm-
generation and other activities could wreak unintended harms. Although
the establishment of the WTO enjoyed the support of international busi-
ness groups, the establishment of a WEO has not drawn symmetric support
from international environmental groups. Surely, one reason why is that
many nongovernmental organizations (NGOs) fear that a WEO might make
it harder for citizens and associations to influence environmental policy.
Chapter 56 Page 180 September 11, 200820:0

180 International Environmental Issues and Debates

The doubts about accountability are sometimes expressed as concerns about

a democratic deficit or a gap in legitimacy.
Legitimacy has many facets. The least controversial claim is that an inter-
national organization should act legitimately with respect to its member
governments. The WTO has a mixed record on that facet of legitimacy. On
the one hand, it has rule-based legitimacy in that all actions putatively have to
gain the consensus of all member governments. On the other hand, the tradi-
tional WTO practice of handling controversies by having key governments
convene privately in a green room has not yet ceased (Global Accountability
Report 2003, 15). In the environment regime, there has been an effort over
the past few years to reconstitute the UNEP Governing Council to include
all governments based on the rationale that universal participation is more
legitimate than representative participation (see Johannesburg Plan of Imple-
mentation 2002, para. 140d). That rationale is questionable, however, if a
larger assembly would make decision-making harder.
A more contested claim is that the constituents of an international organi-
zation include the public in each country, if not a global public.5 The counter-
argument is that international organizations are sufficiently accountable
to each individual in a transitive fashion through his or her own govern-
ment. Whether or not that is true as a matter of democratic theory, many
individuals and groups (e.g., anti-globalization protestors) believe that the
legitimacy of an international agency is undermined by the lack of a direct
connection to an electorate. NGOs have also criticized the insularity and
secretiveness of some international organizations, such as the WTO.
When the problems considered are complex and solutions emerge slowly
the common predicament on global environmental issuesthe value of trans-
parency and regular public input becomes obvious as a way of securing better
information and ideas. The WTO achieves some transparency, but scores
badly on eliciting public input. Thus, while a WEO might be able to copy
the WTO approach for how governments participate, a WEO would need
to be far more open to civic society and business than the WTO is. This
would be in line with the Rio Declaration which states that environmental
issues are best handled with the participation of all concerned citizens, at the
relevant level (Rio Declaration 1992, Principle 10).
Given the long-time practice of nongovernmental participation in envi-
ronmental governance, there would seem to be little point in establishing
a WEO if based on the common intergovernmental model in which NGOs
participate as a sideshow. Indeed, the prospects for a broad-based WEO
could enhance public support for adopting a new organization. If a WEO
is to be created, it should reflect the learning from the path of the Inter-
national Labour Organization (ILO), a body created in 1919. In the ILO,
workers and employers participate equally with government representatives,
a feature known as tripartism.
ILO-style tripartism, however, is no longer fully adequate as a template
for gaining NGO participation. An effective WEO would have to provide
space for government representatives to work with a multiplex of
Chapter 56 Page 181 September 11, 200820:0

Charnovitz Toward a World Environment Organization 181

stakeholders including environmental NGOs, human rights groups, busi-

nesses, scientists, religious leaders, mayors of cities, and many other stake-
holders. Such inclusiveness would be in line with the doctrine of subsidiarity,
which states that social activity ought of its very nature to furnish help to
the members of the body social, and never destroy and absorb them (Pius
XI 1931, para. 79).
Should governments be unwilling to extend participation in a WEO to
private social and economic actors, then that would tip the scale in favour
of those who demur that the cost of a massive reorganization into a WEO
would be too high relative to the expected gains. Why go to the trouble of
setting up a new international organization if it is to be composed merely of
government officials and bureaucrats? One does not need an international
organization for governments to cooperate; they can do so bilaterally, or
through emerging transgovernmental networks (see Raustiala 2002). Yet on
many international problems, governments will be too cautious and nation-
alistic to reach integrative solutions without the catalyst of nongovernmental
Establishing a participatory WEO would be a challenge, and no recipe
for it exists. Perhaps the most difficult task is to find a way to combine
broad participation with a decision-making capacity for the organization.6
The sorry experience with the UN Commission on Sustainable Development
(CSD) (see Elliott, this volume) stands as a stark reminder of the pointless-
ness of fostering broad participation detached from any decision-making
Sometimes, analysts argue that the failures in international environment
policy are not caused by poor organization, but rather are caused by lack of
political will (see Najam 2002, 8). Yet that diagnosis seems to miss the point
that well-designed international institutions can help generate political will
by constructing new social norms.

Environmental Protection and Sustainable Development

Sustainable development is a useful concept (Tarlock 2001; Holliday,

Schmidheiny and Watts 2002). It marries two important insights: that
economic development should be ecologically viable and that environmental
protection does not preclude development. Sustainable development also has
value in positing an answer to the trade-off between the welfare of the people
today and the welfare of the people of the future.7 Thus, the goal of sustain-
ability should inform the work of all international agencies (Dowdeswell
and Charnovitz 1997, 101).
Yet it is one thing for sustainable development to be an inspiration,
and quite another for that vague term to be the organizing principle for
governmental action. In the years since the Rio Conference of 1992, during
which sustainable development has reached mantra status,8 we have not
seen many examples of how the concept has made much of a policy difference
(Esty 2001). It may be unfair to point to the CSD as an example of such
Chapter 56 Page 182 September 11, 200820:0

182 International Environmental Issues and Debates

failure because the CSD was not set up to do anything except be a talk shop.
Yet that is exactly the point: when the international community glorifies a
concept as expansive and ambiguous as sustainable development, perhaps
the best place for it is a talk shop.
One unfortunate manifestation of the sustainable development concept
is that it is elbowing out environmental protection at the international level.
Recall that in 1972, the world community held a UN Conference on the
Human Environment (in Stockholm), and, in 1992, held the UN Conference
on Environment and Development (MacDonald 2003, 16668). In 2002,
however, there was a decennial conference called the World Summit on
Sustainable Development. Thus, over the years, we have seen the premier
global environmental event transmogrify from a conference focused on the
environment, to a conference about environment and development, and then
to a Summit where the term environment has been banished from the
events title. Ironically, the Johannesburg Conference was held just a few
months after another major development conference, the UN Conference on
financing for Development (in Monterrey). That propinquity itself provided
a reason to rehabilitate the Johannesburg Summit back to environment, but
the governments did not even consider that. In recent years, UNEP, too,
has shifted its attention more toward development, and has adopted a new
motto, Environment for Development (UNEP 2002,4).
In stating that environmental problems should be dealt with directly, I do
not mean to sound anti-development or anti-poverty reduction. Certainly,
the plight of the poorest countries may be the central economic and moral
issue of our time. In my view, the United Nations and other international
organizations, such as the WTO, should be doing a lot more to alleviate
poverty. But that does not justify usurping an environmental agenda with a
poverty reduction agenda. Both agendas are important and distinguishable
from one another. Trying to meld them ends up short-changing both. To
be sure, defenders of the Johannesburg Summit would argue that poverty
reduction can be a potent environmental strategy, and I agree. Yet that is
hardly a reason to refrain from holding a Summit to zero in on environmental
If the Summit had succeeded in delivering significant environmental
benefits, then the continued reliance of a paradigm of sustainable devel-
opment might be more supportable. The absence of such a positive result
demonstrates the inefficacy of organizing global meetings around the
sustainable development objective. Space does not permit a full auditing of
the meagre output from Johannesburg, so I call attention to the document
Key Outcomes of the Summit, located on the UN web-site.9 The eight key
outcomes listed can be summarized as: (1) reaffirming sustainable devel-
opment, (2) broadening sustainable development to include poverty link-
ages, (3) issuing concrete commitments and targets for action, (4) giving
attention to energy and sanitation issues, (5) supporting a world solidarity
fund for eradication of poverty, (6) focusing on the development needs of
Africa, (7) taking into account the views of major groups, and (8) boosting
Chapter 56 Page 183 September 11, 200820:0

Charnovitz Toward a World Environment Organization 183

partnerships with the private sector. Of those, outcomes 2, 4, 5, and 6 are

not really environmental. Outcome 2 and 7 are about process, and outcome
1 is regurgitive.
Thus, if there was any environmental policy advance in the Summit, it
has to be in outcome number 3, the concrete commitments and targets. Yet
according to the accompanying fact sheet regarding those commitments,
many of the targets hew to development rather than environment (e.g.,
poverty eradication, sanitation, infant mortality, and energy), and some of
the ones that are environment (e.g., safe drinking water) are just restate-
ments of goals previously established by the United Nations. Boiling all this
down, there are just a few new environmental targets for chemicals (2020),
water efficiency (2005), oceans (2010), fish stocks (2015), and biodiver-
sity (2010) yet even there, no specific goal is backed up with an action
plan likely to achieve the goal. Given the many months of planning for
Johannesburg and the numerous PrepComs and planning sessions held, the
wispy output of the Summit did not help environmental policy escape the
doldrums of the past few years (see Gutman 2003).10
Were sustainable development a viable programmatic objective, one
could expect to see in the 12 years since the Rio Conferencenumerous
governments setting up ministries of sustainable development. While some
do exist (e.g., France), that is hardly a common feature of national admin-
istration. Instead, governments have continued to maintain environmental
ministries that are separate from trade ministries and energy ministries.
Recognizing the separateness of environment and sustainable development
objectives is certainly consistent with the recommendations of the Brundt-
land Commission which stated in Our Common Future that Environ-
mental protection and sustainable development must be an integral part
of the mandates of all agencies of governments, of international organiza-
tions, and of major private-sector institutions (World Commission on Envi-
ronment and Development 1987, 312 emphasis added). In my view, most
governments made the right choice in avoiding conglomeration through a
sustainable development ministry because such a ministry would probably
be ineffective.
Progress in national government and international governance over the
years has come through specialization (Gardner 1974, 558). This functional
approach is not controversial in most international bodies. One expects the
ILO to bring together labour ministers to focus on workers and employ-
ment One expects the WTO to bring together trade ministers to focus on
trade. One expects the World Health Organization (WHO) to bring together
health officials to focus on disease and public health (see Stein 2003, A15).
By contrast, for the environment, when the United Nations holds a world
conference, it is apparently not politically correct to bring together govern-
ments to focus on ecological threats. This skittishness has gotten worse in
recent years, and is a main reason for the miasma in international environ-
mental governance. To be sure, the meetings of the UNEP Governing Council
and the GMEF do convene environment ministers.11 Such a meeting was
Chapter 56 Page 184 September 11, 200820:0

184 International Environmental Issues and Debates

held in February 2003 in Nairobi, but did not accomplish much (see Nanda
2003). Another meeting was held in March 2004 in Jeju, but whatever result
it generated has not yet been released to the UNEP website.12
Despite the initially high expectations, the WSSD brought the worst of
both worlds for the environmental regime. Although widely perceived as
the once-in-a-decade opportunity for national leaders to address the envi-
ronment, the Summit ended up being more about poverty (and then only
rhetorically). With that political space now having been used up, the envi-
ronment will probably not gain another Summit this decade. Any attempt to
upgrade environmental governance will have to contend with the albatross
of sustainable development, and demands by developing countries that any
new organization have sustainable development as the core objective. So
what is wrong with that: why not take sustainable development seriously
and organize internationally around that overarching concept? The reason
why not is that governments are loathe to let an organization for sustainable
development interfere with other functional international organizations.
This attitude was apparent at Johannesburg. Many developing countries,
quite logically, took the position that a conference on sustainable develop-
ment had the competence and responsibility to address international trade
policy. These governments pushed for new commitments on trade liberal-
ization, but ran into the buzzsaw of European and US delegates who did not
want even to discuss going beyond what had been agreed to at the WTO one
year earlier (Gray 2003, 25865). As a result, although there is a great deal
of verbiage in the Johannesburg Plan of Implementation about trade, all of
it merely rephrases prior agreements that have been reached at the WTO.
The same concern about turf exists in the World Bank, the International
Monetary Fund, and other UN organizations. All would resist having a
Sustainable Development Summit intrude on their policy mandates. Despite
the ostensible allegiance to sustainable development, governments are not
going to use a Johannesburg Summit or analogous future event to supervene
the competence of other international agencies.
Yet if a Summit about sustainable development cannot negotiate on
trade, or development funding, or intellectual property, then it is a hollow
Summit. The same point holds for a prospective World Organization for
Environment and Development (for one proposal, see Simonis and Brhl
2002, 12223). The necessary comprehensiveness would make it impossible
to create such an Organization with a meaningful mandate.
In conclusion, if a WEO is to be set up, its mission should be to address
the top environmental risks facing the planet. A WEO could also address the
growing inconsistencies between environmental conventions, a problem that
now lacks an organizational solution (see Wolfrum and Matz 2003).

