Download Socialist alternatives to capitalism II: Vienna to Santa Fe

Socialist alternatives to capitalism II:
Vienna to Santa Fe
by Duncan K. Foley∗
1
Top-down
Barone and Pareto tackled the problem of socialist allocation using simple
methods of the analysis of static economic equilibrium, centering on the idea
that competition will force an equality between marginal cost and price. In
the nineteen-twenties and thirties a group of mathematicians, statisticians,
and economists, including notably Abraham Wald, John von Neumann and
Oskar Morgenstern coalesced around Karl Menger’s “Vienna Circle”. Their
discussions laid the groundwork for the mathematical approach to economics
that dominated economic theory in the second half of the twentieth century.
In particular this work clarified the problem of how to keep prices from becoming negative through the application of “complementary slackness” conditions which allow for non-zero prices only when resources are fully utilized.
One result of this line of thinking was to situate the problem of economic
allocation of resources in the broader context of optimal control theory. The
mathematical theory of optimal control concerns the design of feedback mechanisms that can stabilize complex dynamical systems to achieve particular
goals. A paradigmatic example is the problem of anti-aircraft gunnery, which
became a key aspect of military combat in the Second World War. As Philip
Mirowski observes (Mirowski, 2002), the duel between the anti-aircraft gunner (who tries to aim at the point where he thinks the bomber will be when
the shell reaches the aircraft’s altitude) and the bomber pilot (who tries
Leo Model Professor, Department of Economics, New School for Social Research, 6
East 16th Street, New York, NY 10003 ([email protected]) and External Professor,
Santa Fe Institute. This talk was prepared for lectures at the Havens Institute at the
University of Wisconsin, Madison, April 6–7, 2011.
∗
1
to evade the shell through maneuvering the plane) inspired von Neumann
and Morgenstern’s zero-sum game theory and the “cybernetics” of Norbert
Wiener. (It is also the root of modern financial theory, where the problem
appears as the task of predicting where asset prices will be when a trader’s
transactions are actually consummated.)
One continuing tradition of thought takes the technocratic view that the
task of socialism is to identify and implement a socially optimal allocation
of resources and distribution of products. This is a “top-down” vision of
socialism, in which the participants in the division of labor surrender their
autonomy to a (presumably) benevolent centralized mechanism of social control. The development of sophisticated optimal control theory supports this
top-down vision by revealing mathematically precise characterizations of the
conditions for optimal allocation that are in principle generalizable to an
arbitrary number of produced goods and capable in theory of dealing with
problems of time and uncertainty. In this perspective, establishing socialism amounts to putting in place social institutions that can implement the
solutions of optimal control problems. The power of these mathematical
methods and the prestige of mathematics in the natural sciences disposed
many sophisticated thinkers to accept the eventual triumph of optimal control methods as inescapable. Joseph Schumpeter (Schumpeter, 1942), for
example, seems to have seen himself as fighting a doomed rear-guard action
on behalf of an increasingly obsolescent capitalist entrepreneurial impulse in
the face of an inevitably encroaching bureaucratic rationalization of social
production.
The top-down optimal control program gained momentum from two other
important developments of the nineteen-thirties that the Second World War
greatly accelerated.
One was Wassily Leontief’s input-output system (Leontief, 1966), a compilation of data on inter-sectoral transactions at a previously inconceivable
level of detail (on the order of dozens or even hundreds of economic sectors).
To the technocratic socialist it seemed only a short leap from Leontief’s tables
to implementable plans of production and distribution. Indeed, during the
Second World War, Leontief’s methods contributed significantly to resolving
real-life bottlenecks created by planned armament production.
A second was the invention of the digital computer by Alan Turing, von
Neumann, and others, a result of converging streams from the pure mathematics of proof theory, the combinatorics of code-breaking, devices for guiding artillery, anti-aircraft guns and aerial bombing, and rapid advances in
2
electronics. von Mises had felt confident in attacking the concept of centralplanning socialism in the nineteen-thirties on the ground that it was humanly
impossible to solve all the equations that characterize competitive market
equilibrium except through market interactions. But just a few years later
technocratic socialism had at its disposal a sophisticated mathematical representation of the allocation of social resources as an optimal control problem,
a rapidly developing computing technology precisely adapted to solving such
mathematical problems, and a rudimentary but generalizable input-output
data base on which to base calculations relevant to real-world allocation.
The constituency for developing this nascent technology of social planning extended far beyond head-in-the-clouds socialist idealists. Like nuclear
bombs, these social engineering innovations had played a centrally effective
role in organizing the successful war effort. (In fact, a good argument could
be made that the cumulative impact of the social engineering and planning
innovations on the actual prosecution of the war far exceeded the marginal
importance of nuclear weapons.) In the paranoid mirror-world of the Cold
War, the extension of rational planning institutions seemed to offer a critical
edge in the long twilight struggle.
