Vous êtes sur la page 1sur 19


Water Resources and Economics xxx (2017) 1​–​10

Contents lists available at ​ScienceDirect

and Economics
j o u r n a l h o m e p a g e : ​w w w . e l s e v i e r . c o m / l o c a t e / w r e

A resource-based view of utilities: The key-determinant factors for

customer prices and organizational costs in the Portuguese water

Hugo Consci​^​encia Silvestre a​​ ,​ *​​ , Ricardo Corr​^​ea Gomes b​

Center for Policy ​&​ Public Administration, Universidade da Integraçao​~​ Internacional da Lusofonia Afro-Brasileira (UNILAB), Campus
da Liberdade, Avenida da Aboliçao,​~​ 3 ​–​ Centro, CEP: 62.790-000, Redençao,​~​ CE, Brazil

University of Brasília, Faculdade de Economia, Administraçao​~​ e Contabilidade (FACE), Campus Darcy Ribeiro, Ala Norte, Subsolo,
Modulo 25, CEP: 70910-900, Brasília, Federal District, Brazil


The resource-based view supports the idea that ​fi​rms that make better use of
Resource-based view their existing resources have superior performance, which depends on
organizational processes and routines to lower organizational costs and
Competitive advantage prices to customers. By applying this approach to the Portuguese water
industry based on a cross-sectional research strategy, the ​fi​ndings
Customer prices demonstrate that public ​fi​rms' with lower organizational costs are able to
charge lower prices. There are two implications ​of the current ​fi​ndings. The
Organizational costs fi​rst is political in nature and involves the discussion of the best way to
improve the provision of public services by utilities. From an economic,
Portugal managerial and social approach, all the arrangements, whether public or
private, should be rethought. The second implication is theoretical, since
fi​rm effects can be established based on the existence of superior resources,
the key variable to explain differences in organizational performance.
1. Introduction

Under what conditions can water utilities achieve sustained competitive advantage? (see Ref. ​[1]​. This question results from a theoretical
discussion about management strategy that considers the competitive advantages of a ​fi​rm in comparison with its competitors. In this particular,
the resource-based view highlights the internal resources as the major drivers of organizational performance ​[2]​. Internal resources can be
measured as the ​“fi​rm's management skills, its organizational processes and routines, and the information and knowledge it controls​” ​[1]​; p. 625).
Firms in possession of these superior resources will be able to decrease their production costs ​[3]​, thus increasing their competitive advantage.

An ongoing theoretical discussion exists about which factors are associated with ​fi​rms' superior performance. As far as we know, this
theoretical discussion does not include utilities, especially water companies ​[4]​. An attempt was carried out by Ref. ​[5]​; but his focus was a
natural resource based view of ​fi​rms and the in​fl​uence on ​fi​rms' performance. Regarding water utilities, several studies have been conducted to
understand which contextual factors are related to water consumption through pricing schemes, but very few have examined the factors that
in​fl​uence customer prices ​[6​–​12]​; as examples). Therefore, studies in this sector still need to be developed to provide insights for practitioners
and politicians.

For those reasons, this study's main aim is to understand which factors are related with residential customer prices and

* Corresponding author.

E-mail addresses: ​hmcsilvestre@gmail.com​ ​(H.C. Silvestre), ​gomesric.rg@gmail.com​ ​(R.C. Gomes).


Received 16 February 2017; Received in revised form 4 August 2017; Accepted 7 September 2017
Available online xxxx

2212-4284/​©​ 2017 Elsevier B.V. All rights reserved.

Please cite this article in press as: H.C. Silvestre, R.C. Gomes, A resource-based view of utilities: The key-determinant factors for customer
prices and organizational costs in the Portuguese water industry, Water Resources and Economics (2017), http://
H.C. Silvestre, R.C. Gomes Water Resources and Economics xxx (2017) 1​–​10

organizational costs for Portuguese water utilities. As an innovation, this study applies a resource-based view approach, to shed light on this
phenomenon from a different perspective ​[53] ​and bene​fi​ting from cross-fertilization that sustains future decisions. Applying a cross-sectional
research design ​[13]​, customer prices and organizational costs for 2013 are the dependent variables in this study. As explanatory variables,
institutional and governance factors such as organizational costs, ownership structure, management model and economic heterogeneity were
considered, while socio-demographic, political and local government economic and ​fi​nancial results were selected as predictors for measurement
of the resource-based view and structure-conduct-performance, respectively.

This manuscript is organized as follows: ​fi​rst, the literature concerning strategic management is discussed, with details of the resource-based
view; then predictors for user prices and organizational costs are presented and debated. After that, details of research design and variables are
discussed. Lastly, the results and conclusions are presented.

2. The resource-based view for the water industry

For Penrose ​[14]​; ​fi​rms are distinct due to their internal resources. The differentiation of resources and their application is what makes ​fi​rms
stand apart from each other. Starting with strategic decision-making, ​fi​rms' are able to better apply their resources to reach competitive advantage
in light of organizational returns. The main emphasis of the resource-based view is to understand competitive heterogeneity, which ultimately
will in​fl​uence ​fi​rms' performance ​[15]​. This approach seeks, grounded on different organizational resources, to explain performance of competing
fi​rms ​[16]​. Such ​fi​rm effects can be established based on the existence of superior resources, the key variable to explain differences in
organizational performance. When superior resources are properly applied, they will generate lower production costs and/or higher bene​fi​ts ​[17]​.

In the water sector, private involvement has been considered as adding value by promoting organizational ef​fi​ciency and cost savings ​[18]​.
According to this view, the private sector has superior resources in relation to other institutional arrangements. These resources ​result from a
for-pro​fi​t management perspective, which ultimately aims to make better use of scarce resources.

Based on business policy roots, it is believed that organizations are different in their attributes. Those attributes are heterogeneous, since it
“​implies that some ​fi​rms have resources that generate more value than others​” ​[3]​; p. 317). In that respect, their resources and internal capabilities
must be considered for successful strategic management of the ​fi​rm ​[19]​. Those resources, however, need to be rare, unable to be copied easily
and superior to others to lead to a competitive advantage. The assets can be tangible and intangible, ​“​including a fi​rm's management skills, its
organizational processes and routines, and the information and knowledge it controls​” ​[1]​;​ ​p.​ ​625).

For the water sector, private operators are viewed as more ef​fi​cient since in theory they have the ability to improve their operational results
[20]​. In turn, for public operators, one of the greatest problems is the need to follow bureaucratic rules, strongly in​fl​uencing the processes and
routines of these organizations that are invariably considered inef​fi​cient ​[51]​.