Organization and Its Discontents

This leads to the final consideration: Is the O in WEO realistic? Critics of

such an Organization have made two salient points. One is that incremental
Chapter 56 Page 185 September 11, 200820:0

Charnovitz Toward a World Environment Organization 185

improvements in current governance will have to be adequate because

nothing else is feasible. The other is that the environment regime is too
complex for one WEO.
The notion that the current environment regime is the best that humans
can accomplish would be a preposterous claim. At present, the regime is one
of disorganization with hundreds of agencies and treaties operating unlinked
to each other. Not once have governments taken the time to design an ideal
management structure. Instead, whenever a new environmental problem
arose, a new entity was opportunistically added. Rarely have entities been
dismantled, even when they are so obviously ineffective, such as the CSD.
Another disappointing experience is with the G-7/G-8 Environment
Ministers, who have been meeting annually for nine years without much
to show for it.13 At the most recent meeting in April 2003, the Minis-
ters announced support for increased environmental coordination at the
international level through broad policy guidance and advice of the UNEP
Governing Council/GMEF and full respect for the independent legal status
and governance structure of other entities    (G8 Environment Ministers
Communiqu 2003). That G-8 meeting heard a proposal from France to
consider establishing a new UN Organization for Environment, but there
was apparently minimal support for the idea (Speer 2003).
The organizational failure in environmental governance is especially
disturbing when one compares it to the more rationally-organized trade
regime. An organigram of the WTO shows a political ministerial body, a
hierarchy of policy committees, a dispute settlement system, and a group
that reviews national government policies.14 To my knowledge, no organi-
zation chart for the complete environment regime(s) even exists. If it did, it
would be a mishmash, with numerous boxes unconnected to each others.15
The costs of such organizational anomie are high.
Whether governments and stakeholders are saddled forever with this
disorganization is a matter of conjecture. While I agree that establishing a
WEO would be difficult politically, I cannot accept that it is the organiza-
tion of the impossible to use Konrad von Moltkes memorable phrase (see
von Moltke 2001). Good environmental policy is no longer just a preoccupa-
tion of the rich countries; it is equally sought by new environmental leaders
in developing countries too (French 2003).
Certainly, a WEO will not be set up unless there is a large group of govern-
ments and stakeholders who demand it. Unfortunately, we are far away
from that. The major multilateral environmental agreements have spawned
distinct epistemic networks that seem to have a vested interest in maintaining
a highly compartmentalized system. The uneasiness among environmental-
ists about a WEO will have to be reversed before any progress can be made.
Although a claim that the current regime is sufficient is unsupportable,
a strong argument does exist that the totality of environmental issues and
international environmental entities is far too extensive to be immediately
joined into one organization. Thus, any initial WEO will necessarily have to
be far less than comprehensive. Based on this reality, a good first step toward
Chapter 56 Page 186 September 11, 200820:0

186 International Environmental Issues and Debates

reform might be to cluster related multilateral environmental agreements

(MEAs) into three or four groups (see von Moltke, this volume), to build
new environmental organizations around them, and then perhaps to abolish
or redefine UNEP. Such a plan would emphasize the linkages among related
treaties and environmental entities.
Nevertheless, clustering has its own pitfalls (see Biermann, this volume).
Whatever clusters are designed will leave out important links between the
cluster. Furthermore, all of the functional tasks identified by Esty and
Ivanova (2002), such as data collection and monitoring, would seem to be a
cross-cutting feature of each cluster. Similarly, the bargaining proposed by
Whalley and Zissimos (2002) would be stunted if it had to occur within each
cluster, rather than across clusters.
Alternatively, the first step could be to establish the WEO initially with
only planning and budget functions. The WEO could seek to address the
biggest flaw in the status quo, which is that no ongoing mechanism exists to
identify the most serious gaps in the stewardship of environmental resources
and to determine where new environmental investments are most needed.
Such a WEO could hold annual conferences at the ministerial level and
more frequent meetings on particular topics. It could also set up a inter-
parliamentary assembly to serve a consultative role.
Of all the existing international environmental entities, the Global Envi-
ronment Facility (GEF) is perhaps the best model for a more extensive
environmental organization (Streck 2001). It focuses on six critical global
environmental threats biodiversity loss, climate change, degradation of
international waters, ozone depletion, land degradation, and persistent
organic pollutants. It acts as a funding entity whose implementing agen-
cies comprise UN agencies and the World Bank. It is run with a small
bureaucracy. It has achieved close relationships with the major MEAs. It
permits some participation by NGOs, including at Council meetings. It has
begun to gain more private sector involvement. The GEF operates transpar-
ently. With 176 member countries, the GEF has adopted a creative solution
for solving the problem of internal democratic governance. The governing
Council is reasonably-sized (32 members) with more from developing than
developed countries. Even more innovatively, the members on the Council
are appointed by a constituency of states for whom they represent (with some
large states representing only themselves). As the GEF continues to mature,
and its projects are evaluated, there may come a time when the perennial calls
to strengthen UNEP are replaced by a more apt proposal to broaden GEF.

3. The Paradigm of WEO as a Competitor

In the literature advocating a WEO, the rationale for the organization is

described as promoting coherence within the regime, achieving economies
and efficiencies, or carrying out cooperation with other international orga-
nizations (for example, Bergsten 1994, 364). Thus, much of the emphasis
Chapter 56 Page 187 September 11, 200820:0

Charnovitz Toward a World Environment Organization 187

has been about what functions should be included within the WEO and what
functions should be excluded.
Instead of designing a WEO with an eye only to internal coherence and
external cooperation, the models for a WEO should better reflect an under-
lying goal of making the WEO a more effective competitor against other
regimes. The notion that a WEO would be an institutional counterweight
to the trading system was a key insight in Estys early analysis (see Esty 1994,
230; see also Esty 1999, 156061). Yet Esty, even while emphasizing the
value of competition in other contexts, has not highlighted competition as a
paradigmatic feature of the WEO. In my view, a WEO ought to champion
environmental causes as it interacts with other international organizations
such as the WTO, the World Bank, and the UN Development Programme.
The idea that the architecture of governance requires competition is an old
one, going back to James Madison, if not earlier. In Federalist Papers No. 51,
Madison explains that the United States Constitution should contrive the
structure of the government so that its several constituent parts may, by their
mutual relations, be the means of keeping each other in their proper places
(Madison 1788). He further explains that in all subordinate distributions of
power, the aim is to divide the several offices in such a manner as that each
may be a check on the other. Although Madison writes about a national
constitution, the same principle could apply to an international system or
Overlapping competence of agencies is a characteristic feature of inter-
national governance. Most analytical attention seems to go to managing the
overlaps at different vertical levels of authority. Yet the horizontal overlaps
are equally challenging, and require active efforts to seek coherence (see
Sampson 2003). My point here is that while cooperation is one avenue to
obtain coherence, it is not an exclusive one. Coherence can also be achieved
through competition.
The need for competition is most apparent in the relationship between
international environmental and international economic governance. In
recent years, the WTO has climbed to a dominant position from which it
seeks to insinuate its norms into other organizations. The claim is often
made that WTO law trumps other bodies of law, and that environmental
treaties need to conform to trade law. Since the WTO went into force in
1995, environmental treaty negotiations have been monitored carefully to
make sure that they do not contravene WTO rules. This adversarial stance
by the trading system has led to a chill in environmental policy-making.
Close observers of the WTO recognize that it suffers a superiority complex
(Pauwelyn 2003, 1177). Currently, the WTO is negotiating several environ-
mental issues in the Doha Round. Although UNEP and several secretariats
of multilateral environmental agreements have been invited to some of these
negotiating sessions as ad hoc observers, the environment regime is powerless
within the WTO to exercise any influence.
Because the trading regime is likely to give much more weight to commer-
cial rather than ecological values, what has been missing is an evenly matched
Chapter 56 Page 188 September 11, 200820:0

188 International Environmental Issues and Debates

environment regime that can promote its norms in other arenas, and stand up
to resist any overreaching by the WTO, or by new free trade agreements. Just
as the WTO is now delving into trade-related environment policy, it would
be appropriate for a WEO to delve into environment-related trade policy. For
example, if a WEO existed now, it could be monitoring WTO negotiations
on services to make sure that any new disciplines do not undercut environ-
mental regulation. The recent proposal for a WEO included in the Heinrich
Bll Foundations Joburg Memo takes account of the value of horizontal
competition at the international level. The Memo states that No system of
checks and balances can be installed unless organizations like the ILO, the
WHO, and the WTO are joined by an environmental organization of equal
standing (Sachs et al. 2002, 65).
It is beyond the scope of this article to present an organizational blueprint
for a WEO that would prescribe a method of decision-making and means of
enforcement. Good lessons can be learned from the GEF, and from the multi-
lateral environmental agreements, which have pioneered institutional inno-
vations (see Churchill and Ulstein 2000). The new network of environmental
regulators is another important development that should be considered in
designing a WEO.16 In an era where hierarchy is giving way to networks,
insights can come from any direction.
In calling for a WEO that could serve as a counterweight to the WTO
and other institutions of economic governance, this article is not endorsing
the WTO constitution as a template for a WEO. Certainly, the WTO has
strengths that might be copyable. But the WTO also has many weaknesses,
most notably its adherence to consensus-based decision-making that has
recently arrested progress in the current Doha round of negotiations.
In its competition with the WTO, a WEO would have one important
advantage. In contrast to the WTO, in which the vertical relationship with
national trade ministries is one of supervision rather than cooperation, a
WEO could develop a more cooperative association with national environ-
ment ministries, which themselves would be in a cooperative relationship
with each other. Such environmental interdigitation toward common goals
would be a feature not present in the trade regime, which is hard-wired
for economic nationalism. Because the environment regime is so weak at
the international level, most of the possibilities for fruitful vertical coopera-
tion remain to be harvested. For example, improvements to environmental
legal norms could occur through more systematic vertical borrowing (Wiener
2001). New ideas at the local level could be evaluated and, if successful,
offered to other countries.

4. Conclusion

In penning this reflection, I am mindful that progress in ecological protec-

tion continues to occur, and that the tiny steps at Johannesburg may yield
dividends not yet apparent. What worries me though is that the remarkable
Chapter 56 Page 189 September 11, 200820:0

Charnovitz Toward a World Environment Organization 189

resilience of the biosphere is being taken for granted. Many opportunities to

prevent a loss of natural resources are being missed.
I am also mindful that diversity within the environment regime can
be valuable (Sand 2001, 297), and that many environmental tasks are
disjoinable from others. Nevertheless, the fragmented nature of todays
environmental governance defies organizational logic and perpetuates weak
responses. If, over the next decade, UNEP is cabined to its present status
and no better methods ensue for carrying out international environmental
policy, then governments may fail to make much progress in responding to
global challenges.
Supporters of a WEO should renew efforts to make the case for why orga-
nizational change can improve policies. In this article, I have addressed each
aspect of the W-E-O, and pointed out the danger of allowing environmental
governance to muddle along while economic governance grows stronger.