Perhaps it was an awareness of this threatening consensus that panicked
Hayek into writing The Road to Serfdom, which conflates all varieties of
collective social action into a monolithic “collectivism” threatening individual
human liberty. In any case, it is not hard to see from a Marxist perspective
that the program of technocratic centralized rational allocation would run
into heavy weather with the capitalists themselves. The capitalist backlash
left its mark on the economics profession, in the effective suppression of
Lorie Tarshis’ Keynesian textbook and its replacement by Paul Samuelson’s
carefully sanitized neoclassical synthesis, and in unremitting and ultimately
successful pressure to marginalize and defund Leontief’s research program.
The top-down “hard” program of technocratic centralized rational allocation of resources guided by real data probably reached its high-water mark
in the late nineteen-forties, though it continues to exert a real influence on
political economic thinking and policy discourse to the present day.
2
Bottom-up
There is a curiously close connection between the mathematics of optimal
control theory and the mathematics of thermodynamics. Thermodynamics
3
is the somewhat unlikely but hugely successful branch of physics that deals
with systems that are too complex to be attacked directly by the methods
of dynamics. A paradigmatic example is the behavior of molecules in diffuse
gases such as the earth’s atmosphere: a relatively small volume of such a
gas contains such an enormous number (on the order of 1023 ) molecules that
it would be beyond human capability to calculate their dynamics in detail
(shades of von Mises’ critique of central-planning socialism). The ingenious
application of statistical reasoning to this type of problem leads to very
powerful and useful predictions about the macroscopic behavior of gases,
regardless of the details of their molecular motions. Thermodynamic methods
lead to deep insights into an enormous range of complex phenomena, though
the results are generally not as simple as the behavior of diffuse gases.
When the interactions of the constituents of complex systems are more
intricate than the collisions through which gas molecules exchange energy,
such as commodity purchases and sales, the system can produce macroscopic
emergent, adaptive behavior. Except in unusually tractable cases like diffuse
gases, it is possible to study these emergent properties of complex systems
only through simulation methods of limited precision. There is considerable
reason to believe that the division of labor and the capitalist economy are
best understood as complex systems (see Foley, 2003; Duncan K. Foley (ed.)
and Peter S. Albin, 1998). In fact, as I have suggested, classical political
economy can be understood as an attempt to reason through the implications
of regarding commodity production as such a complex system.
The last half of the twentieth century saw an upsurge in interest in
bottom-up self-organized systems, such as ant-hills and bees’ nests, bird
flocks, computer networks, unsupervised learning algorithms such as neural nets, the capitalist economy, possibly the human brain and a host of
other similar examples. Several features of these bottom-up systems draw
the attention of systems thinkers.
For one thing, compared to top-down optimal control systems, bottomup self-organized systems tend to be more “resilient” or “robust”. Self organized systems often (though not invariably) continue to function (possibly
in a degraded mode) despite the disruption or even destruction of important
subsystems, and in important cases can recover from such damage spontaneously. Top-down systems can be built with considerably “redundancy”,
but tend to be more vulnerable to disruption of key elements, particularly
the feedbacks that implement their optimal control properties.
Another intriguing property of bottom-up self-organized systems is that
4
although they are not designed to achieve optimal performance, they often
do perform surprisingly well. Ant-hills, for example, are remarkably efficient
in locating and exploiting food sources.
Bottom-up self-organized systems also exhibit a high degree of adaptability to new situations. Optimal control systems tend to be optimized for some
particular context in which they are designed to operate. Self-organized systems can adapt to a wide range of environmental changes (though there are
typically limits to the magnitude of shocks any such system can survive).
Capitalism itself is a good example of all these features of spontaneously
organized bottom-up systems: historically capitalist institutions tend to reproduce themselves even after the destruction wrought by wars, revolutions
and financial meltdowns; capitalist allocation of resources approximates efficiency to some degree; and capitalism has proven to be highly adaptable
in the face of massive environmental changes (many of which, such as world
population growth and technical innovations, arise from the dynamics of
capital accumulation itself).
The vision of the economy as a complex system gives rise to a “bottomup” understanding of how commodity production is organized, a la Hayek.
It also suggests a parallel bottom-up vision of socialism as arising from the
spontaneous organization of production through some framework of institutions different from private appropriation and exchange of products.
This bottom-up vision of social organization resonates with important political currents of the late twentieth century. The “New Left” movements of
the nineteen-sixties rebelled against the centralizing and regimenting tendencies of “Old Left” socialism and communism by calling for decentralization,
participation, and the political primacy of individual freedom and expression.
The Left, such as it is in the contemporary world of globalized capitalism, is
much more attracted to the vision of spontaneous political expression than
to the unpleasant chore of organizing political institutions.
This enthusiasm for spontaneity without central coordination or direction
has attached different parts of the Left to various economic experiments,
plans, philosophies, and models. Some of these, such as micro-credit institutions, amount to very minor modifications of the capitalist commodity
system (or could indeed be regarded as systematic extensions of the logic of
the commodity system).
