Refurbishing the Ricardo Effect

The Ricardo Effect has been a specter in the Austrian theory of the business cycle
for decades. Despite its most notable proponent F.A Hayek writing several essays and
chapters on its importance, it has fallen into relative obscurity. The reason for this is
twofold. On the one hand orthodox economists are simply not aware of it, or reject it on
empirical grounds. On the other, Austrians have misinterpreted the nuances of the
Ricardo Effect and have quietly acknowledged its ambiguity, continuing to support their
business cycle theory without it. In fact, Mises had no use for it in his explanation of the
boom and bust cycle in his economic treatise, Human Action. It is only mentioned once
and is a non-cyclical version of the one found in Hayek's work. It's possible that the
majority of the fault lies on Hayek's shoulders. He never came to a definitive conclusion
on the topic, taking only a piecemeal approach in its defense and clarification, and
eventually focused his efforts on political philosophy instead. Thus his opponents have
forgotten his work on the subject, and his allies have largely ignored it. This essay hopes
to revitalize this integral link that can unify several different economic effects in a single
framework, so that a coherent story can be told in terms of business fluctuations. The
purpose of our inquiry is not to restate an entire theory of this instability, but merely to
analyze one aspect of it and its importance to what has been called the 'turning point' of
the trade cycle.

Business Cycle Canon

A proper description of the Ricardo Effect is necessary. Throughout Hayek's work

he subtly changed parts of the concept to address specific objections made by
competing economists, although these changes were mostly just in terminology. From
Prices and Production to The Pure Theory of Capital, Hayek emphasized workers' wages
in relation to prices, yet often used aggregate concepts like real wages which somewhat
contradicted his methodological viewpoint on heterogenous capital and individual prices.
This most likely made his argument more muddled and allowed for misinterpretations.
However in his lesser cited 1942 article in Individualism and Economic Order, he
provides a useful working definition:

The proposition here described as the Ricardo effect asserts that a general change in wages
relative to the prices of the products will alter the relative profitability of different industries or methods of
production which employ labor and capital ("indirect labor") in different proportions.
 While a narrow understanding of the Ricardo effect is desired ultimately, this
broad explanation suits our exploration of its differing versions. In its most popular form it
states that during an economic boom stimulated by artificial credit expansion, incomes
from wages and entrepreneurial profits enable spending on consumer goods, which
raises demand and thus prices. This causes wages to lose purchasing power, thus
lowering real wages. When businesses see that real wages are falling, this indicates that
their selling prices are rising relative to labor costs, which incentivizes them to use more
labor in proportion to capital when creating their products. This definition is supported by
a few notable Austrians such as Roger Garrison and Jesus Huerta De Soto. In Money,
Bank Credit, and Economic Cycles De Soto givers an in depth look at the business cycle,
and explains the difference between a progressing economy based on increased
savings from citizens, and a credit expansion from the fractional reserve banking system
which causes a temporary boom. The Ricardo Effect has an essential role in both
scenarios. In the first, citizens save more money based on their own time preferences
without any interference from the central bank. This means that the money supply is
inelastic, or fixed. Increased savings implies decreased consumption. Lower interest
rates(and the lower natural rate of interest) will ensue, which increases investment in the
economy. With consumer spending reduced, the prices of consumer goods falls, while at
the same time with increasing investment, labor increases in productivity due to more
capital being used. With labor becoming more productive, while at the same time the
prices of consumer goods falling, wage rates rise both in nominal and real terms. The
latter are what entrepreneurs supposedly follow when looking at labor costs, and since
these are now higher, businesses prefer to use more capitalistic methods of production
due to lower borrowing costs and hence, higher potential profitability.