Competitive advantage comes from economic and productive ef​fi​ciency and customer satisfaction ​[21]​. In theory, making better use of scarce
resources will bene​fi​t customers through lower prices. Even though organizational resources and internal capabilities are crucial factors for
organizational strategy, managers cannot forget external factors. Among those factors are ​“​changes in demand, science and technology,
availability of raw materials, and government policy,​”​ which in​fl​uence organizational resources and internal capa-bilities ​[22]​; p. 1004).

In sum, the institutional and governance factors where the ​fi​rm operates need to be considered for strategic decisions. This is to say that
internal factors are crucial, but other factors must not be forgotten, especially in the water industry ​[6]​.

3. A review of institutional and contextual factors in the water industry

In this section, studies of the water supply industry are highlighted, taking into account the resource-based view in order to understand which
factors are related with water ​fi​rms' performance. According to the resource-based view, superior resources within a ​fi​rm will lead to lower
production costs and ultimately lower customer prices: “​In the majority of cases, the operational costs of water ​companies form the ​‘​indicator​’ for
de​fi​ning prices​”​ ​[23]​; p. 224).
The 1980s brought many changes in assessing the performance of utility companies. Legislators introduced ​“​explicit standards and measures
of performance​” while stressing the ​“​discipline and parsimony in the resource use​” ​[24]​; pp. 4​–​5) in order to improve utilities' performance.
Major structural reforms, such as privatization, were accomplished to raise competition and ultimately lead to lower prices ​[25]​.

3.1. Institutional and governance factors

Organizational performance can be in​fl​uenced due to particular institutional and governance characteristics, which are assumed to differ in
each organization ​[6]​. According to Carter ​[26]​; ownership can in​fl​uence ​fi​rms' performance, since public and private com-panies can produce
different output levels. In that sense, and in theory, private companies tend to perform better than public organ-izations ​[27]​. Since private ​fi​rms'
aim to raise pro​fi​ts, they will try to lower their operational costs in order to improve their revenues ​[20]​. ​However, in comparison with other
variables, according to Carvalho and Marques ​[11]​; ​ownership does not seem to in​fl​uence ​fi​rms' ef​fi​ciency in the Portuguese water industry.
Similar results were found by Romano et al. ​[28] ​in ​Italy. By itself, ownership was not ​a determinant factor for water consumption, but
consumption levels were higher under public utilities, while customer prices were lower. In Spain, customer prices are also higher when the
operator is a private company and lower when it is a municipally owned utility ​[29]​.

H.C. Silvestre, R.C. Gomes Water Resources and Economics xxx (2017) 1​–​10

Nowadays, it is not just improvement in ef​fi​ciency that is demanded, since the political and managerial logic has changed, from public value
to sale of a commodity service. As Bakker ​[27] ​argues, water supply involves a market and business approach to the provision of a scarce
resource on a full cost-recovery basis. Bearing in mind these arguments, the author stresses the need to promote public ownership of water
utilities in England and Wales in order to improve sustainability and equity.

Based on the discussion about the ownership of public utilities, new institutional arrangements for public sector organizations have been
introduced without resorting to privatization ​[30]​. For this purpose, public agencies are allowed to perform like private ones, in the belief that if
public-sector organizations are managed in accordance with private law, they will be able to more easily decide about the resources necessary to
improve their performance. These arrangements are described as single-purpose bodies which are ​fi​nancially and legally independent from local
government structures, thus having greater political decision-making independence ​[9]​. They are also dependent on user fees instead of budget
transfers from local government, increasing the need for managerial responsibility. Lowering organizational costs is crucial to keeping customer
prices low. In sum, public agencies emerged as a way of enhancing ​fi​nancial and economic performance, the assumption being that public-sector
organizations can be as ef​fi​cient and productive as private ​fi​rms, but only if political interference is minimized and management is
professionalized. Moreover, selecting public-sector arrangements avoids private sector commercial incentives to abuse market position ​[30]​. In
fact, it seems that customer prices are lower under public-sector organizations ​[8]​. However, political interference in setting customer prices is
still a major concern ​[31]​. In opposition, García-Valinas​~​ et al. ​[29]​ ​found that under the guidance of Spanish townships, customer prices are
lower. However, public agencies charge more than townships and even private ​fi​rms, which ​“​could be due to public utilities not being managed
as ef​fi​ciently as private ones.​”​ Another interesting ​fi​nding from this study is the existence of contractual public-private partnerships, which
promote lower customer prices in comparison with purely public agencies and private organizations.

Besides ownership and the adopted organizational arrangement, economic heterogeneity also needs to be considered ​[26]​. According to
Carvalho and Marques ​[11]​; it is possible to identify scope economies for this particular industry. Scope economies can be found when different
services are performed simultaneously, such as water supply and sewer services. Adding sewage collection and treatment to the delivery of water
can facilitate ef​fi​ciency gains due to the joint production. Through the analysis of the Portuguese water industry, Carvalho and Marques ​[32] ​fi​nd
that when combining water delivery and sewage collection and treatment, ef​fi​ciency gains exist in comparison with those organizations that only
supply water. In other words, if the company only delivers water to its customers, without any other services, then the economic heterogeneity
level is low. In contrast, if the company, besides water delivery, also deals with the collection and treatment of sewage, the economic
heterogeneity level will be medium ​[26]​.

3.2. Demographic factors

Demographic factors, such as population density, seem to be one of the most important factors, due to the in​fl​uence on water consumption
levels and thus costs for production ​[33]​. That can be seen in Spain, where population density is positively associated with customer prices.
Contrary to expectations, customer prices were found to be higher in more populated areas ​[29]​. According to Carvalho and Marques ​[11]​; the
density of customers and the number of inhabitants in​fl​uences ​fi​rms' performance in the Portuguese water industry. Additionally, the size of the
supplied area also affects organizational costs, where larger areas increase production costs (see Ref. ​[34]​. In sum, larger areas with lower
population density will tend to raise water delivery costs.

According to Marques and De Whitte ​[35]​; the existence of scale economies (based on the delivery of water) are supposed, in theory, to be
ef​fi​ciency facilitators if the water volume, the number of customers and supplied areas are considered. By themselves, scale economies can
facilitate the improvement of ef​fi​ciency levels due to the reduction of production costs per delivered unit. The main ​fi​ndings of these authors'
study point to improved ef​fi​ciency levels under scale economies. However, according to Chettri and Ven-katesan ​[37]​; economies of scale can
drop when other resources are being used ef​fi​ciently. After reaching an optimum output, ​fi​rms can face higher production costs and decreasing
scale economies. The lesson to be drawn relies on the existence of other factors that in​fl​uence the cost of the service, e.g., ​“​cost-of-service studies
generally support lower rates for larger volume utility customers (both energy and water), based on operational economies and favorable load
factors​”​ ​[8]​; p. 42). However, García-Valinas​~​ et al. ​[29]​ ​found no statistical difference between scale economies and customer prices in Spain.