One idea which flowers perennially on the Left as an alternative to capitalism is workers’ control. In its simplest form this vision accepts the
commodity-money framework of the organization of the division of labor,
5
but proposes to replace capitalists as the organizers of production at the
point of production by workers. Enterprises would be owned by collectives
or cooperatives of workers, who rent or borrow capital and constitute what
economists call the “residual claimants” of commodity production. Workers’ control has features of spontaneity, adaptability, and resiliency, partly
as a result of inheriting them from capitalist-commodity production itself.
Like capitalist firms, worker-controlled enterprises could form and dissolve
in response to market signals to accomplish a decentralized reallocation of
resources as society changes and evolves. It is possible to imagine capitalallocating institutions that would permit the free formation of new enterprises and hence a substantial degree of competition. Worker cooperatives,
as producers of commodities, can co-exist with (and compete with) capitalist
firms, so it is possible to imagine an evolutionary process of transformation
of a capitalist economy into a socialist economy, if the content of socialism
is workers’ control. The vision of workers’ control, by accepting the broad
framework of commodity production and money, solves many of the puzzles
that haunt more radical visions of socialism.
In fact, workers’ cooperatives do function as a subsidiary sector of most
capitalist economies, and even have flourished in some circumstances. They
have a long history in the Basque country of Spain, the American midwest,
and Italy, among many other places. For some years a major European economy, Yugoslavia, was committed to the workers’ control form of organization
of social production, and its economy functioned largely on this basis, providing some invaluable historic experience to shed light on the actual workings
of such a system. Economic theorists (for example Vanek, 1970) have produced a rigorous analysis of the properties of workers’ controlled enterprises
and economies.
Some points stand out in surveying the history and experience of workers’ control. One is central to the economic analysis: the role of the actual workers in enterprises as residual claimants of enterprise income biases
worker-controlled enterprises against hiring because new workers dilute the
claim of existing workers to the surplus revenues of the enterprise. This
phenomenon in turn gives rise to pressures to differentiate claims to surplus
revenues among different strata of workers. When worker-controlled enterprises become very successful, the resulting hierarchy of workers begins to
resemble that of capitalist firms. When the contradictions between the juridical form of organization of the enterprise and the substantive claims of
different strata of workers to the surplus revenues become acute, it is com6
mon for worker-controlled enterprises to privatize themselves spontaneously,
converting into traditional capitalist firms. A second point is that despite
the demonstrated capability of worker-controlled firms to compete successfully with capitalist firms under particular circumstances, there has been no
tendency for worker controlled enterprises to grow spontaneously to crowd
out capitalist firms in any economy: workers’ control remains a relatively
small sector even in societies where tradition and practice favor this form of
productive organization.
Another noteworthy bottom-up alternative model to capitalist firms as
organizers of production is the “open source” or “peer production” movement
that has been highly successful in developing computer software and generalized to some other similar fields such as pharmaceutical drug development
and gene technology. Peer production is a decentralized, spontaneous form
of organization of productive resources, through which a group of interested
and participants contribute time, energy, and capital to the production of socially useful products. TeX and LaTeX, type-setting programs widely used
for scholarly and scientific writing, Apache web servers, and the Linux operating system are products of peer production. The software created through
peer production, aside from being “free”, works enough better and more reliably than for-profit competing products to dominate important branches of
computer technology. Peer production is a quite convincing demonstration of
the feasibility (and even superiority) of organizing a technically sophisticated
division of labor through spontaneous and primarily non-commodity/money
mediated social relations of production.
There are some major obstacles to peer production as a universal model
of social production. The capital involved in software development is relatively small compared to human labor input. Much of the capital required to
carry on software development, mainly computers and network connections,
are cheap and already available to participants in open source projects. (The
network itself, the internet, is provided socially through a complicated mixture of top-down and bottom-up institutions.) The participants in open
source projects typically have “day jobs” that provide them with incomes
to pay the rent and keep food on the table through participating in the
commodity-capitalist organized division of labor. Participants in peer production are motivated by their own needs and interests, like more humble
hobbyists, as well as by powerful forces of social approbation and recognition for achievement. Many “volunteers” in peer production are actually
participating at the instruction of their capitalist for-profit employers, who
7
see commercial advantages in being involved in open source program development. The products of many peer-production efforts are what economists
call “non-rival” goods like programs or information for which the marginal
cost of provision to an additional user is negligible compared to the fixed
cost, for example, of program development. If we think about generalizing
peer production to sectors like food or textiles or construction many contradictions are immediately apparent.