In the artificial boom story, the Ricardo effect works in the opposite direction.
Rather than an increase in voluntary savings by citizens, the banking system injects
"unbacked" credit by lowering interest rates below the unobservable natural rate. This
increases investment during an economic slowdown or recession. Income in the form of
rents, profits, and wages rises. The main difference between this story and the former is
that consumption has not fallen. While interest rates have fallen, the demand for
consumer goods on the part of citizens has not. This creates a mismatch between social
time preferences and the natural rate. The owners of the factors of production have
increased incomes, and since their chosen demands have not actually changed relative
to before the increase in credit, consumer goods industries are stimulated, followed by
their prices rising. In the former situation, there was a temporarily reduced supply of
consumer goods since the desire to spend on longer term investment rose. This was met
by a more than offsetting reduction in the demand for these final goods, allowing their
prices to fall and real wages to rise. Currently the demand for consumer goods does not
fall, thus their prices rise ahead of wages. This causes a fall in real wages, and the use of
more labor-intensive methods of production since the cost of labor has fallen in terms of
their percentage of profits. This is the process Hayek first described from his book
Monetary Theory and the Trade Cycle and elaborated further in Profits, Interest and
Investment. While this version of the Ricardo effect seems quite plausible, it has been
met with considerable criticism. Perhaps that is why it has not become a part of the
canonical Austrian theory of the business cycle. Economist Tyler Cowen has attacked
this mechanism of "cyclical dynamics" on several fronts which have not been formally
answered. We will use his criticisms as a way to formulate a more robust exposition of
Hayek's Ricardo effect.

The Problem of Real Wages

In his book Risk and Business Cycles: New and Old Austrian Perspectives,
Cowen makes the case that Hayek's argument was flawed in terms of both theory and
history. His major reason for rejecting the Ricardo effect has to do with the motion of real
wages throughout the boom. Hayek made the case that real wages would fall after a
period of time following the expansion of credit in the loan market. What the evidence
finds, however, is that real wages are procyclical and actually rise during the boom. This
directly contradicts the notion that reduced wages in terms of selling prices will incite
businesses to substitute labor for machinery in production. He states:

Some real wages may rise during the move towards longer term investments, as discussed above
in the section on forced savings. Workers in capital-intensive industries may find their wages rise more
quickly than do the prices of consumer goods (Hicks 1967); the net effect depends on the relevant
elasticities. To that extent the Ricardo effect does not reverse the capital-intensive investments.

While this is particularly devastating to the mechanical understanding of this

position, it is not a fully accurate depiction of what happens. This is due, again, to
Hayek's unclear and patchy descriptions found in most of his major books on business
cycle theory. Even in his most detailed and lucid writings on the topic in Profits, Interest
and Investment, he makes two mistakes. He unfortunately adheres uncharacteristically
to the concept of real wages in the aggregate, and wrongly asserts that they fall during
the cycle:

For the present we shall just take it as a fact--and it is probably one of the best established
empirical generalizations about industrial fluctuations--that at this stage prices of consumers' goods do as a
rule rise and real wages fall.

However if we take a more nuanced reading of Hayek, as well as a more

business-oriented or microeconomic perspective, we can see that Cowen's argument is
ineffective. The most important thing to note is that businesses do not concern
themselves with what economists call real wages. These are calculated by deflating the
nominal wage(rate) with the price level. There are a few shortcomings to this concept
that we will cover briefly before Cowen's point can be critiqued. First, there is no such
thing as a 'price level'. There are only individual prices, which are more relevant to
estimating profits relative to business costs. This also applies to wages. There are
thousands of different jobs and industries that employ a multitude of employees with a
different range of skills and pay. In essence, real wages are too much of an aggregation
for entrepreneurs to be able to calculate their profits accurately. And second, real wages
in this context are not relevant because businesses look more at their own profits
compared to their employees' wages, not the rate of consumer goods price increases,
also known as the consumer price index (CPI). This is because the capital structure
consists of many businesses that are higher up in the supply chain, or in earlier stages of
production before consumer goods industries. In being consistent with Austrian theory,
both nominal and real wages (and therefore costs) do rise during the credit expansion
and ensuing economic boom. Unfortunately, Hayek did not seem to recognize that real
wages are not the issue, and wrongly insisted that they would always fall. In fact, he saw
the case of rising real wages as a 'special' case, and not the general case in his later
work.