3.3. Social factors

Customers' socioeconomic characteristics are potential contextual factor and they can in​fl​uence the ​fi​rm's production costs ​[38]​. In that way,
customers' purchasing power can lead to higher consumption levels while providers can lower their organizational production costs. In Italy, for
example, greater water consumption levels are positively linked with higher income per capita, since higher pur-chasing power raises water
consumption ​[28]​. According to Marques et al. ​[12]​; serving areas with higher purchase power levels allows the organization to improve its cash
fl​ow, which ultimately enables system upgrade and consequently the investment in operational ef​fi​ciency. In opposition, low-income households
have lower water consumption levels, especially when the European directive demands full cost recovery, jeopardizing the access to water,
which is problematic from a social perspective. Thus, public policies need to be developed to target low-income households ​[39]​.
Furthermore, the number of inhabitants with no labor earnings (unemployed) can have a relationship with the consumption levels that can
in​fl​uence organizational costs. Bakkers ​[27] ​mentions that in England and Wales, low-income households spend an average of four percent of
their weekly budget on water consumption, while the national average is around one percent.

H.C. Silvestre, R.C. Gomes Water Resources and Economics xxx (2017) 1​–​10

3.4. Political and local government ​fi​nancial sustainability factors

According to Niskanen ​[40]​; politicians' main goal is to be re-elected. In order to achieve this, politicians will not support cutting public
services ​[25]​. In this particular, the number and effective participation of the electorate can in​fl​uence user prices when rates are decided on
political grounds instead of managerial ones ​[7]​. This is due to politicians' desire to win or remain in of​fi​ce, which does not allow establishment
of real and effective organizational cost recovery based on customer prices ​[34]​.

Through analysis of Austrian water pricing policy, Klein ​[41] ​found that European directives are clear when dealing with estab-lishment of
water rates. In theory, water utilities' prices should follow the full cost recovery principle, but deviations exist due to political reasons. For
example, when in election, water tariffs are reduced around 3%, on average, for Austrian local government water services ​[10]​. ​Moreover, even
if water rates increase, they are not announced to avoid political costs.

However, political decision-making must include local governments' ​fi​nancial considerations. According to Bel et al. ​[42] ​and Carpentier et
al. ​[43]​; local governments must pay attention to the tax revenues when performing their services. Therefore, local government leaders also need
to knowledgeable about expenses for providing public services such as water, such as civil servant salaries. Thus, local government ​fi​nancial debt
and cost structure must be balanced with taxes and service tariffs. If earnings from these two sources fall short of expenditures, water prices will
be pressured upward ​[41]​.

Having described the institutional, governance and contextual factors identi​fi​ed in the literature regarding water prices, we now describe the
method adopted in this study.

4. Research design and method

4.1. Data and variables

This study is based on a cross-sectional research design ​[13]​, which means that the data collection was carried out in a single period (2013)
for several units of analysis. This year was chosen for several reasons. The ​fi​rst is due to the available data both for water utilities

– ​our dependent variable and institutional and governance variables – ​and contextual factors. For internal factors, data were obtained ​from the
Portuguese Water Sector Regulator's Annual Report ​[44]​ ​for 2014. This report contains information on 284 operators from the
mainland (Azores and Madeira were excluded). Like Bel et al. ​[42]​; our empirical analysis uses different dependent variables for 2013: user
3 3
prices for a 5 m​ consumption level (codi​fi​ed as CP5m3); user prices for a 10 m​ consumption level (codi​fi​ed as CP10m3); and user prices for a
15 m​ consumption level (codi​fi​ed as CP15m3). The three dependent variables measure the total amount to be paid for monthly consumption
including ​fi​xed charges. As discussed earlier, choosing customer prices as dependent variable is due to the ​fi​rms' strategy that aims to lower their
operational costs, allowing more aggressive pricing schemes ​[5]​.

Applying the resource-based view to identify and measure internal factors, we ​fi​nd that key variables are the ​“​a ​fi​rm's management skills, its
organizational processes and routines, and the information and knowledge it controls​” ​[1]​; p. 625). For that reason, and as a best proxy,
organizational costs aim to measure ​“​organizational processes and routines​” (codi​fi​ed as OrganiCosts). Due to the lack of other operational and
fi​nancial numbers, data on organizational costs (measured in mean ​€​/month) appear to be the best proxy since real unit cost of production must
be considered. It is expected that ​fi​rms in possession of superior resources will be able to decrease their production costs ​[3]​. Secondarily,
organizational costs become a dependent variable, which can be justi​fi​ed due to the resource-based view in supporting the idea that superior
resources will lead to competitive advantage through reduction of operational costs. With lower costs, customer prices will also be lower ​[19]​.
However, we lack the knowledge about which institutional, governance and contextual factors explain them, so we perform a second regression
Other identi​fi​ed internal factors are ownership, management model and economic heterogeneity. For Carvalho and Marques ​[11]​; economic
heterogeneity (EcoHeteroge) and ownership (Ownership) must be considered as external factors since the political decision is taken for granted.
However, when politicians decide to change ownership, management model or even economic heterogeneity, they are de​fi​ning the ​fi​rms'
structures and designs. In such a way, we are dealing with coercive isomorphism ​[52]​. Those ​fi​rms face particular environments and must be
managed under those circumstances. For those reasons, we considered these variables as institutional and governance factors. Ownership, for
example, of each operator is considered as an institutional and governance variable. For meas-urement, a binary variable was applied, where 1
indicates public ownership and 2 indicates private ownership.

Regarding the management models, dummy variables were applied for bureaucracy alone (coded as Bureau, where local public structures are
responsible for running the activity), which equals 1, and 0 otherwise; for municipalized services (coded as MunServi, they are administratively
and ​fi​nancially autonomous from local governments, but they lack legal autonomy, so they are controlled by the local bureaucracy), which equals
1, and 0 otherwise; for public agencies (coded as PubliAgenc, they are administratively, ​fi​nancially and asset independent and are managed under
commercial law), which equals 1, and 0 otherwise; for a municipal concession contract to a private ​fi​rm (coded as Private), which equals 1, or 0
otherwise; and for multi-concession model (coded as Multi-con, where the central government instead of local government establishes a
concession contract with a public enterprise), which equals 1, or 0 otherwise.
The last identi​fi​ed internal factor is economic heterogeneity, where ​fi​rms performing in the water supply and sewage collection and treatment
might bene​fi​t from economies of scope (see Ref. ​[35]​. In this case, a binary variable was applied, where 1 indicates an organization solely
supplying water and 2 indicates an organization also providing sewage collection and treatment services.
On the side of contextual factors, data were collected from the Portuguese Data site (​http://www.pordata.pt/​, last consulted on 3 February
2016). This online database contains data from several sources, such as the National Statistics Institute, the budget division of the Ministry of
Economics and Finance and the Ministry of Home Affairs, among others. Data collection followed the dimensions and indicators for each
municipality in 2013, based on the discussion presented at section ​3​. Social-demographic indicators included: total