The explosive increase in computing power and the emergence of computer networks potentially solves problems for the bottom-up approach to
socialism just as it does for the top-down approach. The central issue in
organizing a complex social division of labor is, as Hayek pointed out, information. (That is not the only issue, of course, since incentives and coordination of effort at the point of production also pose important questions.) The
capitalist-commodity system depends to a considerable degree on markets
and the formation of market prices to generate and diffuse the information
required for the organization of the division of labor. A socialist system
that either de-emphasizes markets or dispenses with them altogether will
have to depend on other sophisticated decentralized methods of processing
and diffusing large quantities of information. The availability of powerful
computers and networks at least holds out possibility of creating interactive
systems that could function in place of markets in this respect.
The top-down vision of an optimally controlled society stemming from
the Vienna Circle’s mathematicians has morphed through thermodynamics
into a Santa Fe Institute bottom-up vision of a self-organized, decentralized
system of social dynamics.
3
What do we talk about when we talk about
socialism?
Socialism is the dialectic negation of capitalist commodity production, and
it is inevitable that the discussion of socialism is motivated by the desire
to mitigate or eliminate features of capitalist commodity production that
socialists dislike. Since there are many things about human social interaction
to dislike, this discussion can become quite diffuse. (Can socialism eliminate
war? racial and ethnic discrimination? gender oppression?) In order to
focus our thinking, let me list three key aspects of capitalism that play a
8
central role in socialist discussions of radical economic change: commodity
production and inequality; private ownership of the means of production;
and wage labor and exploitation.
3.1
Commodity production and inequality
In principle, as neoclassical general equilibrium theory asserts, we could imagine a commodity production system in which the distribution of wealth and
income were highly egalitarian. In the traditional economic Walrasian model,
for example, inequalities in the final allocation of produced goods and services
arise only from inequalities in the ownership of productive resources, like labor, means of production, and land. One popular fantasy among egalitarianminded economists and other social scientists is a society of decentralized
commodity production with equally distributed ownership of productive resources, which would magically reconcile the goals of fairness, efficiency, and
individual liberty.
The first point to make about this idea is that it is an historic empty
box. One of the better-attested “regularities” of human social life is that
commodity production is associated with a highly unequal distribution of
income and an even more unequal distribution of wealth. Typically the top
fifth of incomes and the top tenth of wealth amount to about one half of the
totals in commodity producing societies.
There are several plausible explanations of just why this sharp degree of
inequality, which is perhaps the most salient experiential feature of capitalist
society, exists.
Human beings differ, for example, in their genetic endowments. Perhaps
these enormous differences in economic outcomes reflect these differences.
Some highly visible instances, such as the high incomes of sports stars, lend
credence to this notion. It is interesting that the classical political economists
tended to view inborn differences in human talent as negligible in their thinking about human society (and certainly neither Smith nor Ricardo were radical critics of capitalist social relations). One reason to doubt this explanation
of observed inequality is that independent non-economic measures of human
capacities tend to show rather small, normally distributed differences among
individuals. It is not easy to understand how the commodity production
system could systematically amplify these differences and transform their
statistical properties into the extremely skewed exponential and power-law
distributions of income and wealth we observe.
9
The inheritance of wealth, either directly in the form of ownership of
resources, or indirectly through inter-generational transfers in the form of
education and other social advantages, is another popular explanation of
the highly skewed distribution of income and wealth. But bourgeois revolutionary values are hostile to inherited privilege, and not very consistently
supportive of concentration of wealth through inheritance. Large family estates are regularly dissipated through the multiplication of descendants, mismanagement, and extravagant consumption. It is not easy to produce models
based only on inheritance that predict extreme concentrations of wealth and
income. Furthermore, some quasi-experiments dramatically call this hypothesis into question. In Eastern European countries in the nineteen-nineties, for
example, privatization of very substantial state assets was sometimes based
on distributing “vouchers” or shares on an egalitarian basis and allowing
the recipients to sell their claims (or buy others’ claims) freely according to
the rules of commodity exchange. Within a very short time, much shorter
than the dynamics of inheritance would imply, the ownership of these claims
became highly concentrated through speculation. Similar outcomes are common after land reforms and other redistributive measures. This suggests that
even if it were possible hypothetically to equalize ownership of productive resources completely at one moment of time, there are powerful equilibrium
tendencies of commodity exchange that would tend to reproduce a highly
unequal distribution of income and wealth.
These equilibrium tendencies are, in fact, observed in thermodynamic
systems with large numbers of degrees of freedom, which tend to reproduce
exponential and power-law distributions. It seems likely that mass, decentralized social interactions, such as market exchange, share these powerful
equilibrium tendencies. The Walrasian theorem that final allocations merely
reflect the underlying distribution of ownership of productive resources depends crucially on the incorrect assumption that all market transaction take
place at equilibrium prices. When we make the more realistic assumption
that commodity owners very often exchange at disequilibrium prices as they
grope their way toward the determination of the equilibrium price, abstract
models of commodity exchange imply skewed, highly unequal distributions
of income and wealth (Foley, 2010).