1. The idea of an international organization for the environment certainly predates 1992,
but this article focuses on the period since then. The quest for better international institutions for
the environment began in 1909 with separate efforts by Theodore Roosevelt and Paul Sarasin. A
recent volume by Martin Holdgate discusses this early history briefly, and has good chapters on
the progress made transgovernmentally in the late 1940s (Holdgate 1999). The next renaissance
for environmental governance was the early 1970s. One analyst of that era was Abram Chayes
(see Chayes 1972).
2. This volume uses the term World Environment Organization (WEO), and so that
convention will be followed in this article. Nevertheless, a GEO would be a much better name
for such an institution. The term GEO is also the name for the Global Environment Outlook
prepared by the UN Environment Programme.
3. See http://www.iisd.ca/unepgc/gmef3/.
4. The UN University held a panel session on governance at the third WSSD PrepCom,
and the papers are now in publication (see Chambers and Green 2004).
5. It is interesting to note that the first paragraph of the Johannesburg Declaration on
Sustainable Development begins We, the representatives of the peoples of the world   
(Johannesburg Declaration 2002). One sees a similar populist theme on the home page of the
UN website which begins, United Nations. Its Your World.
6. The ILO achieves this. It is true that ILO Conventions are not law until they are ratified
by governments. But a requirement for approval of new law at the national level is the norm
for all specialized organizations, including the WTO and the World Health Organization. The
only major international organization with authority to write new rules that are automatically
obligatory is the UN Security Council.
7. Sustainable development is commonly dened as development that meets the needs of
the present without compromising the ability of future generations to meet their own needs.
See http://www.un.org/esa/sustdev/about_us/aboutus.htm.
8. For example, the ILO Declaration on Fundamental Principles and Rights at Work
(1998) refers in preambular language to the goal of broad-based sustainable development.
See http://www.ilo.org/public/english/standards/decl/declaration/text/index.htm. See also the
UN Millennium Goal no. 7 to Integrate the principles of sustainable development into
country policies and programmes; reverse the loss of environmental resources, available at
9. See http://www.un.org/partners/civil_society/calendar/c-dvcop.htm. Another self-
congratulation on the UN website about the Implementation Summit says that one of
Chapter 56 Page 190 September 11, 200820:0

190 International Environmental Issues and Debates

the major accomplishments was strengthening of the concept of sustainable develop-

ment and the important linkages between poverty, the environment and the use of natural
resources. See The Road from Johannesburg: What was achieved and the way forward, at
10. A few positive outcomes from Johannesburg are worth noting. One was the program-
matic emphasis on initiating partnerships between governments, business, civil society, and
international organizations. Another was the Global Judges Symposium which brought together
senior judges from 59 countries and from international courts and tribunals. The judges adopted
the Johannesburg Principles on the Role of Law and Sustainable Development (available in
UNEP/GC.22/INF/14). Still another was the inauguration of the Partnership for Principle 10 to
promote good environmental governance at the national level, including transparency, partic-
ipation, and access to justice. Principle 10 was part of the Rio Declaration of 1992. The new
Partnership includes governments, international organizations, and NGOs. See www.pp10.org.
11. This is generally true, but not in the United States which lacks an environmental ministry
with international competence. The US Department of State represented the United States at
the Johannesburg Summit, and also does so at the GMEF and the UNEP Governing Council.
12. As of 27 June 2004.
13. For a more positive view, see Kirton, this volume.
14. See http://www.wto.org/english/thewto_e/whatis_e/tif_e/org2_e.htm.
15. By contrast, the new World Resources report glorifies the current structure as a
Symphony of Organizations (UN Development Programme et al. 2003,139).
16. International Network for Environmental Compliance and Enforcement, available at


Bergsten, C.F. (1994), Managing the World Economy of the Future, in P.B. Kenen (ed.),
Managing the World Economy: Fifty Years after Bretton Woods, Institute for International
Economics, Washington, 34174.
Bermann, G.A. (1994), Taking Subsidiarity Seriously: Federalism in the European Community
and the United States, Columbia Law Review, vol. 94,332456.
Biermann, F. (2000), The Case for a World Environment Organization, Environment, vol. 42
(9), 2331.
Biermann, F. (2001), The Emerging Debate on the Need for a World Environment Organiza-
tion, Global Environmental Politics,vol. 1 (1), 4555.
Biermann, F. (2005), The Rationale for a World Environment Organization, in F. Biermann
and S. Bauer (eds), A World Environment Organization. Solution or Threat for Effective
International Environmental Governance?, Ashgate, Aldershot, 117144.
Carozza, P.G. (2003), Subsidiarity as a Structural Principle of International Human Rights
Law, American Journal of International Law, vol. 97, 3879.
Chambers, W.B. and J.F. Green (eds) (2004), Reforming International Environmental Gover-
nance: From Institutional Limits to Innovative Solutions, UNU Press, Tokyo.
Chayes, A. (1972), International Institutions for the Environment, in J. Lawrence
Hargrove (ed.), Law, Institutions and the Global Environment, Oceana Publications,
Dobbs Ferry, 126.
Churchill, R.R. and G. Ulstein (2000), Autonomous Institutional Arrangements in Multilateral
Environmental Agreements: A Little-Noticed Phenomenon in International Law, American
Journal of International Law,vol. 94, 62359.
Dowdeswell, E. and S. Charnovitz (1997), Globalization, Trade and Interdependence, in
M.R. Chertow and D.C. Esty (eds), Thinking Ecologically. The Next Generation of Envi-
ronmental Policy,Yale University Press, New Haven.
Dua, A. and D.C. Esty (1997), Sustaining the Asia Pacic Miracle, Institute for International
Economics, Washington.
Chapter 56 Page 191 September 11, 200820:0

Charnovitz Toward a World Environment Organization 191

Elliott, L. (2005), The United Nations Record on Environmental Governance: An Assessment,

in F. Biermann and S. Bauer (eds), A World Environment Organization. Solution or Threat
for Effective International Environmental Governance?, Ashgate, Aldershot, 2756.
Esty, D.C. (1994), Greening the GATT. Institute for International Economics, Washington.
Esty, D.C. (1999), Toward Optimal Environmental Governance, New York University Law
Review, vol. 74, 1495574.
Esty, D.C. (2001), A Terms Limits, Foreign Policy, September/October, 7475.
Esty, D.C. and M.H. Ivanova (2002), Revitalizing Global Environmental Governance:
A Function-Driven Approach, in Esty and Ivanova (eds), Global Environmental Gover-
nance. Options and Opportunities, Yale School of Forestry and Environmental Studies,
New Haven (Conn.), 181203.
French, H. (2003), New Leadership from the South, World Watch, JulyAugust, 2.
G8 Environment Ministers Communiqu (2003), available at http://www.g7.utoronto.ca/-
Gaines, S.E. (2003), The Problem of Enforcing Environmental Norms in the WTO and What
To Do About It, Hastings International and Comparative Law Review, Spring 2003,
Gardner, R.N. (1974), The Hard Road to World Order, Foreign Affairs, April 1974, 55676.
German Advisory Council on Global Change (2001), New Structures for Global Environmental
Policy, Earthscan, London.
Global Accountability Report (2003), Power Without Accountability?, One World Trust,
[available at http://www.oneworldtrust.org/].
Gray, K.R. (2003), World Summit on Sustainable Development: Accomplishments and New
Directions? International and Comparative Law Quarterly, vol. 52, 25668.
Gutman, P. (2003), What Did WSSD Accomplish? An NGO Perspective, Environment, vol. 45
(2), 2128.
Haas, P.M. (2004), Addressing the Global Governance Decit, unpublished paper.
Hierlmeier, J. (2002), UNEP: Retrospect and Prospect: Options for Reforming the Global Envi-
ronmental Governance Regime, Georgetown International Environmental Law Review,
vol. 14, 767805.
High-Level Panel on Financing for Development, Report, in UN General Assembly, A/55/1000
(26 June 2001).
Holdgate, M. (1999), The Green Web, Earthscan, London.
Holliday Jr., C.O., S. Schmidheiny and P. Watts (2002), Walking the Talk. The Business Case
for Sustainable Development, Greenleaf Publishing, Shefeld.
Johannesburg Declaration and Plan of Implementation (2002), available at http://www.
Juma, C. (2000), The Perils of Centralizing Global Environmental Governance, Environ-
ment Matters (World Bank), 2002, 1315, [available at http://www-wds.worldbank.-
Keohane, R.O., P.M. Haas and M.A. Levy (1993), The Effectiveness of International Environ-
mental Institutions, in P.M. Haas, R.O. Keohane and M.A. Levy (eds), Institutions for the
Earth, MIT Press, Cambridge (Mass.), 324.
Kirton, J. (2005), Generating Effective Global Environmental Governance: The Norths Need
for a WEO, in F. Biermann and S. Bauer (eds), A World Environment Organization. Solu-
tion or Threat for Effective International Environmental Governance?, Ashgate, Aldershot,
Levy, M.A., R.O. Keohane and P.M. Haas, (1993), Improving the Effectiveness of International
Environmental Institutions, in Haas, Keohane and Levy (eds), Institutions for the Earth,
MIT Press, Cambridge (Mass.), 397426.
Lodefalk, M. and J. Whalley, (2002), Reviewing Proposals for a World Environmental Organ-
isation, The World Economy, vol. 25, 60117.
MacDonald, G. J. (2003), Environment: Evolution of a Concept, Journal of Environment and
Development, vol. 12, 15176.
Chapter 56 Page 192 September 11, 200820:0

192 International Environmental Issues and Debates

Madison, J. (1788), No. 51, available at http://memory.loc.gov/const/fed/fedpapers.html.

Malm Declaration (2000), available at www.unep.org/malmo/malmo_ministerial.htm.
Marshall, D. (2002), An Organization for the World Environment: Three Models and Analysis,
Georgetown International Environmental Law Review, vol. 15, 7903.
McNeill, J.R. (2000), Something New Under the Sun. An Environmental History of the
Twentieth-Century World, W.W. Norton and Company, New York.
Najam, A. (2002), Why We Dont Need a New International Environmental Organiza-
tion (Draft), [available at http://www.bu.edu/cees/research/workingp/pdfs/Najam-GEO-
Najam, A. (2005), Neither Necessary, Nor Sufcient: Why Organizational Tinkering Wont
Improve Environmental Governance, in F. Biermann and S. Bauer (eds), A World Environ-
ment Organization. Solution or Threat for Effective International Environmental Gover-
nance?, Ashgate, Aldershot, 235256.
Nanda, H. S. (2003), India Convinces UNEP to Eliminate Asian "Brown Cloud" from Meeting
Agenda, International Environment Reporter, vol. 26, 246.
Newell, P. (2002), A World Environmental Organisation: The Wrong Solution to the Wrong
Problem, The World Economy, vol. 25, 65971.
Oberthr, S. and T. Gehring (2005), Reforming International Environmental Governance: An
Institutional Perspective on Proposals to Establish a World Environment Organization, in
F. Biermann and S. Bauer (eds), A World Environment Organization. Solution or Threat
for Effective International Environmental Governance?, Ashgate, Aldershot, 205234.
Ostry, S. (2001), The WTO and International Governance, in K.G. Deutsch and B. Speyer
(eds), The World Trade Organization Millennium Round, Routledge, London, 28594.
Palmer, G. (1992), New Ways to Make International Environmental Law, American Journal
of International Law, vol. 86, 25983.
Pauwelyn, J. (2003), What to Make of the WTO Waiver for Conict Diamonds: WTO
Compassion of Superiority Complex? Michigan Journal of International Law, vol. 24,
Pius XI, Pope (1931), Quadragesimo Anno, on Reconstruction of the Social Order, avail-
able at http://www.vatican.va/holy_father/pius_xi/encyclicals/documents/hf_p-xi_enc_-
Raustiala, K. (2002), The Architecture of International Cooperation: Transgovernmental
Networks and the Future of International Law, Virginia Journal of International Law,
vol. 43, 192.
Rio Declaration on Environment and Development (1992), in United Nations, Earth Summit
Agenda 21, United Nations, New York.
Runge, C.F. (1994), Freer Trade, Protected Environment, Council on Foreign Relations, New
Runge, C.F. (2001), A Global Environment Organization (GEO) and the World Trading
System, Journal of World Trade, vol. 35, 399426.
Sachs, W. et al. (2002), The Joburg Memo. Fairness in a Fragile World, Heinrich Bll Foun-
dation, available at http://www.joburgmemo.org/.
Sampson, G.P. (2003), Is there a Need for Restructuring the Collaboration among the WTO and
UN Agencies so as to Harness Their Complementarities, Paper presented at the European
University Institute, June 2003.
Sand, P.H. (2001), Environment: Nature Cooperation, in P.J. Simmons and C. de J. Oudraat
(eds), Managing Global Issues. Lessons Learned, Carnegie Endowment for International
Peace, Washington, 281309.
Shadow G8 (2003), Restoring G8 Leadership of the World Economy, May 1993, available
at http://www.ifri.org/les/policy_briefs/WP_SHADOW_G8.pdf.
Simonis, U.E. and T. Brhl (2002), World EcologyStructures and Trends, in P. Kennedy,
D. Messner and F. Nuscheler (eds), Global Trends and Global Governance, Pluto Press,
London, 97124.
Speer, L.J. (2003), G-8 Environment Ministers Pledge Better Maritime Safety, Disagree on
Timing, BNA Daily Environment Report, 29 April 2003, A-4.
Chapter 56 Page 193 September 11, 200820:0