H.C. Silvestre, R.C. Gomes Water Resources and Economics xxx (2017) 1​–​10

2 2
number of inhabitants (numinhabi), the mean number of inhabitants/km​ (numinha/km2); number of cities (numbercities); total area in km​
(TotalAreakm2); inhabitants' mean salary per month in euros (InhabiMSala); and total unemployed (TUnem) (see Refs. ​[9,10,33,35,42]​; ​among
others). For the political and local government ​fi​nancial sustainability we added: total number of registered ​voters for 2013 local government
elections (RegisVoLGEle); total number of those who actually voted in the 2013 local government election (VotersLGEle); local government
total expenses in thousand ​€ (LGTExp); local government ​fi​nancial liability in thousand ​€ (LGFinanLiabi); local government payment to local
employees per capita in ​€ (LGEmploysala); total local government revenues in thousand ​€ (LGTotalReven); total local government taxes
collected in thousand ​€ (LGTCOleTaxes); local government ​fi​nancial balance in thousand ​€ (LGFinanBalanc) (see Refs. ​[12,41,42]​; among
others). All variables are detailed in ​Table 1​ ​and their descriptive statistics are presented in ​Table 2​.

4.2. Analytical tools

For data analysis, we applied hierarchical multiple linear regression, which is useful to estimate the coef​fi​cients of the linear equation that
best predict the value of the dependent variable. This technique includes several independent variables, of which we included the institutional
and governance variables in a ​fi​rst stage, followed by the contextual factors. This option is justi​fi​able, and in line with the resource-based view's
assumptions, we tested the effects of the ​fi​rst set of predictors independently of the in​fl​uence of contextual factors. Additionally, we applied the
t-test to indicate if the regression coef​fi​cients were statistically signi​fi​cant (at a level of p ​< 0.05). By applying analysis of variance (ANOVA -
F-test), which indicates the signi​fi​cance level of the regression (p ​< 0.05), we were able to ascertain which independent variables ​fi​t the
regression model. Also, we calculated R-squared (R​ ) scores to understand the explanatory power of the model, where the closer to 1 the score
is, the stronger explanatory power is.

Additionally, we applied the stepwise method. On its own, this method searches for independent variables that will (not) be part of the
equation. The independent variables that present largest probability of F are removed from the equation while the others are not. Put simply, it
tries to ​fi​nd the best matching variables for the given dependent variables. Besides this, and in order to distinguish the null hypothesis for the
independent variables and customer prices, we applied the Durbin-Watson test. This statistical test's main concern is

Table 1

Variables: description and sources.

Variable code Description Source

CP5m​3 Dependent variable: Monthly water consumption in ​€​/month. This price Portuguese Water Sector Regulator's Annual Report (ERSAR)
includes both the ​fi​xed quota and the price for the water consumed.
CP10m​3 Dependent variable: Monthly water consumption in ​€​/month. This price
includes both the ​fi​xed quota and the price for the water consumed.
CP15m​3 Dependent variable: Monthly water consumption in ​€​/month. This price
includes both the ​fi​xed quota and the price for the water consumed.
OrganiCosts (In)dependent organizational costs (measured in mean ​€​/month) appears
to be the best proxy for real unit cost of production.

Institutional and governance factors
EcoHeteroge Dummy variable that takes a value of 1 for organizations providing
Ownership Binary variable that takes a value of 1 for public organizations, and 2
for and sewer services, and 0 otherwise.

private ​fi​rms. Political and local government

MuniServi Dummy variable that takes a value of 1 for municipalized services, LGTExp Local government total expenses in ​€/​ thousand for 2013.
and 0
LGFinanLiabi Local government Financial liability in ​€​/thousand for 2013.
LGEmploysala Local government payment to local employees per capita in ​€​ for
Bureau Dummy variable that takes a value of 1 for bureaucracies, and 0 2013.
LGTotalReven Local government total revenues in ​€​/thousand for 2013.
PubliAgenc Dummy variable that takes a value of 1 for public agencies, and 0
LGTCOleTaxes Local government total collected taxes in ​€/​ thousand for 2013.

LGFinanBalanc Local government ​fi​nancial balance in ​€​/thousand for 2013.

Private Dummy variable that takes a value of 1 for private ​fi​rms, and 0 RegisVoLGEle Total number of registered voters in 2013 for local government
otherwise. elections.

Multi-con Dummy variable that takes a value of 1 for multi-concession, and 0 VotersLGEle Total number of those who actually voted in 2013 in local
elections. Portuguese Water Sector Regulator's Annual Report (ERSAR)


InhabiMSalar Inhabitants' mean salary per month in ​€​ for 2013.

TUnemBeneSB Total unemployed in 2011.

TotalArea km2 Km2 in 2013.

Numbercities Number of cities.

Numinhabi Total number of inhabitants in 2013. National Statistics Institute, the budget division from the Ministry of Economics
and Finance, the Ministry of Home Affairs, among others.

Numinhabi/ The mean of the number of inhabitants/km​2​ in 2013.


National Statistics Institute, the budget division from the Ministry of Economics
and Finance, the Ministry of Home Affairs, among others.


H.C. Silvestre, R.C. Gomes Water Resources and Economics xxx (2017) 1​–​10

Table 2
Summary of descriptive statistics.
Variable Obs. Mean SD Minimum Maximum

CP5m3 281 5.2349 2.28331 0.75 16.08
CP10m3 281 9.2395 3.24630 1.50 20.38
CP15m3 281 14.1066 4.68098 2.80 27.28
Institutional and governance
OrganiCosts (in) 281 2.7236 1.83444 0.00 14.18
Ownership (binary) 284 1.11 – – –
Dummy for MuniServi 284 0.08 – – –
Dummy for Bureau 284 0.65 – – –
Dummy for PubliAgenc 284 0.11 – – –
Dummy for Private 284 0.12 – – –
Dummy for Multi-con 284 0.04 – – –
EcoHeteroge (dummy) 283 1.05 – – –
Political and Local Government
LGTExp 284 27113.8627 40611.79665 3553.61 544333.26
LGFinanLiabi 284 2234.4487 4924.36679 0.00 69840.33
LGEmploysala 284 323.9591 177.36898 95.76 1084.23
LGTotalReven 284 27465.6312 41355.95627 3735.48 546365.85
LGTCOleTaxes 284 8406.9808 19662.67774 156.77 242297.44
LGFinanBalanc 284 2800.5158 6462.08449 12543.85 64085.77
RegisVoLGEle 284 32507.8662 50451.85851 1437.00 507495.00
VotersLGEle 284 17061.4014 23988.40505 1123.00 228682.00
InhabiMSalar 284 882.2479 165.24546 0.00 1788.00
TUnemBeneSB 284 2284.7148 3990.67835 83.00 30839.00
TotalArea km2 284 328.7958 289.82866 8.00 1721.00
Numbercities 284 0.5423 0.70958 0.00 4.00
Numinhabi 284 35936.5775 56886.49059 1775.00 520549.00
Numinhabikm2 284 300.4285 825.98247 4.70 7378.40

to detect autocorrelation of the residuals. If the values stand between 1.5 ​< p ​< 2.5, the null hypothesis can be rejected and the existence of
differences accepted. That was the case for the regression analysis, validating the models. The data were analyzed by the Statistical Package for
the Social Sciences (SPSS).