These considerations suggest that Marx was correct in arguing that the
socialist reforms of the commodity system of exchange could not by themselves transform social relations. Similar conclusions flow from both the theoretical analysis of schemes like worker-controlled enterprises (which predict
10
sharp inequalities) and even the historical experience of Yugoslav society,
which experienced sharp increases in regional and class inequality during its
experience with workers’ control.
3.2
Private ownership of the means of production
Classical Marxist analysis tends to characterize capitalism in terms of the
private ownership of the means of production. In Marx’s theory of capitalism, the “monopoly” (in a class sense, not in the usual economic theory
sense) control of means of production by a small capitalist class is a necessary condition for exploitation (though not a sufficient condition, since the
availability of exploitable wage laborers is also essential).
Thus classic socialism has centered its political energies on the “expropriation of the expropriators” through the elimination of private property in
means of production.
The elimination of private property in the means of production has ramifications in various dimensions. One question is what happens to any incomes
(profits) that accrue to owners of the means of production. Another question
is how ongoing or new productive enterprises get access to means of production required to carry on production. A third, related, question is how scarce
means of production are to be allocated among competing potential uses.
The experience of the Soviet experiment and various state socialist episodes
make it clear that eliminating private ownership of the means of production
in and of itself does not eliminate exploitation of the direct producers. In
the Soviet case there is strong evidence that a class division of society arose
between members of the ruling elite who practically controlled the means
of production on the one hand and direct producers on the other. A case
can be made that the collapse of the Soviet system was the result of classic
Marxist pressures to bring juridical relationships into line with substantive
social relations of production by formalizing the “private” control over productive resources. (The historic transformation of major parts of the Soviet
nomenklatura into the nascent capitalist class of post-Soviet society is possible confirmation of this view.) When western European nations flirted with
state ownership of the means of production, the facilities nationalized were
often in declining industries where the state enterprises could not make profits anyway. The motive of nationalization in these episodes was more often
the preservation of jobs threatened by competition than redistribution of
profits among social classes.
11
The relevance of private property in the means of production is closely
related to the distributional questions just discussed. If the distribution of
ownership of productive resources were equal, the system of private ownership
of means of production in itself would not pose so many social drawbacks.
From the point of view of bottom-up, decentralized visions of socialism,
private ownership, or at any rate some type of private control over means
of production could be a means of relatively free access of new or ongoing productive enterprises to means of production. (The trick would be to
prevent a market in means of production, or even some system of barter
exchanges, from leading to concentrations of wealth. An appealing idea is
to set things up so that productive enterprises could claim control only of
means of production they actually used, but there might be incentives to
evade such restrictions, which are difficult to enforce.)
3.3
Wage labor and exploitation
Marx’s analysis of capitalist exploitation is based on his original and powerful analysis of the social relation of wage labor. Marx based this analysis
on the distinction between “labor-power”, the human capacity to perform
useful labor, and hence add value to other inputs, and “labor”, the actual
expenditure of human capacities to achieve useful production. There are two
fundamental aspects to this division.
On the one hand, the distinction between labor-power and labor illuminates the source of surplus value in the capitalist system. The value of
labor-power, the labor-time equivalent of the wage paid, on average is smaller
than the value labor creates. This difference is the source of surplus value (including profits, rents, interest, and the compensation of unproductive labor)
for the system as a whole. From this point of view it looks like a no-brainer
for socialism to eliminate the wage labor form as a way of eliminating exploitation.
On the other hand, however, the distinction between labor-power and labor points to a critical issue in the organization of the division of labor. The
capitalist has to pay for labor-power whether or not any useful productive
labor turns out in the end to result. Thus the capitalist faces the problem of
extracting labor from labor power, which involves both motivating and controlling the behavior of the wage laborers. Historically the wage-labor form
appears to have been stunningly successful in these respects. Labor productivity, partly as the result of technical change, but also because of high levels
12
of labor discipline, regularity, and skill acquisition, has increased enormously
over the years of capitalist domination of social production. In many job
contexts, the key element in the extraction of labor from labor-power has
been the threat of firing. As we have come to understand through the economic analysis of principal-agent interactions (see, for example Bowles, 2004)
this mechanism works because capitalist employers offer a wage premium, or
rent, to workers above their prospective incomes if they lose the job.
The problem for socialism is how to get rid of the exploitation inherent
in the wage-labor form without also losing the enormous social advantages
of labor discipline and motivation the wage-labor form carries with it. In
the Soviet era this dilemma was discussed under the rubric of “moral and
material incentives”.