Rates of Turnover and Uncertainty

The problem of real wages aside, businessmen do not look at real wages in any
way when calculating labor costs. What they look at closely is the relative ratio of prices
between specific wage rates that are based on the marginal productivity of labor and
marginal profits. In a typical business cycle which usually lasts for several years,
entrepreneurs are focused on relatively short run production. During this period of time,
only nominal prices are relevent. Calculating the real value of goods or labor is only
useful for comparisons over time. Therefore there is no need to adjust for inflation,
according to business accounting standards. The relative price of labor is compared not
only to the respective employer's appraisement of a given employee's marginal
productivity, but also in relation to sales and profits. This ensures that workers that are
being hired are more than compensating for their wages earned. But this need not be
falling as a percentage of final aggregate consumer prices or profits. Real wages can be
rising while consumer goods profits increase during the late stages of the boom. What
truly matters to businesses in terms of whether to hire more workers or increase the use
of capital in production is the element of time, which often supercedes their desire for
productivity due to the possibility of increased profit. This is true due to the concept that
Hayek referred to as 'turnover'. It is a businessman's turnover, and refers to the period of
time it takes an entrepreneur to get back his initial investment, in value terms.

Thus the rate of turnover is the number of times an amount of money that is spent
on a particular investment is earned back through revenue as a result from that specific
input. The profit earned on the turnover of a given amount of labor is the difference
between the specific wages and the marginal product that labor. If the price of a product
rises the turnover profit does as well regardless of the length of said turnover. However
the time rate of profit increases proportionately more for labor used in shorter periods of
turnover than in longer periods. A simple numerical example of this is explained by Tyler
Beck Goodspeed:

Defining the per annum percentage return (Hayek's time rate of return) as the "internal rate of
return," I, and the profit margin, M, as the proportional gain on each turnover, we then have the relationship
I=TM.
From this relationship, it is easy to compute the effects of a rise in price margins on the internal
rates of return of various inputs, depending upon the rate at which they are "turned over." For instance, if
the price of a product rises by 2 percent, then, holding input prices constant, there will be an addition to the
profit margins earned on each turnover equal to the amount of the rise. The firm turning over its capital
twice a year will find its profit margin increased from 3 to 5 percent, while that of the firm turing over its
capital once a year will rise from 6 to 8 percent. Multiplying these profit margins by the corresponding rates
of turnover, we obtain the new internal rates of return of 10 percent per annum for the former and 8 percent
per annum for the latter.

As we can see, the technique that has a higher turnover also obtains a higher
profit per unit of time. This applies to labor and is called the time rate of profit. Notice the
technique that has the lower rate of turnover has a higher profit margin of 6 percent at
first than the technique with the higher rate of turnover at 3 percent. When profits
increase, the latter yeilds a 10 percent rate of return and the former 8 percent. This will
cause entrepreneurs to use less capitalistic methods in production. Toward the end of
the boom, consumer good demand rises and so do profits in the industries that produce
these goods because of the increased employment of workers in the earlier stages of
production. The producers of consumer goods want to increase production to meet this
growing demand. They can either hire more labor to do it or they can buy more
equipment. The important point here is that labor as an input has a considerably shorter
turnover than capital. Put differently, it takes more time for capital to get a return on
investment in monetary terms than it does labor. Thus, in the short run, businesses will
choose to use more labor in the form of extra shifts or overtime hours (in addition to hiring
more workers). They can get more profit by increasing the labor/capital ratio in favor of
labor as demand for their products rises. This addresses Cowen's other point that the
Ricardo effect emphasizes capital and labor substitutability, rather can complementarity.
Increased use of labor with existing (nondurable) machinery can and may occur if the
turnover of labor is high relative to more capitalistic production methods. This does not
pose a problem to the Austrian position as he claims. Calculating real wages is
something economists do in order to utilize data for determining appropriate policies to
mitigate stagnating wages, lower incomes, unemployment, and declining economic
growth. This was a mistake even Hayek made in some of his work on the topic, and
explains the misinterpretation critics have for the Ricardo effect. The business world, on
the other hand, looks simply at nominal profits and revenue. They most often do not
account for inflation-adjusted prices during the boom, not do they pay attention to
aggregate nominal or real wages.This dovetails nicely with other aspects of Austrian
Business Cycle theory which will be discussed later.