Charnovitz Toward a World Environment Organization 193

Speth, J.G. (2002a), The Global Environmental Agenda: Origins and Prospects, in Esty
and Ivanova (eds), Global Environmental Governance. Options and Opportunities, Yale
School of Forestry and Environmental Studies, New Haven, 1130.
Speth, J.G. (2002b), Attacking the Root Causes of Global Environmental Deterioration, Envi-
ronment, vol. 44 (7), 1625.
Speth, J.G. (2004), Red Sky at Morning, Yale University Press, New Haven.
Stein, R. (2003), WHO Gets Wider Power to Fight Global Health Threats, Washington Post,
28 May 2003.
Stone, CD. (1993), The Gnat is Older than Man, Princeton University Press, Princeton.
Streck, C. (2001), The Global Environment FacilityA Role Model for International Gover-
nance?, Global Environmental Politics, vol. 1 (2), 7194.
Supachai, P. (2001), The Evolving Multilateral Trading System in the New Millennium,
George Washington International Law Review, vol. 33, 41949.
Tarlock, A.D. (2001), Ideas without Institutions: The Paradox of Sustainable Development,
Indiana Journal of Global Legal Studies, vol. 9, 3549.
Tolba, M.K. (1998), Global Environmental Diplomacy, MIT Press, Cambridge (Mass.).
Turner, T. (2000), Critical Crossroads, Our Planet, vol. 10 (5), 45.
UNDP [United Nations Development Programme], UNEP [United Nations Environment
Programme], World Bank, World Resources Institute (2003), World Resources 20022004,
World Resources Institute, Washington.
UNEP [United Nations Environment Programme] (2002), UNEP in 2002, available at
Verges Bausili, A. (2002), Rethinking the Methods of Dividing and Exercising Powers in the
EU: Reforming Subsidiarity and National Parliaments Jean Monnet Working Paper 9/02,
available at www.jeanmonnetprogram.org/papers.
von Moltke, K. (2001), The Organization of the Impossible, Global Environmental Politics,
vol. 1 (1), 2328.
von Moltke, K. (2005), Clustering International Environmental Agreements as an Alterna-
tive to a World Environment Organization, in F. Biermann and S. Bauer (eds), A World
Environment Organization. Solution or Threat for Effective International Environmental
Governance?, Ashgate, Aldershot, 175204.
Whalley, J. and B. Zissimos (2001), What Could a World Environmental Organization Do?
Global Environmental Politics, vol. 1 (1), 2934.
Whalley, J. and B. Zissimos (2002), Making Environmental Deals: The Economic Case for a
World Environmental Organization, in D. Esty and M. Ivanova (eds), Global Environ-
mental Governance. Options and Opportunities, Yale School of Forestry and Environ-
mental Studies, New Haven (Conn.), 16380.
Wiener, J.B. (2001), Something Borrowed for Something Blue: Legal Transplants and the
Evolution of Global Environmental Law, Ecology Law Quarterly, vol. 27, 1295371.
Wolfrum, R. and N. Matz (2003), Conicts in International Environmental Law, Springer,
World Commission on Environment and Development (1987), Our Common Future, Oxford
University Press, Oxford.
Worldwatch (2003), State of the World 2003, W.W. Norton and Company, New York.
Chapter 57 Page 194 September 11, 200820:12

Addressing the Global Governance Deficit
Peter M. Haas

he Brundtland Commission report wrote, The globe is one, but
the earth is not.1 The challenge for effective governance is how to
encourage governments to pursue comprehensive policies to achieve
Sustainable Development within an international political context that has
inhibited the pursuit of comprehensive and long-term goals. An interna-
tional society of states founded on the principle of national sovereignty alone
discouraged serious state attention to international environmental external-
ities. It suppressed the voices of those in ecologically threatened areas, often
the poor within societies, and led states to ignore the global commons outside
national jurisdictions. While developing countries have not been pollution
havens for the rich, they do lack the ability to forcefully present their posi-
tions at international negotiations on sustainable development and environ-
mental protection.

Brave New World

The contemporary international political system faces two new geopolit-

ical realities, outlined below, that challenge the old geographical principles
of national sovereignty.2 Consequently there is the potential for replacing
the traditional dichotomous concepts of global governance organized hier-
archically or anarchically3 with a network model of decentralized global
governance. Such a networked model involves the performance of a variety
of discrete functions by multiple actors, whose interactive effects in prac-
tice can yield more effective global coordination and performance of major
governance functions.4 This is a political project or vision of incremental
multilateralism, as more parties become part of a growing project of global-
ization over which each has an interest and a say. Such an approach would
establish the institutional mechanisms for promoting the beneficial features
or globalization while minimizing the more egregious negative effects.
The first new geopolitical reality is the growing complexity of a glob-
alizing world, whose management requires more holistic or comprehensive
policies to address environmental externalities (a diplomatic term for ecolog-
ical collapse) and to support sustainable development. Most international

Source: Global Environmental Politics, 4(4) (2004): 115.

Chapter 57 Page 195 September 11, 200820:12

Haas Addressing the Global Governance Deficit 195

and national institutions were designed historically to address discrete prob-

lems, whereas the current globalized agenda consists of intertwined issues
(or what organizational theorists term non-decomposable or partially non-
decomposable problems). Effective management of these new intertwined
issues requires procedures for responsible agencies (either nationally or inter-
nationally, or states as a whole) to think about how their actions will
affect the responsibilities of other autonomous agencies and how their policy
domain may be affected by decisions taken in or by other bodies.5
The second new geopolitical reality is the proliferation of new political
actors and the diffusion of political authority over major governance func-
tions, particularly in the environmental sphere. These new actors include
NGOs, MNCs, organized transnational scientific networks known as epis-
temic communities, global policy networks, and selective international
institutions that are capable of exercising discretionary behavior indepen-
dently of the wishes of their dominant member states.6 These include, for
example, UNEP, the World Bank since 1987, the ECJ, and possibly the EU

International Reponses to Global Complexity

Since 1972 there have been many efforts to design international institu-
tions to better harmonize international decision-making to promote envi-
ronmental protection, and, later, to promote sustainable development.7 The
United Nations Environment Programme (UNEP) was established in 1973
to be catalytic and performed remarkably well with fairly scarce resources.
It helped to develop a significant body of international environmental
law, encouraged other international institutions to take account of the
environmental consequences of their programmatic activities, and trained
hundreds of developing country officials in techniques of ecological resource
But things have changed dramatically in the last 30 years with the spread
of environmental consciousness and the proliferation of Multilateral Envi-
ronmental Agreements (MEAs). UNEP is now under-funded, over-loaded
and remote.8 It is relatively obsolete, eclipsed in resources and prestige
by other international institutions that have taken on new environmental
responsibilities, such as the World Bank, the EU for Europe, and even, to
the dismay of many environmentalists, the WTO. Indeed many NGOs, such
as the World Resources Institute, Greenpeace and TRAFFIC have assumed
some of the functions of global governance, including environmental moni-
toring and policy verification.
Recent years have seen several major reform proposals for redesigning
the United Nations and Bretton Woods systems which recognize these new
political realities and address the perceived gaps in the performance of
some key governance functions in the realms of environmental protection
and sustainable development. In other words, these proposals identify a
Chapter 57 Page 196 September 11, 200820:12

196 International Environmental Issues and Debates

governance-deficit. The primary functions of environmental governance are

presented in Table 1. below.

Why Reform?

Three broad arguments are provided for the need for such reform.9
The first focuses on redundancy and overlapping responsibilities and
tasks among international institutions. These are presumed to be a bad
thing out of a general rationalist impulse for simplification and centraliza-
tion. There is seen to be the result of inefficient use of scarce resources by
competing international institutions, and the logistical difficulties faced by
small bureaucracies who have to attend and prepare for too many meet-
ings at the institutions and the MEA Conferences of Parties. For the ratio-
nalist redundancy, inefficiency and logistical difficulties constitute profound
impediments for effective international governance. I find this argument
unpersuasive because I think that some degree of redundancy is actually
desirable in the international system, as it provides insurance against the
decline of any individual international institution and fits better with an
ecological institutional design vision of requisite diversity. Moreover, redun-
dancy provides for more contact and linkage between institutions. If the
governance deficit is due to performance gaps then responses should be
addressed through capacity building. If the governance deficit is due to
redundancies between international institutions then responses should await
a clear inventory and assessment of the performance of vital governance
The second argument is a straightforward efficiency argument. More
activities could be conducted if there was less competition for resources
between organizations. I find this unpersuasive because it would have the
effect of consolidating political influence in the international system in a
smaller number of major IOs, increasing the possibility of political capture
and the actual decline in the efficient use of financial resources due to capri-
cious national budgetary cycles, as has been a recurrent problem for UNEP.
If the governance deficit is simply a problem of UNEPs resources, then it
should be elevated to UN Specialized Agency status and given more money.
The third argument is the need for a strong environmental presence in the
international system, especially as an environmental advocate at the WTO,
or as a counterweight to the WTO in trade and environment disputes. I find
this argument persuasive but it is not clear that a counterweight to the WTOs
presumptive bias towards trade liberalization over environmental protec-
tion in its trade and environment arbitration decisions requires the creation
of a massive countervailing institution, particularly since the WTOs recent
dispute panel record has upheld some environmental protection in its deci-
sions, such as the protection of sea-turtles.10 Rather, as I argue below, such
a counter-weight can be more pragmatically pursued through institutional
Table 1: Environmental Governance Function Performance by Different Actors
Chapter 57 Page 197

Governance International Scientific

Function States Institutions MNCs NGOs Communities

Agenda setting X (from national X (from monitoring, X (i.e. the media) X X (publications,
monitoring) publicizing testimony)
others alerts)
Framing X (by first movers X (i.e. UNEP, also X (the media) X (new information Monitoring
and through financial provided by
negotiations) incentives and epistemic

demonstration communities)
effects from IFIs)
Monitoring X X X (through some X X
ISO standards)
Verification X X X X
Rule Making X (regime X (guidelines, X (voluntary X (principled X (indirectly through
development) regime sectoral standards, examples consulting to states
development) guidelines) include Brent Spar, & international
whaling moratorium) institutions)
Norm Development X (soft law) X X (i.e. Amnesty
Earth Council?)
Enforcement X X X X (i.e. TRAFFIC)
Capacity Building X (ODA) X (various support X (DFI) X (subcontracting to X (technical training)
activities) states & IOs, and
direct grassroots
activities. Training
Addressing the Global Governance Deficit

Financing X (ODA) X (GEF, IBRD, X (DFI)

September 11, 200820:12
Chapter 57 Page 198 September 11, 200820:12

198 International Environmental Issues and Debates

reforms that amplify environmental voices within WTO trade and envi-
ronment dispute resolution panels. This is largely because the political will
behind the creation of a mammoth new international organization is clearly
lacking in the US, which would have to be largest funding source for such an
institution. Moreover, the organizational design of a powerful centralized
body is poorly suited to deftly coordinate complex decentralized tasks.