It seems to me reasonable to imagine that a deep transformation of social relations of production would lead to parallel deep transformations in
the behavior and feelings of human beings. For one thing, even where the
dominance of capitalist institutions and bourgeois ideology is overwhelming,
people themselves do not completely internalize the laws of commodity production: they worry about the environment, the future of their and other
people’s children, and the welfare of other people. Capitalists themselves, of
course, highly prize workers who internalize high standards of productivity
and cooperation without the stick of threats of dismissal. (At least they like
these characteristics in the workers they actually have hired, though they
would probably prefer their competitors’ workers to be responsive to purely
material incentives.) It is a bit too easy, however, to wish away the principalagent problem of bourgeois economics as irrelevant to a socialist society. The
conflicts of interest, or moral hazards, a socialist organization of production
would face might be very different from those of capitalist-commodity production, but they are unlikely to disappear altogether. Perhaps, as in peer
production, socialist producers might be motivated more by the positive approbation and recognition of their colleagues than by fear of job loss, but
socialists had better have some idea of how the social psychology of productive effort is going to work.
4
Fantastic conclusion
To bring this talk to an end, I will indulge in the luxury of fantasizing
about some alternatives to capitalist-commodity production. There are some
13
ground rules for this exercise.1
This is speculation about changing what Marx calls social relations of production. In thinking about alternatives in this (very broad) range of issues
I am not required to consider every social problem (foreign policy, religion,
education, political institutions, mental illness, drug abuse, culture). In general, the basic supposition is that life goes on as usual. People get up, go to
work or school, form families, reproduce, age, consume, and so forth. There
are natural disasters and domestic and foreign conflicts, regional divisions,
and the like. But it is not fair for me to wish away critical aspects of the
problem of organizing a complex division of labor to sustain the type of technologically sophisticated society post-industrial capitalism has created, with
relatively high levels of consumption and economic security. For example, it
would not be very interesting for me to posit a system of universal abundance
of use-values created by some imaginary robot work force, and then to spin
a vision of the resulting glittering possibilities for human social life. It would
also not be of much interest for me to ignore the problem of a “transition”
by assuming a complete self-contained and self-reproducing alternative system. A socialist fantasy alternative has to include the notion of a “mixed”
economy, in which commodity-capitalist social relations continue to prevail
in some sectors, or regions. The socialist sector of the economy has to include
methods for dealing with a capitalist-commodity other. Furthermore, I am
most interested in ideas that would allow a socialist alternative to co-exist
with (and in a larger sense compete with) capitalist-commodity social relations indefinitely. Thus it is not fair for me to assume some “revolutionary”
transformation of the whole society (even with the dystopian backdrop of
a looming or ongoing ecological crisis) to resolve basic problems of social
relations. It is equally uninteresting for me, however, to ignore the deep hostility of bourgeois ideology to radical alternatives to capitalist-commodity
production. I am interested primarily in how ordinary, regular suburbanurban people who by and large have jobs, incomes, and more or less stable
lives through the capitalist-commodity system might transform their social
relations of production.
It is not going to be hard to criticize the particular fantasy I put forward. The main function of such exercises is to induce dialectic thinking
by making concrete assumptions that can be knocked down. On the other
For a survey of other discussions in this vein see Devine (2002), kindly brought to my
attention by David Kotz, and Wright (2010).
1
14
hand, it is not in the spirit of this kind of exercise to invoke “human nature” in one or another of its many incarnations as an objection; or to wish
away through assertion the fact that capitalist-commodity production has
many parallel unresolved contradictions. It is not reasonable to insist, for
example, that opportunistic behavior would be any less mixed with socialand other-regarding behavior in a socialist fantasy than it is in capitalistcommodity reality. In fact, it seems reasonable to suppose that a change in
social relations could re-balance human behavior towards others at least at
the margin. It also seems reasonable to suppose that people are grown-up
enough and aware enough of the facts of economic life to understand that
at an aggregate level consumption depends on production, and that they are
willing to accept social responsibility to work productively as a condition of
consuming.
4.1
Lifenet
So let us imagine an alternative set of social relations, which I will call
“Lifenet” which can organize social production to meet at least some part of
the spectrum of human needs outside capitalist-commodity institutions.
Lifenet production is organized through peer-production enterprises. People start or join existing peer-production initiatives without any direct material compensation. The output of these enterprises is made available freely
to participants in the system through a Lifenet distribution system, just as
open-source software is published freely through open source licenses. One
key point here is that users or consumers of Lifenet outputs agree not to
re-sell them as commodities for money. The aim is to create an alternative
system of production and distribution that can be effectively separated from
capitalist-commodity production.
Participants in Lifenet enterprises have a Lifenet account maintained in
a central Lifenet database. The Lifenet account records individual contributions to Lifenet enterprises, and also individual withdrawals of Lifenet
products. Thus an individual who participated in a food production enterprise would have that participation recorded in her Lifenet account; if she
received food or clothing or software from Lifenet distribution centers (perhaps something like food coops or farmers’ markets today) these withdrawals
would also be recorded in the Lifenet account. (One can imagine something
like a credit card system to maintain these accounts automatically.) The
purpose of the Lifenet account would not be to enforce an exact balance
15
between individual contributions to the production system on the one hand
and withdrawals through the distribution system on the other. It would,
however, allow the system to monitor the balance of production and distribution as a whole, and to identify and discourage or prevent possible abusive
over-drawing on distribution. (If some level of the individual accounts were
publicly accessible, peer pressure could play a major role in this respect.)