Furthermore, since the Ricardo effect usually begins at the peak of the boom, the
economy is at full capacity or close to it. This could also allow businessmen to use less
efficient and older equipment which were previously laying idle or in storage. What will
result is lower productivity, particularly in the later stages of production closest to
consumption. Since employing more labor with the same amount of capital, or using less
productive capital equipment causes diminishing marginal returns, this lowers overall
productivity in the economy. But businessmen are willing to sacrifice productivity for the
sake of short terms profits. One clarification is necessary, however. Procyclical real
wages are a sign of increased productivity, but apply mostly to labor in the early stages of
production, not in consumer goods industries. This is because artificial credit expansion
brings about increased technology and other more 'roundabout' methods of production
which increase productivity. The same is not true in consumer goods for the reasons
made above. Perhaps the reason why real wages rise only modestly during the upswing
of the business cycle is due to inflationary pressure from price stabilization policies. With
increasing growth in the economy due to a supposedly proportional increase in
productivity, real wages should rise in conjunction with growth. However this is often not
the case. Thus real wages will remain procyclical, but the rate of their growth may be
suppressed considerably due to inflationary policies by the central bank. This explains
the phenomenon of comovement in terms of wages and production that concerns
Cowen. It is not real wages that businesses consider in the short run during the business
cycle, but nominal wages and prices. Real wages can rise during the later part of
economic expansion due to increasing productivity, but they will fall relative to revenue.

Another implication of the Ricardo effect is the relevance of uncertainty, a concept

emphasized by both Austrians and post Keynesians. Prospective profits are always
more attractive in the short run, particularly in production stages that rely more heavily on
higher social time preferences (consumer goods industries). With the risk of losing
business, both in the near future with short run sales and the further future with
permanent customers, entrepreneurs have a higher incentive to 'make hay while the sun
shines'. In other words, immediately increasing production to meet short run demands
becomes a greater importance than long terms endeavors. Businesses may do this for
different reasons. As the boom continues and prices and profits increase in the
consumer good industries, entrepreneurs will not be certain as to how long this trend will
continue. Short term demand may be maintained for a considerable amount of time, but
it is less certain the longer the period under consideration. Thus businesses will have a
justified reason to increase short run production at the expense of capital spending,
reinvestment, or other longer terms advancements. Professor Fiona Maclachlan, a post
Keynesian, mentions this in her book Keynes' General Theory of Interest: A
Reconsideration. Since the future is uncertain and production has a significant time
element, the Ricardo effect proper, as Maclachlan calls it, leads to an increase in the
output of consumer goods since businesses do not know how long the increase in
consumer demand will last, so they take advantage of these short term profit
opportunities. While she finds several faults with Hayek's Ricardo effect, she agrees that
an increase in consumption demand will lead to a fall in investment, particularly in capital
goods industries.