Proposals for Reform

The most ambitious reform is the creation of a Global Environmental

Organization, or World Environmental Organization. This idea has been
proposed most forcefully by the German Advisory Council on Global
Change, the German academics Frank Biermann and Udo Simonis, and
Dan Esty and the Yale Center for Environmental Law and Policy.11 Chan-
cellor Schroeder and President Chirac have publicly supported this initiative,
although it has absolutely no support in the United States government. The
US remains selectively committed to most elements of multilateral environ-
mental diplomacy, despite its reversal on the Kyoto Protocol, but it is not
interested in potentially expensive institutional reform or the creation of new
international institutions until 2005 at the earliest.
A more modest suggestion came from the 1997 Task Force on Environ-
ment and Human Settlements, which, not surprisingly, has been adopted by
UNEPs Governing Council. It suggests strengthening UNEP by elevating it
to a specialized agency (and thus being entitled to a fixed and regular budget)
and by improving its ability to coordinate activities with other specialized
agencies. However, no clear guidelines were given about how such coordi-
nation was to be achieved in the absence of strong political will by member
governments or the heads of the agencies. France is currently circulating a
slightly more comprehensive proposal for a strengthened UNEP that would
conduct more scientific activities without shedding any of its present respon-
sibilities. It is doubtful that there is much US government support for this
proposal, or much concern by the US electorate. Positions outside the US
have not yet been clearly formulated.
Global Ministerial Environment Forums have been meeting in conjunc-
tion with the UNEP Governing Council since 1998, and the Sixth Special
Session of the Governing Council/Global Environmental Forum issued the
Malmo Declaration which called for strengthening general international
environmental governance and deeper institutionalization of the Global
Ministerial Environment Forums. These meetings are essentially a periodic
set of summits for environmental ministers. While such high-profile meetings
are a good way to encourage the adoption of high-sounding commitments
when the domestic climate is favorable in the major countries, they lack any
ongoing administrative abilities or institutional memory for how to conduct
effective multilateral environmental diplomacy.
Chapter 57 Page 199 September 11, 200820:12

Haas Addressing the Global Governance Deficit 199

These proposals constitute an overly narrow conception of potential

responses to the governance deficit. UNEP reform still puts too much reliance
on a small and remote organization. Yet there is no political future for elimi-
nating UNEP, because developing countries insist on having the headquarters
of a UN agency in a developing country. The question then is what to do
with UNEP, since it is currently overstretched.
On the other hand, creating a new centralized GEO appears utopian,
given the lack of political will in the US and abroad, lack of popular concern
in the mass publics, and the general distraction for the international commu-
nity provided by the ght against terrorism.
Proposals for increased centralization of responsibilities or the creation
of a new monolithic body run counter to the key insight of the most sophis-
ticated current organizational theorists about the best institutional design
for managing complex problems like global environmental issues. The best
institutional structure for dealing with complex and uncertain policy envi-
ronments is loose, decentralized, dense networks of institutions and actors
that are able to quickly relay information, and provide sufficient redundan-
cies in the performance of functions so that the elimination or inactivity
by one institution does not jeopardize the entire network.12 Decentralized
information-rich systems are the best design for addressing highly complex
and tightly-coupled problems. In short, strong centralized institutions are
fundamentally unecological. They run counter to the ecological principle of
requisite diversity or flexibility; inhibit random mutation, or policy innova-
tion; and are easily captured by single powerful parties.
A better way of reforming the global environmental governance system,
I argue, takes this new decentralized governance design principle seriously. In
international circles it is now referred to as multilevel governance. The United
Nations Global Compact, for example, is a recent effort to institutionalize
multilevel governance within a network of networks that includes a variety of
non-state actors without sacrificing the principle of national sovereignty.13
Such a network model would involve the streamlining and improvement
of the performance of existing governance efforts, rather than creating new
governance bodies. A lot of governance is clearly already going on, the trick
is to improve it and to enhance the synergies between the performance of
these different functions. What is needed is a clearer map of the actual divi-
sion of labor between governments, NGOs, the private sector, scientific
networks and international institutions in the performance of various func-
tions of governance. Also needed is an assessment of both the comparative
advantages of these various actors and how well they actually perform these
activities. Further study is necessary to get a better understanding of the
comparative advantage enjoyed by different actors in the performance of
different functions and to learn more about creative institutional designs
in which multiple actors perform the same function and thus keep each
other honest. Democratic theory would suggest that such an adversarial
design would inhibit regulatory capture and shirking, as well as reduce
the likelihood of false positives and false negatives in the performance of
Chapter 57 Page 200 September 11, 200820:12

200 International Environmental Issues and Debates

monitoring and verification.14 Many MEAs in fact rely on such institutional

techniques as third party verification and independent institutional authority
for conducting monitoring and verification.
Effective governance rests on the performance of multiple governance
functions. Some of the key functions include agenda setting, framing, moni-
toring, verification, rule making, norm development, enforcement, capacity
building, and financing. Agenda setting entails identifying and sounding
alarms about potential international environmental risks. Framing entails
establishing the array of causal linkages associated with the management of
a particular issue, and thus the warranted substantive linkages that may be
made in negotiating a regime or set of policies for that issue. For instance,
the Forestry Stewardship Council and the UN Global Compact combine
multiple stake-holders (actors) in a constructive tension so that information
is shared and each holds the other accountable for their public commitments.
More profoundly, framing entails the assignment of anthropogenic respon-
sibility for an issue that may previously have been regarded as an act of god.
Such framing has two vital political consequences: one is to assign moral
blame, the second is to assign legal blame, so that if global climate change
were to be widely accepted as a phenomenon caused by human action, then
the insurance industry might be liable for paying for natural disasters that
had previously been treated as force-majeure or acts of god. Monitoring
involves collecting data about environmental quality. Verification is associ-
ated with assessing state policies in compliance with their obligations. Rule
making consists of establishing common standards for collective behavior:
regimes or policies. Norm development means the establishment of common
norms of expected behavior for a variety of different actors. Enforcement
is ensuring that actors actually comply with their obligations through delib-
erate actions. Capacity building covers a variety of techniques that enable
actors to live up to their obligations: including technology transfer, tech-
nical training, public education, and the collection of clearinghouses about
best available technology and best established practices. Financing relates to
the provision of money for actors to perform their functions and to pursue
sustainable development.
Some functions are formally performed: that is, the international commu-
nity directly tasks some agent to explicitly perform them. Others may be
performed indirectly: action is not the consequence of explicit instructions
by those contracting some set of activities to be performed by the relevant
actors. Some activities may have multiple indirect effects. For instance, by
publicizing issues, norm development and standard setting may be achieved.
By verifying and providing resources, one may achieve compliance. By mobi-
lizing civil society, governance efforts may promote agenda setting and
framing, and thus define new national preferences that narrow the range of
feasible negotiated outcomes. Educating elites and governments may have
similar effects.
Serious account of these new ideas is necessary in order to effectively
address the governance deficit. Governance should rely on a differentiated
Chapter 57 Page 201 September 11, 200820:12

Haas Addressing the Global Governance Deficit 201

division of labor among various elements of international civil society, with

clearer attention paid to coordinating their efforts, assuring rapid and accu-
rate information flow between them. Seriously applying this network vision
of governance would entail some reorganization, some consolidation, and
the creation of a limited set of new organizations.
I think that some core set of responsibilities should be left with UNEP,
such as that associated with its initial scientific research and monitoring
responsibilities, but that other governance functions should be redistributed
among other international actors. UNEP should be elevated to the status
of a UN agency in order to stabilize its financing. Its membership should
not be global though, as that would impede its decision making ability by
effectively introducing a unit-veto system for decisions made by consensus.
UNEP could help draft a global ecosystem assessment seeking to develop a
priority list of global environmental threats of interest to the international
community, and coordinate ongoing standing international scientific panels
to conduct environmental research and monitoring. Such a concentrated
and reinvigorated UNEP would contribute to improved agenda setting for
international environmental governance.


Agenda Setting

A more systematic early warning system is needed for alerting the interna-
tional community to impending environmental threats, such as suggested
above for UNEP. Current agenda setting is largely performed by NGOs,
who often provide exaggerated claims or false warnings. The challenge is to
develop early warning signals that are accurate that is, that do not miss
threats or cite potential urgent threats that subsequently prove unfounded.
Better environmental monitoring might improve agenda setting, but the
creation of standing international scientific panels responsible for evalu-
ating the state of the environment would be a valuable reform, similar to
the already existing IPCC, the Millennium Ecosystem Assessment, and the
Ozone Trends Panel.
Alternatively, more sophisticated measurement techniques for agenda
setting using social indicators of populations at risk could provide an early
warning system of ecosystem threats. Examples would be monitoring migra-
tion patterns of groups living in target ecosystems, tracking prices of scarce
resources, measuring keystone species for signs of threats to marine species,
or NGOs and scientists monitoring coral reefs as an early warning sign of
climate change.


Current arrangements for verifying state compliance with international

environmental obligations is very weak. NGOs could help keep track of
Chapter 57 Page 202 September 11, 200820:12

202 International Environmental Issues and Debates

governmental adherence to their international obligations, and a revital-

ized Earth Council could serve this purpose. The Earth Council was created
after the 1992 Rio Earth Summit, modeled on Amnesty International and its
verification role in human rights regimes, but has subsequently languished.

Technology Transfer and Financial Resource Transfers

There is clearly a gaping and continued need for financial transfers in support
of sustainable development. The GEF exists for this purpose, as does the
Montreal Ozone Fund, but the financial resources for these institutions are
inadequate to the task. Recent trends in overseas development assistance and
foreign investment also suggest that insufficient amounts of money are going
to countries in need of significant capacity building for sustainable develop-
ment. Incentives from industrialized country governments could encourage
MNCs to transfer green technologies to developing countries, and informa-
tional clearinghouses about green technologies in the public domain could
serve a public good here, created either by international institutions such
as the International Atomic Energy Agency (IAEA), NGOs, or the private


It is a common lament that many MEAs are not enforced, or that govern-
ments dont submit data on enforcement so that there is insufficient infor-
mation to enable informed judgments about the extent of enforcement with
MEAs. NGOs can help monitor enforcement, as well as the creation of
impartial 3rd party inspectors who would be able to inspect facilities for
compliance, such as the IAEA does, though with limited success.


Efficiency gains and the creation of new usable knowledge could be achieved
through the consolidation of the way that many of the current scientific (i.e.
research and monitoring) functions are performed at present.

Environmental Monitoring

Environmental monitoring should be consolidated by environmental

medium, to be conducted by consortia of international institutions and
scientific networks, and possibly NGOs, and even MNCs if they would be
willing to provide emission data. For instance, in the area of ocean moni-
toring there is a plethora of monitoring activities conducted by the IOC,
GESAMP, UNEPs Regional Seas Programme, and US NOAA. And this is
only a partial list. Different bodies are responsible for monitoring different
oceans. All of these activities should be formally consolidated, although the
Chapter 57 Page 203 September 11, 200820:12

Haas Addressing the Global Governance Deficit 203

Table 2: Secretariat Locations of Major MEA Secretariats

Location Number Comments

Geneva 11 Many of the interim secretariats for MEAs that are not yet
in force are in Geneva
London 9 Largely shipping related by IMO, OSPARCOM
Vienna 4 Related to nuclear safety
Rome 4 Administered by FAO
Montreal 3 ICAO, Montreal Ozone Fund, CBD

Source: Yearbook of International Cooperation for Environment & Development 2002/2003.