The premise of the accounts is that Lifenet as a whole is solvent, and has
products available for distribution.
A Lifenet system is going to work only if the individuals participating and
the system itself develop an ethos of thrift, prudence, and waste-aversion.
Lifenet must prudently keep large reserves of products to secure its ability to respond to contingencies (natural disasters, local breakdowns of the
system) flexibly and reliably. The participants in the system must use its
resources sparingly and thriftily, both as producers and consumers. The idea
of “enough” has to pervade the system from top to bottom.
Now, as we have seen, the basic issue of a division of labor is what feedback
mechanism can adjust shortages and surpluses of particular products. (People cannot live exclusively on high-performance open-source software if there
is no food or drink in the distribution system.) In the capitalist-commodity
system the market provides this feedback (not perfectly as in the fantasies
of attained neoclassical equilibrium theory, but at least well enough to keep
complex societies going most of the time). One of the goals of the Lifenet
system is to avoid the inevitable redistribution and concentration of money
inherent in the Hayekian system that requires individuals to have a stake in
market transactions to elicit their private information. Thus it would not be
acceptable (in this fantasy logic) to charge users prices for goods and services
and credit enterprises with revenue for providing them. But here we are with
the Lifenet food distribution coop in Berkeley short of wine, and the Baltimore shoe distribution outlet glutted with shoes (not to speak of the likely
chronic oversupply of Lifenet software). Or, even more to the point, with a
society-wide shortage of food and oversupply of clothing.
Addressing this problem requires a signal to producers to shift away from
the over-produced outputs and toward the under-produced outputs. (The
capitalist-commodity system signal is excess profits in the shortage sector
and losses in the glutted sector.) These signals might be provided simply
by producers monitoring the Lifenet accounts and noticing that the stocks of
wine are running low, while the stocks of shoes are building up. Of course it is
not very easy for shoe producers in Albany to shift to wine production. Fur16
thermore, decentralized adjustment to observed shortages or over-stocking
can easily lead to unstable responses.
One recourse for Lifenet might be to set up an interactive computer game
with real-time inputs based on the Lifenet accounts where a kind of virtual
market could provide price signals to producers and users. Simulation of this
kind could help avoid instabilities in adjustment.
The most serious problem for Lifenet would be the reluctance (or inertia)
of producers of products in chronic excess supply to make a major investment
in new productive capacities, which might involve relocation and other social
costs. This type of adjustment would test the motivation, imagination, and
willingness to adapt of the Lifenet participants. Lifenet would have the
advantage over capitalist-commodity production that it would not have to
achieve the demanding marginal equalities implied by the equalization of the
rate of profit as precisely as capitalist competition requires. The point of
Lifenet would not be to optimize, but to satisfice, that is, to do well enough
to get along and let people devote more of their life energies to things they
are interested in. If Lifenet could just manage to cope, it would succeed.
4.2
Lifenet boundaries
One idea of Lifenet or something like it is to create an alternative to capitalistcommodity social relations that could coexist and compete with capitalism.
The Lifenet could expand (or contract) in several dimensions in relation
to a coexisting capitalist-commodity system.
At an individual level, Lifenet would not require an all-or-nothing commitment. A person could devote some or all of her energies to Lifenet for
some or all of her life-cycle, just as participants in peer-production have
“day jobs”. This implies that Lifenet participants would typically have not
only Lifenet accounts, but also regular bank accounts, and would meet their
consumption/reproduction requirements not just from Lifenet distribution
centers, but also by spending money on commodities. This division would
depend on individual choices, but of necessity would imply an aggregate division of social economy between Lifenet production and capitalist-commodity
production. If Lifenet turned out to be extremely successful, it would starve
capitalist-commodity production both by withdrawing productive labor and
by meeting a larger proportion of human needs through the Lifenet distribution system.
17
There is inevitably another side to this dialectical coin, the “moral hazard” of individuals drawing much of their consumption from Lifenet without
making a proportionate contribution to Lifenet production, in the short run
subsidizing their private commodity expenditure, but also in the long run
effectively subsidizing the capitalist-commodity system. (Marxist analysis
would argue that in the long run the capitalist-commodity system would
reduce wages for labor-power by just the excess of their subsistence wage
workers would draw from the Lifenet.)
The coexistence of Lifenet with the capitalist-commodity system would
provide an important safety-valve both at the individual and social level. If
the Lifenet system experiences performance glitches (say, in the availability
of goods and services in the distribution system) they can be smoothed at the
individual level by shifting to the capitalist-commodity system. A smaller
capitalist-commodity system as a result of effective competition from Lifenet,
would have less capacity to play this shock-absorber role. One way Lifenet
distribution could adjust to shortages would be supplement its own production with commodities purchased from the capitalist system. This would be
self-defeating if it became a constantly reproduced pattern, but it could also
be one of the informational signals to Lifenet to adjust its internal allocation
of resources.