5. Results and discussion

The ​fi​rst test, using stepwise hierarchical regression analysis, showed that the institutional and governance variables were stat-istically
2 3​ 2 3 2
signi​fi​cant and best ​fi​t the models (p ​< 0.05) ​– see ​Table 3​. The high explanatory power (R​ ​¼ 0.887 for CP5m​ , R​ ​¼ 0.757 CP10 m​ and R​ ​¼
0.566 for CP15 m​ ) indicates the strength of the relationship between ​‘​internal factors​’ and ​‘​customer prices​’​. Organizational costs, the variable
that measures ​‘​organizational processes and routines​’,​ among the considered factors was the one that presented the highest statistical scores for
all customer prices levels. For that reason, Peteraf and Barney's ​[3] ​fi​ndings can be applied to the Portuguese water supply industry. In fact,
superior resources were the key determinants for lower operational costs, thus lowering user prices.

Ownership was the second variable present in all models. Mean customer prices under public ownership were lower (M​¼€​4.87 for CP5m​ ;
3​ 3​
M​¼€​8.68 for CP10 m​ ; and M​¼€​13.25 for CP15 m​ ) in comparison with the mean customer prices under private ownership (M​¼€​8.10 for
3​ 3​ 3​
CP5m​ ; M​¼€​13.62 for CP10 m​ ; and M​¼€​20.76 for CP15 m​ ). These ​fi​ndings contradict the belief that private ​fi​rms outperform public-sector
organizations ​[27]​. According to Bel et al. ​[42]​; one of the main arguments of those who are against private-sector participation in water services
is higher rates. For example, Chong et al. ​[45] ​found that under private sector participation, water prices are higher by 25​€ in comparison with
public organizations. Some caution is needed in this analysis: since private operators signed concession contracts, where initial investments were
demanded, so their operational costs include such items ​[20]​. Besides this, and according to Carpentier et al. ​[43]​; private ​fi​rms set higher prices
because local governments tend to privatize the service in more complex scenarios where service costs are inevitably higher.

Nevertheless, organizational costs under public ownership were much lower (M​¼€​2.46) than private ​fi​rms' organizational costs (M​¼€​4.76).
In comparison with public-sector organizations, private ​fi​rms would be expected to have lower operating costs, since those initial investments
would allow improving their operational performance while lowering customer prices. However, Pinto et al. ​[6] found that Portuguese
public-sector organizations might not include the real cost of the service in customer prices. Besides this, they showed a positive in​fl​uence on
operational ef​fi​ciency when private ​fi​rms are represented. According to Silvestre ​[20]​; there is another issue that needs to be discussed: sharing
costs between public-sector organizational departments, which leads to lower operational costs, thus allowing lower client prices.
Due to these reasons, there is an ongoing debate regarding the (de)merits of public-sector enterprises, where it might be preferable to keep
management of utilities under direct government control (see Ref. ​[30]​. Given the unexpected social and organizational con-sequences arising
with private sector participation in water services, it becomes less controversial to rely on public-sector organizations

H.C. Silvestre, R.C. Gomes Water Resources and Economics xxx (2017) 1​–​10

Table 3

Multiple regression through the stepwise method, with internal and contextual factors as predictors of water prices in Portugal.

Variable Model 1** Model 2*** Model 3****

Coeff. t-stat Coeff. t-stat Coeff. t-stat

(constant) 0.000* 8.655 0.000* 9.182 0.000* 7.493

Institutional and governance
OrganiCosts (in) 0.000* 40.936 0.000* 23.040 0.000* 12.891
Ownership (binary) 0.000* 4.114 0.000* 5.964 0.000* 6.785
Dummy for MuniServi 0.087 1.720 0.000* 3.684 0.004* 2.937
Dummy for Bureau 0.647 0.458 0.826 0.220 0.127 1.532
Dummy for PubliAgenc 0.485 0.700 0.839 0.204 0.319 0.998
Dummy for Private 0.256 1.137 0.341 0.954 0.988 0.015
Dummy for Multi-con 0.909 0.115 0.702 0.383 0.321 0.993
EcoHeteroge (dummy) 0.095 1.676 0.291 1.058 0.081 1.749
Political and Local Government
LGTExp 0.108 1.612 0.225 1.216 0.336 0.963
LGFinanLiabi 0.126 1.533 0.279 1.085 0.543 0.608
LGEmploysala 0.109 1.609 0.079 1.764 0.306 1.025
LGTotalReven 0.120 1.560 0.243 1.171 0.327 0.982
LGTCOleTaxes 0.109 1.607 0.215 1.244 0.167 1.386
LGFinanBalanc 0.423 0.802 0.305 1.027 0.317 1.003
RegisVoLGEle 0.203 1.275 0.465 0.731 0.634 0.476
VotersLGEle 0.375 0.889 0.723 0.354 0.929 0.089
InhabiMSalar 0.192 1.308 0.835 0.209 0.315 1.006
TUnem 0.296 1.047 0.651 0.453 0.953 0.060
TotalArea km2 0.214 1.244 0.807 0.245 0.729 0.346
Numbercities 0.916 0.106 0.627 0.487 0.940 0.075
Numinhabi 0.245 1.165 0.518 0.648 0.678 0.415
Numinhabikm2 0.586 0.545 0.665 0.433 0.786 0.272
R 0.942 0.870 0.753
R​2 0.887 0.757 0.566
Adjusted R​2 0.886 0.755 0.562
F change 1086.177 288.407 120.614
Df 280 280 280
Durbin-Watson 2.142 2.090 2.075

* sig ​<​.05.