The rest of the 19 secretariats are spread over 17 locations.

logic of consolidation has not yet been established: should it be by simple

environmental medium, by geographic region, by common cause of the envi-
ronmental problem, or by some other justificatory logic?

Rule Making and MEA Administration

Many countries complain of regime saturation that exceeds their ability

to effectively participate in the management and development of particular
environmental regimes. This is largely because the treaty secretariats are
spread around the world and the schedules of major meetings are not coordi-
nated. Indeed, there are over 200 international organizations involved with
administering MEAS, although the number of major influential ones is prob-
ably under a dozen. Similarly the actual secretariats are not as widely spread
as many critics would imply. The following table shows the headquarters of
54 of the major current MEAs (not all of which are yet in force.)
Consolidating the MEA secretariats in one location makes sense in this
regard. Geneva, London, Bonn and many other cities would fulfill these
criteria. Consolidation would make travel easier for government officials,
and would facilitate joint activities between the environmental regimes and
their secretariats. Most importantly, perhaps, it would also have the effect of
creating the equivalent of standing environmental embassies at this central-
ized location which would improve national foreign environmental policy
making as well as elevating the profile of environmental policy makers
within their own foreign ministries and governments. The basic institutional
requirements for this clustering proposal are a location with good telecom-
munications, sufficient office space, and ample conference facilities. It is
unclear to what extent a new organization body would be required, or
whether it could simply consist of a MEA department store, all under one
roof. When the GATT became the WTO there was a massive change in
institutional influence and design, but in practical terms it merely entailed
changing the sign over the front door and printing new business cards for
the secretariat.
Chapter 57 Page 204 September 11, 200820:12

204 International Environmental Issues and Debates


Several governance functions are inadequately performed, and probably

require the creation of new institutions to improve their performance. It is
not clear that one institution needs to serve all these functions, or whether
they could be assigned to different bodies.

Norm Setting

A High Commission for the Environment should be created so that there

would be a high profile gure able to help develop normative principles
for environmental protection and Sustainable Development, akin to the
UNHCR (Refugees) or UNHCHR (Human Rights).

Protecting the Environment from the WTO

The environment needs an advocate before WTO trade and environ-

ment arbitration panels. There are two different institutional options for
performing this function. A more modest one involves the creation of a roster
of potential trade and environment lawyers who would be invited by the
WTO to participate on arbitration panels. A more ambitious option would
be to assign that function to the clustered MEA body.

Proximity and Distance

The institutional design model here is one of multilevel decentralized gover-

nance. Some activities still require old brick and mortar type institutions,
such as consolidated MEA secretariats which can coordinate intergovern-
mental activities and help foster intergovernmental trust. Other functions
or activities may be performed through a looser and more decentralized
network. A switchboard institution must be created to facilitate informa-
tion flow between the different networks and levels of actors engaging in
environmental governance. Such a centralized information coordination and
diffusion body could be small, so long as it is technologically sophisticated
and able to make use of current communications technologies to rapidly
transfer information between the various bodies engaged in performing these
key functions of global environmental governance. It could be largely virtual,
and it is not clear that it need be affiliated with the environmental policy
body. The major monitoring and verification functions would be coordi-
nated through the switchboard, and the findings would be circulated from
the switchboard. Yet the actual collection of information would be done
by diffused units around the world. These two sets of functions (coor-
dination information flow, and conducting or organizing monitoring and
verification) could either be concentrated in one or different formal institu-
Chapter 57 Page 205 September 11, 200820:12

Haas Addressing the Global Governance Deficit 205


We live in a new world. By taking advantage of a decentralized network

of governance functions, global governance may be improved, and the
prospects for achieving sustainable development advanced. By clinging to
models based on an obsolete, exclusively state centric model of governance,
claims of governance deficits will be exaggerated, and corrective designs
erroneously applied that neglect new political realities.
A new complex decentralized international governance system is
emerging from recent changes in the international systems. More actors now
engage in more governance functions at multiple levels of governance. While
such a wide array of decentralized responsibilities may appear disjointed, it
would be an error to conclude that decentralized governance is incoherent.
The task for the architects of the post cold war governance system is to
recognize and take advantage of the complex synergies between networks of
actors operating at multiple levels of international politics.


1. World Commission on Environment and Development 1987.

2. Held, McGrew, Goldblatt, and Perraton 1999; Hoffmann 2002; Keck and Sikkink
1998; Keohane and Nye 2000; Reinecke and Deng 2000; Schaeffer 1997; and Annan 1998,
3. For two somewhat complementary views of organized anarchy at the international level
see Bull 1977; and Oye 1986.
4. For an elaboration of these arguments see Haas and Kanie 2004.
5. Simon 1981; Perrow 1999; Axelrod 1997; Jervis 1997; and Haas 1975.
6. Coleman and Perl 1999; Florini 2000; Slaughter 2004; and Price 2003.
7. Haas and Haas 1995. For contemporaneous views see Skolnikoff and Kay 1972; Kennan
1970; and National Academy of Sciences 1972.
8. Downie and Levy 2000, 355375; Gehring and Buck 2002; and Holmberg 1992.
9. Dodds 2000; and Dowdeswell 2001.
10. DeSombre and Barkin 2002; and Williams 2001.
11. WBGU 2001; Biermann 2002, 2001, 2000; Charnovitz 2002; Esty 1994; and Esty
and Ivanova 2002. Websites include www.globgov.org, www.yale.edu/envirocenter; and
www.ias.unu.edu. For critiques see Najam 2003.
12. Aggarwal 1998; Ostrom 2001; and Ansell and Weber 1999.
13. Ruggie2001.
14. See Victor et al 1998 for studies of systems of environmental implementation and their


Aggarwal, Vinod K., ed. 1998. Institutional Designs for a Complex World. Ithaca, NY: Cornell
University Press.
Annan, Ko. 1998. The Quiet Revolution. Global Governance 4: 123138.
Ansell, Christopher K., and Steven Weber. 1999. Organizing International Politics. Interna-
tional Political Science Review (January).
Axelrod, Robert. 1997. The Complexity of Cooperation. Princeton, NJ: Princeton University
Chapter 57 Page 206 September 11, 200820:12

206 International Environmental Issues and Debates

Biermann, Frank. 2000. The Case for a World Environment Organization. Environment 42 (9):
. 2001. The Emerging Debate on the Need for a World Environment Organization.
Global Environmental Politics 1 (1) February: 4555.
. 2002. Strengthening Green Global Governance in a Disparate World Society. Interna-
tional Environmental Agreements 2: 297315.
Bull, Hedley. 1977. The Anarchical Society. New York, NY: Columbia University Press.
Charnovitz, Steve. 2002. A World Environment Organization. Columbia Journal of Environ-
mental Law 27 (2): 323362.
Coleman, William D., and Anthony Perl. 1999. Internationalized Policy Environments and
Policy Network Analysis. Political Studies XLVII: 691709.
DeSombre Elizabeth R., and J. Samuel Barkin. 2002. Turtles and Trade. Global Environmental
Politics 2 (1): 1218.
Dodds, Felix. 2000. Reforming the International Institutions. In Earth Summit 2002, edited by
Felix Dodds. London: Earthscan.
Dowdeswell, Elizabeth. 2001. Design for the Real World: Ideas for Achieving Sustainable Devel-
opment. Text of a speech delivered at Harvard University, 12 December 2001.
Downie, David L., and Marc A. Levy. 2000. The United Nations Environment Programme at
a Turning Point. In The Global Environment in the Twenty-First Century, edited by Pam
Chasek, 355375. Tokyo: UNU Press.
Esty, Daniel C. 1994. The Case for a Global Environmental Organization. In Managing the
World Economy, edited by Peter B. Kennen. Washington, D.C.: Institute of International
Esty, Daniel C., and Maria H. Ivanova, eds. 2002. Global Environmental Governance: New
Haven, CT: Yale School of Forestry & Environmental Studies.
Florini, Ann, ed. 2000. The Third Force. Washington, D.C.: Carnegie Endowment for Interna-
tional Peace.
Gehring, Thomas, and Matthias Buck. 2002. International and Transatlantic Environmental
Governance. In International Environmental Policymaking, edited by Matthias Buck,
Alexander Carius and Kelly Kollmann, 2143. Munich: Oekom Verlag.
German Advisory Council on Global Change (WBGU). 2001. World in Transition 2. London:
Haas, Ernst B. 1975. Is There a Whole in the Hole? International Organization 29 (3): 827876.
Haas, Ernst B., and Peter M. Haas. 1995. Learning to Learn. Global Governance 1 (3):
Held, David, Anthony McGrew, David Goldblatt, and Jonathan Perraton. 1999. Global Trans-
formations. Stanford, CA: Stanford University Press.
Hoffmann, Stanley. 2002. The Clash of Globalizations. Foreign Affairs July/August.
Holmberg, Johan, ed. 1992. Making Development Sustainable. Washington, D.C.: Island Press.
Jervis, Robert, ed. 1997. System Effects. Princeton, NJ: Princeton University Press.
Kanie, Norichika, and Peter M. Haas, eds. 2004. Emerging Forces in Environmental Gover-
nance. Tokyo: United Nations University Press, distributed in the United States by The
Brookings Institution Press.
Keck, Margaret E., and Kathryn Sikkink. 1998. Activists Beyond Borders. Ithaca, NY: Cornell
University Press.
Kennan, George. 1970. To Prevent a World Wasteland. Foreign Affairs 48 (3): 401413.
Keohane, Robert O., and Joseph S. Nye. 2000. Globalization. Foreign Policy Spring.
Najam, Adil. 2003. The Case Against a New International Environmental Organization. Global
Governance 9: 367384.
National Academy of Sciences. 1972. Institutional Arrangements for International Environ-
mental Cooperation. Washington, D.C.: National Academy of Sciences.
Ostrom, Elinor. 2001. Decentralization and Development: The New Panacea. In Challenges to
Democracy: Ideas, Involvement and Institution, edited by Keith Dowding, James Hughes
and Helen Margetts, 237256. New York: Palgrave Publishers.
Chapter 57 Page 207 September 11, 200820:12

Haas Addressing the Global Governance Deficit 207

Oye, Kenneth, ed. 1986. Cooperation Under Anarchy. Princeton, NJ: Princeton University
Perrow, Charles. 1999. Normal Accidents. Princeton, NJ: Princeton University Press.
Price, Richard. 2003. Transnational Civil Society and Advocacy in World Politics. World Poli-
tics 55 (4): 579606.
Reinecke, Wolfgang H., and Francis M. Deng. 2000. Critical Choices. Toronto: IDRC.
Ruggie, John G. 2001. global_governance.net: The Global Compact as Learning Network.
Global Governance 7: 371378.
Schaeffer, Robert K. 1997. Understanding Globalization. Lanham, MD: Rowman &
Simon, Herbert. 1981. The Sciences of the Articial. Cambridge, MA: MIT Press.
Skolnikoff, Eugene B., and David A. Kay, eds. 1972. World Eco-Crisis. Madison, WI: University
of Wisconsin Press.
Slaughter, Anne-Marie. 2004. A New World Order. Princeton, NJ: Princeton University Press.
Victor, David, Kal Raustiala, and Eugene B. Skolnikoff, eds. 1998. The Implementation and
Effectiveness of International Environmental Commitments. Cambridge, MA: MIT Press.
Williams, Marc. 2001. Trade and Environment in the World Trading System. Global Environ-
mental Politics 1 (4): 19.
World Commission on Environment and Development (WCED). 1987. Our Common Future.
New York: Oxford University Press.
Chapter 58 Page 208 September 11, 200820:12

Bridging the Trade-Environment Divide
Daniel C. Esty

rotection. For free traders, this word represents the consummate evil.