The Substitution of Capital Across Time

A corollary to the notion of increasing production is what professor Simon Bilo

calls the intertemporal substitution of capital. This is not directly connected with the
Ricardo effect, but is related in terms of durabilty and capital maintenance. With lower
real interest rates, the present value of future income streams from early production
stages rises, while the relative profitability of the later stages falls, at least at first. This
causes nonspecific factors of production, primarily labor, to reallocate to those
processes. But, again, there is a time element which applies not only to Hayek's Ricardo
effect, but to Austrian Business Cycle Theory as a whole. Even after interest rates are
lowered (from an increase in the supply of money), this realloction of factors takes time.
This is due to the requirement of complementary and somewhat specific factors that
even earlier stages of production have yet to produce. While the number of potentially
profitabile stages may increase, those stages rely heavily on other inputs from the
current early stages. Hence, a given production process will require produced inputs
from the immediate preceding production stage. In addition, the marginal product of the
nonspecific factors in the later processes will be lower until this stage receives the
complementary inputs from the preceding stage. They will express lower demand for
these nonspecific factors until this accumulation proceeds. Thus there is a waiting
period, or a lag, between the time that interest rates fall, and output in these early stages
can be expanded. The earlier stages of production, such as manufacturing and mining,
will begin to bid for labor and capital from their employment in later stages. The
entrepreneurs from the later stages expect this bidding war from the relatively early
stages. This will mean future higher prices of equipment and especially labor, which is
more nonspecific in nature. Hence, higher costs. But again, this takes time, which gives
later stage entrepreneurs time to adjust their production plans as the structure of
production lengthens.

To the extent that consumer goods businesses use the same machinery, which
can be more durable, they will utilize this equipment more intensely. The reason for this
is that they expect higher costs in the future due to the bidding for nonspecific factors
(labor). The specificity of the durable capital means it cannot be reallocated easily, and
must be used in accordance with the complementary nonspecific labor. With the
increased cost of this labor in the future, businessmen see profitability being more
obtainable now rather than later. This can also be because of our earlier reason of
uncertain future demand for consumer goods. Thus they increase output now using
durable machinery, increasing prodution to the advantage of the lower cost of labor now.
This has several implications related to the Ricardo effect.

First, it explains the comovement in both consumer goods and investment goods
production during the boom. Despite Cowens' assertion that this should not happen in
the Austrian explanation of business fluctuations, we can see that since the increase in
investment takes time after interest rates fall, output in consumer goods industries tends
to increase due to uncertainty and the expected increase in the cost of labor. Second, it
acknowledges that real wages rise later in the boom which, again, shows that Cowen is
mistaken in his claim that the increase in real wages disproves the Hayekian notion of a
capital and labor switch in techniques during the business cycle. According to professor
Bilo, with the increase in demand for complementary nonspecific labor from the early
stages of production, real wages will rise in the future due to the previously mentioned
lag. And this rise in real wages does not mean that profits can not be maintained in the
later stages due to the relatively high rate of turnover for labor. While this seems to lead
to more capital use in both the early stages and later stages of production in the short
run, it will eventually lead to less capitalistic production methods in the future due to
capital depreciation which brings us to our third point.

Insufficient Capital Maintenance

In a simplistic two-stage model of the economy where only consumption and

investment are relevant, there is little room for intertemporal coordination which is the
focus of Hayek's business cycle theory. The problem arises when an increase of
consumer demand, and the production of consumer goods, decreases the funds
available for gross investment due to increased spending out of current profits on labor
and the overutilization of durable capital (and the possible use of less durable types). In
other words, with an increase in the output of final goods, capital consumption occurs
which causes the market value of equipment and machinery to fall. While the increased
utilization of durable capital with a given amount of labor does not in itself cause what
economists call capital consumption, it is the negligence of capital maintenance by
entrepreneurs that initiates the downward turning point of the business cycle. The typical
Austrian explanation is that this is due mainly to an illusory wealth effect that falsifies
economic calculation by market actors. But there is more to it, and neither the money
illusion nor depreciation due to the expected wear and tear of equipment are the main
culprits. When consumer goods industries are stimulated by the increased incomes of
workers and entrepreneurs in the early stages of production, businesses can choose to
increase production by either hiring more labor or employing more capital. Due to the
high rate of turnover of labor, as well as both the higher expected future cost of labor and
the uncertainty of perpetual demand, these businesses choose less capital-intensive
methods over more capital-intensive methods, to use Hayek's terminology. This has a
profound, though negative, affect on the 'intermediate' stages of production, which
produce machinery, equipment, and semi-finished goods which contribute to the
maintenance of the capital stock. These are often complementary to Mengerian higher
order goods, and since increased consumption(and investment) implies that these are
not being renewed, this means that the lack of capital expenditures reduces capital
durability as the cycle progresses. As Hayek explains:

There will, in the first instance, be the possibility of less perfect maintenance and attention,
makeshift instead of thorough repairs, shorter or fewer periods of laying off for inspection and overhaul,
which will reduce the efficiency and shorten the life of the existing machinery but may well be worth while if
current output can thus be increased. There will be, second, the possibility of outright nonrenewal, not, of
course, of essential parts of the equipment, but of the many auxillary laborsaving devices such as
automatic feeders and other gadgets performing operations which can also be done by hand. Third, there
will be the possibility of using obsolete or secondhand machines instead of new ones. Many older factories
have a certain amount of such old machinery for temporary use to meet peak-time demands or in an
emergency, for which it would not pay to keep a new machine in reserve. There exists in many branches a
supply of secondhand machinery which can be used in the same way. Fourth, and last, there will be the
possibility of replacing those machines that wear out by new but cheaper and less efficient ones.

It is important to remember that it is not the physical composition of different types

of capital and labor businesses have aquired in a given period that matter, but at what
proportion total current expenditures are spent on capital and how much is spent on
labor. Yet even if labor is chosen over capital in the later stages of production, is it
feasible that this labor could be used to produce more durable equipment or machinery
to maintain the capital stock so that a sustainable capital structure can progress. But as
we have seen, this is likely not the case. In fact, with higher spending by the later stages
on labor, less is being spent on capital maintenance and capital goods from the early
stages. Furthermore, with a decrease in demand for these capital goods by the relatively
later stages, the sales of these capital goods, and thus their prices, will fall initiating the
bust phase of the business cycle. Not only is this due to the fall in the prices of capital
goods in the early stages, but also the cost of labor rises because of the previously
mentioned bidding for labor services between the early and later stages in the supply
chain. Aggregate real wages rise toward the end of the cycle for which there is
considerable evidence. This equates to lower profits in the stages of production furthest
from consumption as revenues and sales fall but costs rise. As the bust begins, labor is
discharged and unemployment grows. This is because the quality and quantity of capital
equipment has fallen, which makes the demand for such complementary labor fall along
with output, making businesses lay off workers while the unusable capital becomes idle.
A recession follows.

Conclusion

The canonical version of the Ricardo effect can be supplanted by a more robust
model that emphasizes relative nominal wages, and their relation to sales and revenue in
consumer goods industries. This will determine the turnover rates of less capitalistic
methods of production. The uncertainty of future demand coupled with the certainty of
future higher labor costs due to competition reduces the durability and market value of
capital equipment. These forces bring a growing economy based on inflationary
monetary policies to a grinding halt. While the latter increases the likelihood of the bust if
it is assumed aggregate savings are not sufficient to compensate for the lack of renewal
in the capital base, the former changes the entire composition of capital to restructure
due to entrepreneurs' most important endeavor being the profit motive. Hayek deserves
credit for bringing this catalyst of recession to the forefront of Austrian Business Cycle
Theory. He also deserves criticism for not properly integrating it into a more general
explanation of capital theory, forming a comprehensive framework for further research.
Perhaps with less focus on real wages, which are a macroeconomic concept, and more
emphasis on specific wages in terms of profits, which are based on microfoundations,
economists can revisit the Ricardo effect with an open mind as to how an economic
boom can self-reverse merely through the price mechanism.