** dependent variable: Monthly water consumption 5 m​3​.

*** dependent variable: Monthly water consumption 10 m​3​.

**** dependent variable: Monthly water consumption 15 m​3​.

for supply ​[9]​. This study proves that ownership is an important variable when dealing with customer price determinants. It should be
remembered that ownership rights are associated with speci​fi​c goals. From the public side, equity in the access and consumption of water is a
primary public concern, the reason why customer prices are typically lower under public-sector organizations ​[8]​. After all, public sector reform
introduced regulatory agencies as a mechanism to accompany privatization of water utilities.

In that respect, and when considering the management arrangement, our third model (CP15 m​ ) was statistically signi​fi​cant. The only
statistically valid variable regarding the existence of a management model was due to municipalized services, with scores M​¼€​16.59 for CP15
3 3
m​ when existing, and scores M​¼€​13.89 for CP15 m​ when not existing. Additionally, considering organizational costs, when municipalized
services exist, the score was M​¼€​3.50 and when not existing the score was M​¼€​2.65. Without applying for the privatization or private sector
participation in water services, 23 local Portuguese governments (around 8% of the sample) decided to keep an arrangement administratively
autonomous from local government. In theory, this is an alternative to the traditional bureaucracy, aimed to improve operational ef​fi​ciency while
avoiding political interference in performing daily organizational tasks. Regarding the above mentioned discussion, around the (de)merits of
public-sector organizations (see Ref. ​[30]​, this study clearly shows that municipalized services are not an alternative to lower client prices while
improving organizational costs. Without the possibility to share operational costs with other public departments and lacking political autonomy,
they are still controlled by the local bureaucracy. According to ​[9]​; to avoid political interference and when private sector participation is
impossible, corporatized public organizations are a possible alternative.

According to Bithas ​[23]​; ​“​in the majority of cases, the operational costs of water companies form the ​‘​indicator​’ for de​fi​ning prices,​”
something that we also found. But in order to understand which institutional, governance and contextual factors are statistically related with
organizational costs, a hierarchical regression analysis was performed. The ​fi​ndings are presented in ​Table 4​, from where we conclude that
municipalized services (p​¼​0.000), bureaucracy (p​¼​0.000) and public agencies (p​¼​0.028) presented a statistical rela-tionship with
organizational costs, but with low explanatory power than the previous, with p​¼ 39%. Overall, organizational costs are lower under
bureaucracies (with M​¼€​1.92 when present and M​¼€​4.22 when not present), followed by municipalized services and public agencies (with
M​¼€​4.01 when present and M​¼€​2.56 when not present). Regarding private-sector organizations, although not statistically signi​fi​cant, the score
was M​¼€​4.66 when present and M​¼€​2.47 when not present.
Several implications arise from these ​fi​ndings. First of all, and once again, we did not ​fi​nd support for the theory that private

H.C. Silvestre, R.C. Gomes Water Resources and Economics xxx (2017) 1​–​10

Table 4

Multiple regression through the stepwise method, with internal and contextual factors as predictors of organizational costs in Portuguese water

Variable Model 1*
Coeff. t-stat

(constant) 0.000* 21.821

Institutional and governance
Ownership (binary) 0.956 0.055
Dummy for MuniServi 0.000* 3.373
Dummy for Bureau 0.000* 11.674
Dummy for PubliAgenc 0.028* 2.203
Dummy for Private 0.465 0.731
Dummy for Multi-con 0.465 0.731
EcoHeteroge (dummy) 0.329 0.978
Political and Local Government
LGTExp 0.697 0.390
LGFinanLiabi 0.975 0.032
LGEmploysala 0.106 1.620
LGTotalReven 0.717 0.363
LGTCOleTaxes 0.925 0.094
LGFinanBalanc 0.697 0.390
RegisVoLGEle 0.856 0.181
VotersLGEle 0.553 0.593
InhabiMSalar 0.148 1.449
TUnem 0.866 0.169
TotalArea km2 0.172 1.371
Numbercities 0.878 0.154
Numinhabi 0.873 0.160
Numinhabikm2 0.780 0.280
R 0.622
R​2 0.386
Adjusted R​2 0.380
F change 58.120
Df 280
Durbin-Watson 2.269

* sig ​<​.05.

** dependent variable: organizational costs.

provision of public services is more ef​fi​cient. ​“​Discipline and parsimony in the resource use​” ​[24]​; pp. 4​–​5) has not been accomplished under
private management in Portugal. The same is valid for Spanish municipalities, where Suarez-Varela et al. ​[36] ​found that public sector
organizations are less ef​fi​cient in managing human resources in comparison with private organizations. However, our ​fi​ndings also show that
public sector organizations present higher ef​fi​ciency levels at the operational costs management in a comparison with private organizations.
Besides, and for the Portuguese case, if initial investments are argued to be the reason for higher customer prices, the better performance of
multi-concession contracts contradicts this argument (they presented higher organizational costs (M​¼€​5.03) in comparison with private ​fi​rms
(M​¼€​4.66), while customer prices were lower than the latter, especially at CP15 m​ , where the ​fi​rst scored M​¼€​15.92 and the latter
M​¼€​20.48). A second implication is due to public agency scores: their customer prices and organ-izational costs were higher in comparison
with the traditional bureaucracy. If public agencies aim to lower organizational costs, this has not been ful​fi​lled. Once again, a possible
explanation is due to the sharing of costs between departments under the bureaucracy, something that is not possible when public agencies are
managed independently from the local government structure.

The only certainty we have is that under public guidance, commercial exploitation will be avoided ​[30]​. However, comparing bureaucracy
and public agency results, the differences between operational cost and customer prices are not proportionally equal ​– public agencies are twice
as expensive as direct government control by the bureaucracy, but customer prices are not commensurately lower.

6. Conclusions and future research

This study started with a theoretical discussion regarding which factors are important for ​fi​rms' performance based on a resource-based view.
As far as we know, this theoretical discussion has not been applied previously to the water industry. Besides this, few studies have aimed to
understand the factors that in​fl​uence customer prices. Applying the resource-based view to the Portuguese water supply industry, we can
conclude that institutional and governance variables are major drivers for customer prices, namely organizational costs and ownership.
Organizational costs are the crucial factor, something that is assumed by the resource-based view: superior resources reduce operational costs,
making it possible to lower customer prices ​[19]​. External/contextual factors, such as socio-demographic, political and local government
economic and ​fi​nancial results, were not statistically related with customer prices.