P For environmentalists, it is the ultimate good. Of course, for the

trade community, protection conjures up dark images of Smoot
and Hawley, while the environmental camp sees clear mountain streams,
lush green forests, and piercing blue skies. One cannot blame all of
the tensions at the trade-environment interface on linguistic differences,
but these competing perspectives are emblematic of a deep clash of cultures,
theories, and assumptions.
Trade officials often seek to limit efforts to link trade and environmental
policy-making, and sometimes to prohibit such efforts altogether. In this
regard, the narrow focus and modest efforts of the World Trade Organiza-
tions Committee on Trade and Environment are illustrative.1 The launch of
negotiations for a Free Trade Area of the Americas with an express decision
to exclude environmental issues from the agenda provides an even starker
example of the trade communitys hostility toward serious environmental
engagement. Economists have been prominent among those arguing that
pollution control and natural resource management issues are best kept out
of the trade policy-making process (Cooper, 1994; Bhagwati, 1999). Other
economists, however, have tried to set trade policy-making in a broader
context and to build environmental sensitivity into the international trading
system (Runge, 1994; Rodrik, 1997; Summers, 2000).
In fact, there is no real choice about whether to address the trade and envi-
ronment linkage; this linkage is a matter of fact. The only choice is whether
the policies put in place to respond will be designed openly, explicitly, and
thoughtfully, with an eye to economic and political logic or implicitly and
without systematic attention to the demands of good policy-making. This
article seeks to explain why trade liberalization and environmental protec-
tion appear to be in such tension and to push economists to explore more
aggressively what economic theory and practice might do to address the
concerns being raised.

Source: Journal of Economic Perspectives, 15(3) (2001): 11330.

Chapter 58 Page 209 September 11, 200820:12

Esty Bridging the Trade-Environment Divide 209

Trade and Environmental Linkages

Potential Conflicts between Domestic Regulations and Trade

In recent years, the focus of trade liberalization has shifted from lowering
tariffs, which have come down considerably around the world, to the elim-
ination of nontariff barriers to trade (Jackson, 1992). Since many kinds of
domestic regulations can potentially be construed as nontariff barriers, the
extent and impact of the market access commitment and other regulatory
disciplines negotiated in the trade domain has expanded.
A number of the most prominent international trade disputes in the
last decade have concerned the clash between domestic regulations and
trade rules. In the well-known tuna-dolphin case, the United States banned
Mexican tuna imports because the fishing methods resulted in incidental
dolphin deaths. In 1991, Mexico obtained a GATT panel decision declaring
the United States to be in violation of its GATT obligations for imposing such
a ban. In the ongoing beef hormone dispute, the European Union has refused
to adjust its no added hormones in beef food safety standards despite a
series of WTO rulings that its regulations had no scientific foundation and
were in contravention of the rules of international trade. The U.S. sanctions
against Thai shrimp caught using methods that killed endangered sea turtles
were recently deemed to be GATT-illegal. Trade and environment friction
can be found outside the WTO, as well. Witness the enormous effort that
the European Union has put into harmonizing environmental standards over
the past several decades (Vogel, 1994).
There is no end in sight to trade and environment cases. If anything, the
number of disputes seems to be rising (Sampson, 2000). As global economic
integration intensifies, so does the potential for conflict (Lawrence et al.,
1996; Dua and Esty, 1997). Public health standards, food safety require-
ments, emissions limits, waste management and disposal rules, packaging
and recycling regulations, and labeling policies all may shape trade flows.
Trade disciplines may also affect national-scale environmental efforts, espe-
cially to the extent that WTO dispute settlement procedures are used to
challenge pollution control or natural resource management programs.
Thus, while fearmongering about lost sovereignty (Perot, 1993;
Wallach and Sforza, 1999) can be dismissed, the suggestion that trade liber-
alization constrains regulatory flexibility rings true. With new issues like
biotechnology and climate change emerging, the potential for significant
and divisive battles between trade policy and regulatory choices including
environmental rules looms large.

Increasing Trade, Economic Growth, and Environmental Risks

The literature on the interaction between economic growth and pollution

points to what has been called an environmental Kuznets curve. The Kuznets
curve is a inverted-U relationship which shows that environmental conditions
Chapter 58 Page 210 September 11, 200820:12

210 International Environmental Issues and Debates

tend to deteriorate in the early stages of industrialization and then improve

as nations hit middle-income levels, at a per capita GDP of about $5000 to
$8000 (Grossman and Krueger, 1993, 1995; Shafik and Bandyopadhyay,
1992; Seldon and Song, 1994). Since the primary purpose of liberalizing
trade is to increase economic growth, trade unavoidably affects the level
of environmental protection through its impact on the Kuznets curve.
A first concern stemming from the Kuznets curve is that air and water
pollution problems tend to worsen in the early stages of development. Many
developing countries are living through the part of the Kuznets curve in
which environmental conditions deteriorate. In addition, some problems,
especially those that are spread spatially or temporally (such as greenhouse
gas emissions), do not yet appear to have reached the downward-sloping
part of the Kuznets curve in any country. This empirically derived pattern of
ongoing deterioration perhaps reflects the fact that, absent reciprocity, the
benefit-cost ratio for policy interventions in response to diffuse problems are
always negative from a national perspective.
A second concern is that even if expanded trade and economic growth
need not hurt the environment, there is no guarantee that it will not
(Harbaugh, Levinson and Wilson, 2000; Hauer and Runge, 2000). The
effects of economic growth on trade can be broken down into three effects.
Technique effects arise from the tendency toward cleaner production
processes as wealth increases and trade expands access to better technologies
and environmental best practices. Composition effects involve a shift in
preferences toward cleaner goods. Scale effects refer to increased pollution
due to expanded economic activity and greater consumption made possible
by more wealth (Grossman and Krueger 1993; Lopez, 1994). Thus, the
claim that growth improves the quality of environment can be rephrased as a
claim that, above a certain level of per capita income, technique and compo-
sition effects will outweigh scale effects. Empirical evidence on the relative
sizes of these effects is limited. But at least some of the time, it appears that
expanded trade may worsen environmental conditions (Antweiler, Copeland
and Taylor, 1998).
Finally, the odds that increased trade will have net negative environmental
impacts rise if resources are mispriced (Anderson, 1998; Panayotou, 1993).
Around the world, many critical resources like water, timber, oil, coal, fish,
and open space are underpriced (or overpriced) (World Bank, 1997; Earth
Council, 1997). Even the WTO acknowledges in its most recent Trade and
Environment Special Report that expanded trade can exacerbate pollution
harms and natural resource management mistakes in the absence of appro-
priate environmental policies (Nordstrom and Vaughan, 1999).

Transboundary Externalities

Transboundary pollution spillovers make attention to trade-environment

linkages a matter of normative necessity as well as descriptive reality.
Perhaps the most discussed issues involve emissions of ozone-layer depleting
Chapter 58 Page 211 September 11, 200820:12

Esty Bridging the Trade-Environment Divide 211

chlorofluorocarbons and greenhouse gases, which threaten global climate

change. But recent advances in tracing the movement of pollutants have
also demonstrated long-distance impacts from particulates (Grad, 1997),
sulfur dioxide and other precursors of acid rain (Howells, 1995), DDT and
other pesticides (Lawler, 1995; Rappaport et al., 1985), mercury and other
heavy metals (Fitzgerald, 1993), and bioaccumulative toxics (Francis, 1994).
Other transboundary issues involve rules governing shared resources such
as fisheries in the open ocean and biodiversity.
The need to control transboundary externalities makes trade-environment
linkages essential from the point of view of good economic policy-making.
After all, uninternalized externalities not only lead to environmental degra-
dation, but also threaten market failures that will diminish the efficiency
of international economic exchanges, reduce gains from trade, and lower
social welfare. National governments, no matter how well intended, cannot
address inherently international problems such as climate change or fisheries
depletion unilaterally. A functioning Global Environmental Organization,
operating in parallel with the trading system, might be a first-best policy
option in response to these challenges (Esty, 2000a). But no such regime
exists. Thus, the World Trade Organization along with regional trade agree-
ments cannot avoid some shared responsibility for managing ecological inter-

The Political Economy of Trade Liberalization

Taking environmental issues seriously must also be understood as a political

necessity for free traders. Forward momentum in the trade realm is difficult
to sustain (Bergsten, 1992). In this regard, the trade community cannot risk
diminishing further the already narrow coalition in favor of freer trade, espe-
cially in the United States. Dismissing environmental concerns, which results
in broad environmental community opposition to trade agreements, gener-
ates unnecessary and avoidable political resistance to liberalized trade (Esty,
Certain environmentalists will always be opposed to trade liberalization
because they adhere to a limits to growth philosophy. But the environ-
mental community is neither monolithic nor uniformly protectionist. Many
mainstream environmentalists believe in sustainable development and will
support freer trade if they feel that pollution and natural resource manage-
ment concerns are being taken seriously. For example, the congressional vote
in favor of the NAFTA depended critically on the fact that a number of envi-
ronmental groups came out in favor of the agreement, which translated into
support from politicians who define themselves as both pro free trade and
environmentally oriented (Audley, 1997). Concomitantly, the several recent
failures to obtain a majority for new fast track negotiating authority can
be attributed to this swing group voting against the legislation because the
proposals lacked environmental credibility (Destler and Balint, 1999).
Chapter 58 Page 212 September 11, 200820:12

212 International Environmental Issues and Debates

In practice, moreover, there is no empirical support for the suggestion

that environmental linkages detract from trade agreements or trade liberal-
ization. The North American Free Trade Agreement, often considered the
greenest trade pact ever, contains a number of environmental elements and
was adopted with an Environmental Side Agreement. There is no evidence
that these provisions have in any way diminished the post-NAFTA U.S.-
Canada-Mexico trade flows (Araya, 2002; Hufbauer et al., 2000).
One might argue that this political analysis has little to do with
economists role in the trade and environment debate. To the contrary, if the
arguments of economists become disconnected from the reality of political
pressures and policy imperfections, then economic logic is unlikely to prevail
in trade policy-making.

The Arguments for Separating Trade and Environmental Policy

While many no linkage economists and trade officials understand the

arguments for taking up environmental issues in the trade context, they fear a
scenario in which protectionist wolves find their way into the trading system
in environmental sheeps clothing (Bhagwati, 1988; Subramanian, 1992).2
The sight at the 1999 WTO Ministerial Meeting in Seattle of green activists
marching arm-in-arm with avowed protectionists confirmed for many, espe-
cially in the developing world, the suspect motives of those advancing the
environmental agenda.
A related argument for keeping the environment out of the WTO turns
on the fear that trade liberalization will grind to a halt under the weight of
environmental burdens. Why, ask trade economists, must trade measures be
used to enforce international environmental agreements? Shouldnt environ-
mental policy problems be solved with environmental policy tools? Those
who wish to separate trade and environmental policy-making also fear that
high-income countries will impose lofty environmental standards on low-
income countries, depriving them of one aspect of their natural comparative
advantage and subjecting them to trade barriers if they fail to perform up to
developed country standards (Bhagwati, 1999; 2000).
But while these worries have some basis in reality, they do not provide
a justification for complete separation of trade and environmental policies.
Certainly, environmentalism should not be used as a cover to disguise trade
barriers. Certainly, the tactical partnerships of some environmental groups
have been misguided. Certainly, better environmental regulation at both
the national and global levels could markedly reduce trade-environment
tensions.3 Certainly, global-scale environmental efforts should not mean a
reduction in the standard of living for people in low-income countries.
But these are not arguments for ignoring the inescapable linkages between
trade and the environment. They are arguments for trying to integrate trade
and environmental policies in sensible ways. The following sections discuss
key areas for research and policy analysis that could help to narrow the
divide between trade and environmental policy goals and practices. The next
Chapter 58 Page 213 September 11, 200820:12

Esty Bridging the Trade-Environment Divide 213

section focuses on strengthening the foundations of environmental policy,

while the next two sections focus on issues of economic theory and trade

Strengthening Environmental Policy Foundations

A battle rages among environmentalists over how best to address (and even
understand) environmental challenges. Many environmentalists support the
concept of sustainable development (World Commission on Environ-
ment and Development, 1987) and believe that economic growth can,
if managed properly, support environmental improvements. A significant
number of environmental advocates remain committed, however, to a
limits to growth paradigm in which trade liberalization contributes to
more economic activity and therefore more pollution and unsustainable
consumption of natural resources (Meadows et al., 1972; Daly, 1993). But
even those who find the promise of sustainable development attractive worry
that, in practice, environmental policy tools are not up to the pressures of
Economists are likely to have little in common with the advocates of lower
consumption levels, especially when the burdens of such a policy choice
would fall most heavily on those in the poorest countries of the world. But
economists can play a role in answering certain persistent environmental
research and policy questions which could, in turn, help to expand the
common ground between free traders and environmentalists.