When considering the determinants of organizational costs, bureaucracy, municipalized services and public agencies were key

H.C. Silvestre, R.C. Gomes Water Resources and Economics xxx (2017) 1​–​10

statistical determinants, where the ​fi​rst presented the lowest organizational costs. Despite the statistical evidence, it is important to critically
analyze all the identi​fi​ed organizational arrangements. For bureaucracies, cost sharing with other organizational departments might in​fl​uence
organizational costs, thus allowing reduction of customer prices. For municipalized services, the results, in comparison with the others, are
negative. That can be explained by the lack of political split from local government while presenting operational inef​fi​ciencies. For public
agencies, they presented lower prices and organizational costs compared with public-sector organizations, but higher prices and organizational
costs in comparison with other public-sector organizations (except with multi-concession contracts). For private sector organizations, it has been
shown that ef​fi​ciency gains and cost reductions seem to be mixed ​[18] ​or scarce ​[46]​. For these reasons, private sector involvement in provision
of public services has not accomplished the political and theoretical expectations for its use ​[47,48]​. However, when private-sector organizations
sign a a contract with a local government, they usually pay considerable upfront rents, especially when European Union funds and not available
for such cases. Besides this, the accounting rules are not the same. For example, the depreciation periods for private ​fi​rms are shorter (usually
between 25 and 30 years) and therefore highlight higher costs. Last but no less important, multi-concession arrangements are a possible solution
for this industry, as argued by Carvalho et al. ​[49]​; and Kurki et al. ​[50]​.

There are two implications of the current ​fi​ndings. The ​fi​rst is political in nature and involves the discussion of the best way to improve the
provision of public services by utilities. Clearly, from an economic, managerial and social approach, all the arrangements, whether public or
private, should be rethought. The second implication is theoretical. The resource-based view indicates the existence of superior resources, the
key variable to explain differences in organizational performance. The academic community might be interested in knowing which superior
resources are means and what they are based on. Due to these questions, which still lack answers, further research is needed. We believe this
article is a starting point for future developments to better understand this crucial social phenomenon.


[1] J.J. Barney, M. Wright, D.J. Ketchen Jr., The resource-based view of the ​fi​rm: ten years after 1991, J. Manag. 27 (2001) 625​–​641, ​http://dx.doi.org/10.1177/
014920630102700601​.​ ​Retrieved from:.
[2] J.B. Barney, Resource-based theories of competitive advantage: a ten-year retrospective on the resource-based view, J. Manag. 27 (2001) 643​–​650, ​http://
dx.doi.org/10.1177/014920630102700602​.​ ​Retrieved from:.
[3] M.A. Peteraf, J.B. Barney, Unraveling the resource-based tangle, Manag. Decis. Econ. 24 (2003) 309​–​323, ​http://dx.doi.org/10.1002/mde.1126​.

[4] R.Q. Grafton, et al., Responding to global challenges in food, energy, environment and water: risks and options assessment for decision-making, Asia Pac. Policy Stud. 3 (2)
(2016) 275​–​299, ​http://dx.doi.org/10.1002/app5.128​.
[5] S.L. Hart, A natural-resource-based view of the ​fi​rm, Acad. Manag. Rev. 20 (1995) 986​–​1014, ​http://dx.doi.org/10.5465/AMR.1995.9512280033​.

[6] F.S. Pinto, P. Simoes,​~​ R.C. Marques, R. C, Water services performance: do operational environment and quality factors count? Urban Water J. 1​–​9 (2016) ​http://
[7] H.C. Silvestre, R.C. Gomes, Regulatory politics in user prices: evidence for the Portuguese water industry, Water Policy 18 (2016) 1194​–​1208, ​http://dx.doi.org/
[8] J.A. Beecher, J.A. Kalmbach, Structure, regulation, and pricing of water in the United States: a study of the Great Lakes region, Util. Policy 24 (2013) 32​–​47,
[9] M. Klien, Corporatization and the behavior of public ​fi​rms: how shifting control rights affects political interference in water prices, Rev. Ind. Organ. 44 (4) (2014b)
393​–​422, ​http://dx.doi.org/10.1007/s11151-013-9401-0​.
[10] M. Klien, Tariff increases over the electoral cycle: a question of size and salience, Eur. J. Polit. Econ. 36 (2014a) 228​–​242​.

[11] P. Carvalho, R.C. Marques, The in​fl​uence of the operational environment on the ef​fi​ciency of water utilities, J. Environ. Manag. 92 (2011) 2698​–​2707, ​http://
[12] R.C. Marques, S. Berg, S. Yane, Nonparametric benchmarking of Japanese water utilities: institutional and environmental factors affecting ef​fi​ciency, J. Water Resour.
Plan. Manag. 140 (5) (2014) 562​–​571, ​http://dx.doi.org/10.1061/(ASCE)WR.1943-5452.0000366​.
[13] N. Blaikie, Designing Social Research, Polity, New York, 2009​.

[14] E.T. Penrose, The Theory of the Growth of the Firm, John Wiley, New York, 1959​.

[15] N.J. Foss, P.G. Klein, Y.Y. Kor, J.T. Mahoney, Entrepreneurship, subjectivism, and the resource-based view: toward a new synthesis, Strat. Entrepreneursh. J. 2 (1) (2008)
73​–​94, ​http://dx.doi.org/10.1002/sej.41​.
[16] M.H. Kunc, J.D. Morecroft, Managerial decision making and ​fi​rm performance under a resource-based paradigm, Strat. Manag. J. 31 (11) (2010) 1164​–​1182,
[17] K.M. Eisenhardt, J.A. Martin, Dynamic capabilities: what are they? Strat. Manag. J. 21 (10​–​11) (2000) 1105​–​1121, ​http://dx.doi.org/10.1002/1097-0266(200010/11)21​,
[18] G. Bel, X. Fageda, M.E. Warner, Is private production of public services cheaper than public production? A meta-regression analysis of solid waste and water services, J.
Policy Anal. Manag. 29 (3) (2010) 553​–​577, ​http://dx.doi.org/10.1002/pam.20509​.
[19] M.A. Peteraf, The cornerstones of competitive advantage: a resource-based view, Strat. Manag. J. 14 (1993) 179​–​191, ​http://dx.doi.org/10.1002/​ ​smj.4250140303​.

[20] H.C. Silvestre, Public​–​private partnership and corporate public sector organizations: alternative ways to increase social performance in the Portuguese water sector? Util.
Policy 22 (2012) 41​–​49, ​http://dx.doi.org/10.1016/j.jup.2012.01.002​.
[21] J.M. Mol, N.M. Wijnberg, From resources to value and back: competition between and within organizations, Br. J. Manag. 22 (1) (2011) 77​–​95, ​http://dx.doi.org/
[22] C.E. Helfat, M.A. Peteraf, The dynamic resource-based view: capability lifecycles, Strat. Manag. J. 24 (2003) 997​–​1010, ​http://dx.doi.org/10.1002/smj.332​.

[23] K. Bithas, The sustainable residential water use: sustainability, ef​fi​ciency and social equity. The European experience, Ecol. Econ. 68 (2008) 221​–​229, ​http://
[24] C. Hood, A public management for all seasons? Public Adm. 69 (1) (1991) 3​–​19, ​http://dx.doi.org/10.1111/j.1467-9299.1991.tb00779.x​.

[25] C. Pollitt, G. Bouckaert, Public Management Reform: a Comparative Analysis-new Public Management, Governance, and the Neo-weberian State, Oxford Uni-versity
Press, Oxford, 2011​.
[26] N. Carter, Learning to measure performance: the use of indicators in organizations, Public Adm. 69 (1991) 85​–​101, ​http://dx.doi.org/10.1111/j.1467-9299.1991.tb00783.x​.
[27] K. Bakker, Paying for water: water pricing and equity in England and Wales, Trans. Inst. Br. Geogr. 26 (2) (2001) 143​–​164, ​http://dx.doi.org/10.1111/1475-5661.00012​.

[28] G. Romano, N. Salvati, A. Guerrini, Estimating the determinants of residential water demand in Italy, Water 6 (2014) 2929​–​2945, ​http://dx.doi.org/10.3390/​ ​w6102929​.

H.C. Silvestre, R.C. Gomes Water Resources and Economics xxx (2017) 1​–​10

[29] M. García-Valinas,​~​ F. Gonzalez-Gomez, A.J. Picazo-Tadeo, Is the price of water for residential use related to provider ownership? Empirical evidence from Spain, Util.
Policy 24 (2013) 59​–​69, ​http://dx.doi.org/10.1016/j.jup.2012.07.009​.
[30] D. Hall, Public Ownership of the UK Energy System ​–​ Bene​fi​ts, Costs and Processes, Public Services International Research Unit, 2016. Retrieved from, ​http://
[31] R.Q. Grafton, L. Chu, T. Kompas, Optimal water tariffs and supply augmentation for cost-of-service regulated water utilities, Util. Policy 34 (2015) 54​–​62, ​http://
[32] P. Carvalho, R.C. Marques, Computing economies of scope using robust partial frontier nonparametric methods, Water 8 (82) (2016) 1​–​23, ​http://dx.doi.org/
[33] J. Raadschelders, Fresh Water Management as Collective Action in Time: a multi-dimensional and multi-level challenge of governance in an inter-disciplinary

th​ th​
context, in: J. Raadschelders (Ed.), The Institutional Arrangements for Water Management in the 19​ ​and 20​ ​Centuries, IOS Press, Amsterdam, 2005, pp. 1​–​22​.
[34] B. Cooper, L. Crase, N. Pawsey, Best practice pricing principles and the politics of water pricing, Agric. Water Manag. 145 (2014) 92​–​97, ​http://dx.doi.org/

[35] R.C. Marques, K. De Witte, Is big better? On scale and scope economies in the Portuguese water sector, Econ. Model. 28 (2011) 1009​–​1016, ​http://dx.doi.org/
[36] M. Suarez-Varela, M. García-Valinas,​~​ F. Gonzalez-Gomez, A. Picazo-Tadeo, Ownership and performance in water services revisited: does private management really
outperform public? Water Resour. Manag. (2016) 1​–​19, ​http://dx.doi.org/10.1007/s11269-016-1495-3​. September.
[37] R. Chettri, R. Venkatesan, Water resource pricing Policy options, Resour. Policy (1983) 288​–​295, ​http://dx.doi.org/10.1016/0301-4207(83) 90005​–​3​. December.

[38] L. Lynn, C. Heinrich, C. Hill, Studying governance and public management: challenges and prospects, J. Public Adm. Res. Theory 10 (2000) 233​–2​ 61, ​http://
[39] A. Reynaud, Assessing the impact of full cost recovery of water services on European households, Water Resour. Econ. 14 (2016) 65​–​78, ​http://dx.doi.org/
[40] W. Niskanen, Bureaucracy and Representative Government, Aldine-Atherton, Chicago, 1976​.

[41] M. Klien, The political side of public utilities: how opportunistic behaviour and yardstick competition shape water prices in Austria, Pap. Reg. Sci. 94 (4) (2015) 869​–​890,
[42] G. Bel, F. Gonzalez-Gomez, A.J. Picazo-Tadeo, Does market concentration affect prices in the urban water industry? Environ. Plan. C Gov. Policy 33 (6) (2015)
1546​–​1565, ​http://dx.doi.org/10.1177/0263774X15614144​.
[43] A. Carpentier, C. Nauges, A. Reynaud, A. Thomas, Effets de la delegation sur le prix de l​’​eau potable en France: une analyse ​a ​partir de la litterature sur les effets de
traitement, Econ. Prevision 174 (3) (2006) 1​–​20​.
[44] The Water and Waste Services Regulation Authority ERSAR, Annual Report on Water and Waste Services in Portugal e Executive Summary, The Audit Department of
Exchequer, Lisbon, 2014. Retrieved from: ​http://www.ersar.pt/pt/publicacoes/relatorio-anual-do-setor​.
[45] E. Chong, F. Huet, S. Saussier, Auctions, ex post competition and prices: the ef​fi​ciency of public-private partnerships, Ann. Public Coop. Econ. 77 (4) (2006)​ ​521​–​554​.

[46] K.Z. Hossain, S.A. Ahmed, Non-conventional public-private partnerships for water supply to urban slums, Urban Water J. 12 (7) (2015) 570​–​580, ​http://
[47] D. Hall, E. Lobina, R.D.L. Motte, Public resistance to privatisation in water and energy, Dev. Pract. 15 (3​–​4) (2005) 286​–​301, ​http://dx.doi.org/10.1080/
[48] M.E. Warner, Competition or cooperation in urban service delivery? Ann. Public Coop. Econ. 82 (4) (2011) 421​–​435, ​http://dx.doi.org/10.1111/j.1467-8292.2011.00450.x​.

[49] P. Carvalho, I. Pedro, R.C. Marques, The most ef​fi​cient clusters of Brazilian water companies, Water Policy 17 (5) (2015) 902​–​917, ​http://dx.doi.org/10.2166/
[50] V. Kurki, P. Pietil​€​a, T. Katko, Assessing regional cooperation in water services: Finnish lessons compared with international ​fi​ndings, Public Works Manag. Policy 21 (4)
(2016) 368​–​389, ​http://dx.doi.org/10.1177/1087724X16629962​.
[51] C. Talbot, Theories of Performance ​–​ ​Organizational and Service Improvement in the Public Domain, Oxford University Press, Oxford, 2010​.

[52] P. DiMaggio, W.W. Powell, The iron cage revisited: collective rationality and institutional isomorphism in organizational ​fi​elds, Am. Sociol. Rev. 48 (2) (1983)​ ​147​–​160​.

[53] P.H. Gleick, J. Christian-Smith, H. Cooley, Water-use ef​fi​ciency and productivity: rethinking the basin approach, Water Int. 36 (7) (2011) 784​–​798, ​http://