Clarifying Concerns about Sustainable Development

Sustainable development has proven hard to define and even harder to put
into practice. It is clear that poverty can force people to make short-term
choices that degrade the environment, like cutting down nearby trees for
firewood despite the likelihood of future soil erosion. But the hope that
trade liberalization will lead to economic growth that will alleviate poverty
and generate resources for environmental investments sometimes seems to
rely on a tenuous chain of events which may well unravel under real-world
It is useful to examine these issues in terms of the inverted-U environ-
mental Kuznets curve discussed earlier, which shows a general pattern of
increasing environmental degradation up to a certain level of per capita GDP
and environmental improvements beyond that point. Environmentalists will
always be worried about societies which are living through the portion of
the Kuznets curve where growth is accompanied by environmental degrada-
tion, even if it can be shown that people are receiving other welfare gains.
Economists could, however, significantly bridge the gap with green groups if
they were to find ways to reduce the duration and intensity of environmental
deterioration as low-income countries grow to middle income. Economists
Chapter 58 Page 214 September 11, 200820:12

214 International Environmental Issues and Debates

might also confirm that ignoring pollution altogether until middle income
levels are reached is a serious policy mistake. Some environmental invest-
ments, like protecting drinking water or siting polluting factories downwind
of urban areas, have such high benefit-cost ratios that even the poorest coun-
tries should undertake them.
As regards the portion of the environmental Kuznets curve in which
growth and environmental quality are both improving, many mainstream
environmentalists express concerns that either rising wealth or increased
population will drive up consumption in ways that undermine prospects
for sustainable development. Both economic theory and recent empir-
ical evidence could help to assuage these apprehensions. Development
economists have demonstrated that population growth diminishes with
wealth. Economists might do more to demonstrate that poverty alleviation
is critical for population control, which in turn offers significant potential
environmental benefits. More generally, the economics field has had little
to say about how to minimize scale effects and maximize the chances that
growth will improve environmental quality.
Finally, as noted earlier, certain environmental harms do not appear to
diminish with increases in income. Carbon dioxide emissions, for instance,
continue to rise, albeit at a decreasing rate, as GDP per capita goes up. It
may be that, even for carbon dioxide emissions, the downward portion of
the environmental Kuznets curve would be reached at some income level,
but no society has achieved the exalted wealth required. If or until that
occurs, economists could gain credibility by agreeing that wealth is not an
environmental cure-all.
The common theme in this discussion is that the environmental Kuznets
curve need not be destiny. The present shape of the curve, as estimated from
historical experience, reflects a political economy interaction among trade,
growth, and the environment. Trade has a positive effect on the environ-
ment (and perhaps a net welfare benefit more broadly) only if environmental
policy advances alongside trade liberalization (Anderson, 1992, 1998; Esty,
1994). However, institutional failures in the environmental realm often mean
that the requisite strengthening of environmental performance in parallel
with trade liberalization may not occur (Chichilnisky, 1994; Zhao, 2000).
In this regard, economists should take more seriously the need to find policy
strategies that lead to a shorter and flatter Kuznets curve.

Disciplining Free-Riders

Economists and environmental policymakers generally agree on the wisdom

of enforcing the polluter pays principle, which holds that those who
cause environmental degradation should bear the costs. But as a matter of
policy, this goal remains elusive. While economists have demonstrated the
value of market-based environmental strategies, they have by and large not
managed to convince the environmental and political worlds that pollution
Chapter 58 Page 215 September 11, 200820:12

Esty Bridging the Trade-Environment Divide 215

fees, emissions allowances, or other economic incentives will work in prac-

tice. Environmental policy remains underdeveloped in terms of economic
sophistication and largely mired in command and control approaches.
The collapse of the international negotiations over climate change, in part
because of disputes over how far to go in using market mechanisms, demon-
strates the persistence within the environmental policy community of anti-
economics sentiment.
Figuring out how to enact policies that embody the polluter pays prin-
ciple becomes even more difficult when the scope of the environmental harm
is broader than the vista of the regulators. Dua and Esty (1997) argue that
super-externalities, which spill beyond the defined jurisdiction of regu-
latory authorities in either space or time aggravate the collective action
problem.4 A small number of scholars have looked at the spatial distribution
of issues in the trade domain (Krugman, 1991; Bloom and Sachs, 1998) and
at the geographic dimensions of the trade and environment problem (Hauer
and Runge, 2000; Esty, 1994), but more work needs to be done in the realm
of economic geography.
Transboundary environmental spillovers create a risk of allocative
inefficiency and market failure in the international economy. Some mech-
anism for promoting collective action and for disciplining free riders is
therefore required (Baumol and Oates, 1988). Whether free traders like it
or not, trade measures are one potential candidate for this function. Admit-
tedly, trade sanctions are imperfect, costly to those who impose them, and
may backfire. But at least in some cases, trade penalties have worked (Brack,
1996; Barrett, 1997). Moreover, better tools to discipline free riders in the
international environmental domain do not seem readily available. As envi-
ronmentalists point out, the weakness of the extant global environmental
regime cannot be wished away nor dismissed as irrelevant to the question of
how environmental goals get squared with the trade liberalization agenda.
There are a number of issues to be investigated which could shed light
on the use of trade policy as a tool for enforcement of environmental stan-
dards. First, refined theory on the use of trade measures to support environ-
mental cost internalization in the international realm is needed, advancing
the preliminary analyses of Charnovitz (1993), Chang (1995), and Barrett
(1997). Second, more work to find ways to strengthen the international
environmental regime, which could relieve the pressure on the World Trade
Organization to play a major environmental role, would be useful (Esty,
1994, 2000a). Such work might build on efforts to investigate the political
economy of environmental protection (for example, Keohane, Revesz and
Stavins, 1998). Third, the advantages and disadvantages of policy linkages
need to be more fully explored. Concerns are sometimes expressed that if
trade policy becomes entangled with environmental policy, either or both
sets of policies may be unable to advance. Yet the potential benefits of cross-
issue policies and trade-offs have been repeatedly demonstrated (Haas, 1958;
Carrero and Siniscalco, 1994). Finally, those who wish to limit the trade
Chapter 58 Page 216 September 11, 200820:12

216 International Environmental Issues and Debates

systems role in enforcing international environmental agreements would

find their case greatly strengthened if they could point to workable alterna-
tive enforcement mechanisms.5

Refining Trade Theory

Environmental perspectives on trade often clash with the settled views of

economists. Frequently, the problem reflects a degree of economic misun-
derstanding by those in the environmental community. But often, there is
a kernel (or more) of truth in the environmental position with which the
economic community has failed to grapple. In these areas, there are intriguing
research opportunities for economists.

Level Playing Fields

Environmentalists often worry that expanded trade will lead to competitive

pressures which will push down environmental standards. They fear a regu-
latory race toward the bottom as jurisdictions with high environmental
standards relax their rules so as to avoid burdening their industries with
pollution control costs higher than competitors operating in low-standard
jurisdictions. Thus, they call for harmonization of pollution control regula-
tions at stringent levels, the imposition of eco-duties on those with subpar
rules, or other policy interventions to level the playing field.
Economists point out that the existence of divergent circumstances, inclu-
ding variations in societal preferences about the optimal level of environ-
mental protection, is what makes gains from trade possible. If environmental
rules vary because of differences in climate, weather, geography, existing
pollution levels, population density, risk preferences, level of development,
or other natural factors, the variation in standards should be consid-
ered welfare-enhancing and appropriate. Clearly, a sweeping presumption in
favor of uniform standards fails to grasp the insight of comparative advan-
tage and makes no sense (Burtless et al., 1998). More generally, economists
tend to find arguments in favor of regulatory harmonization in a context of
economic integration unpersuasive (Bhagwati, 1996, 2000).
Diversity in circumstances generally makes uniformity less attractive than
standards tailored to the heterogenous conditions that exist (Mendelsohn,
1986; Anderson, 1998). But not always. Divergent standards across jurisdic-
tions may impose transaction costs on traded goods that exceed any benefits
obtained by allowing each jurisdiction to maintain its own requirements.
Sykes (1995, 1999) has demonstrated that market forces will tend, over
time, to eliminate such problems. Vogel (1994) argues, in fact, that upward
harmonization (a race to the top) often occurs. But this logic only applies
to product standards, and standards that relate to production processes or
methods are not subject to the same market pressures.
Chapter 58 Page 217 September 11, 200820:12

Esty Bridging the Trade-Environment Divide 217

Some theoretical work has been done to try to understand the different
harmonization dynamics (Bhagwati and Hudec, 1996; Esty and Geradin,
1998, 2001), but more would be useful, as would empirical evidence on what
happens to environmental standards in the process of trade liberalization.
For example, how often do free trade agreements include commitments to
lower environmental standards and how often to higher standards?
Environmentalists also fear that the rules of international trade are biased
against their interests. They believe that within the trading system both
WTO and regional trade agreements free trade principles always trump
other policy goals such as environmental protection. Some recent analyses
suggest that such a tilt in GATT jurisprudence might once have existed, but is
now less pronounced (Charnovitz, 2000; Wofford, 2000). Efforts to illumi-
nate the facts might diminish fears that trade liberalization runs roughshod
over environmental issues. Some efforts have been made in this regard
(Trachtman, 2000; Burtless et al., 1998), but more would be welcome.

Psychological Spillovers and Ethical Preferences

Most economists acknowledge, at least in theory, that transboundary

pollution externalities need to be addressed, but economists tend to be skep-
tical about claims of psychological spillovers (Blackhurst and Subramanian,
1992; Cooper, 1994). What are we to make of complaints about environ-
mental degradation in China or campaigns to save the rain forest? As long
as the harms are localized, shouldnt environmental policy choices (even
mistakes) in other jurisdictions be accepted? Maybe so from a perspec-
tive of economic theory, but most people do not see the world this way.
The fact that Chinese workers produce goods under adverse environmental
conditions is not celebrated, even if the low standards in China translate into
cheaper products in export markets. Why not?
Perhaps economists assume a utilitarianism that is oversimplified. Sen
(1977) and others have noted the narrow behavioral assumptions on which
most of economics builds, ignoring human realities such as the existence
of interdependent welfare functions. In fact, many people consider them-
selves, at least to some extent, to be part of a global community. In addition,
economists may too readily accept as a given that the policy choices in places
like China are locally optimal and do not stop to ask whether Chinese envi-
ronmental standards truly reflect the will of the people.6 By gliding past
choice of public questions (Esty, 1996), economists simplify their models
but diminish the policy traction of their arguments.
Environmentalist concerns about extraterritorial policy choices frequently
seem to be paternalistic or even imperialistic. Green groups often think that
they know better than the people or governments of other countries, especially
developing nations, what constitutes the right environmental standard or
policy program. Economists have been quick to condemn those who are
keen to impose their own ethical preferences on others, using trade sanctions
to induce or coerce acceptance of such preferences (Bhagwati, 1993).
Chapter 58 Page 218 September 11, 200820:12

218 International Envir