(A neoclassical economist might theorize that there is an “optimal” size
for the Lifenet system at which its marginal benefits in terms of high production quality and higher quality of life decline to equal its marginal costs
in terms of uncertainties of supply and inefficiencies of allocation. Along the
same lines, Lifenet might be regarded as a separate economy that traded
with the capitalist-commodity system.)
Another dimension in which Lifenet could expand or contract is the range
of goods and services it provided. Something like Lifenet does coexist with
capitalist-commodity production in the form of religious institutions, peerproduction, hobby-groups, and the like. Peer-production has demonstrated
its ability to compete effectively with capitalist-commodity production in
producing software (partly because the commodity form creates incentives
that interfere directly with critical aspects of software design and development). There is an open question as to whether it can be generalized to
drug development, particularly for the treatment of rarer diseases that offer
little profit incentive to capitalist pharmaceutical companies. How about local food production, child and senior care, therapy, clothing, textiles, heavy
metals, transportation infrastructure? Each of these cases poses different
18
obstacles to the success of the Lifenet model. (They also pose obstacles to
the success of the capitalist-commodity model in many cases.)
This coexistence model implies the need for Lifenet enterprises to exchange with the capitalist-commodity system. Lifenet could not possibly
supply all the input required for its own production, especially in its early
stages of development. Thus the Lifenet system will have to accumulate,
manage, and spend money on capitalistically produced commodities. From
the ideological point of view this margin looks particularly sensitive, since it
is where the pressures for Lifenet enterprises to shape themselves to the logic
of the commodity system will be greatest. Financial assets of the Lifenet
system would have to be managed as a kind of trust. If Lifenet could not accumulate enough monetary resources from transfers, it would have to export
something of value to the capitalist-commodity system, an activity that also
carries self-evident perils to the integrity of the Lifenet model.
4.3
Life-cycle and economic security issues
From the individual point of view, Lifenet participation presents significant
risks. Someone who devoted her whole productive energies to Lifenet might
be content enough with Lifenet social relations and Lifenet-supplied goods
and services, but what about risks of disability and retirement? Lifenet
production will inevitably involve investment, which will divert productive
resources from current consumption; what form would the corresponding
saving take, and how could individuals in an investing cohort claim the returns from investment in the future? The Lifenet accounts provide a range of
means of addressing these issues, at least in so far as they provided a credible
record of generational contributions to and uses of production. Even in the
realm of structured fantasy, it is clear that there is a risk of the Lifenet system degenerating into an alternative money-commodity accounting system
in facing these issues. If Lifenet were to prove robustly resistant to the development of internal distributional inequality, perhaps it could tolerate some
commodification around the edges without losing its essential character as
an alternative to the capitalist-commodity system.
The model of Lifenet co-existing with and competing with capitalistcommodity production raises another level of risk, the risk that one or the
other system might disappear entirely. Since both capitalist-commodity production and Lifenet production depend crucially on network economies to
be viable, it is all too likely that they will develop instabilities as one or the
19
other shrinks.
5
Questions by way of conclusion
Does the Lifenet fantasy actually have anything to teach us about the ideal
of socialism?
Marx often reiterates the importance of separating the necessary metabolism of human reproduction, the expenditure of labor to create useful
products and the consumption of those products to support human social
life, from the social relations through which reproduction takes place. This
turns out to be very difficult in practice because we tend to perceive production and reproduction through the lens of the specific forms and social
relations we actually experience. If Lifenet or any other version of socialism
is viewed through the laws of the commodity, it will increasingly appear to
be a distorted version of commodity production. This is a familiar effect of
bourgeois ideology. On the one hand, bourgeois ideology would predict the
rapid collapse of the Lifenet system (at least if it tried to cover the whole
gamut of human needs for reproduction) in a chronic and universal shortage
of use-values. On the other hand, bourgeois ideology hints that any alternative system of social relations, to the degree that it is successful in fostering
production and reproduction, would not be very different from the capitalist
commodity system. These boil down to different ways of asserting the identity of capitalist-commodity social relations with human social reproduction.
One advantage of the Lifenet fantasy is that it represents the transition
to socialism as a cumulative process of day-to-day choice. In this way it
emphasizes what seems to me to be the central question posed by the ideal
of socialism, the question of how we as human beings want to live our lives
together.
References
Bowles, S. (2004). Microeconomics: Behavior, Institutions, and Evolution.
Princeton University Press, Princeton, NJ.
Devine, P., editor (2002). Science and Society Special Issue: Building Socialism Theoretically: Alternatives to Capitalism and the Invisible Hand,
volume 66.
20
Duncan K. Foley (ed.) and Peter S. Albin (1998). Barriers and Bounds to
Rationality: Essays on Economic Complexity and Dynamics in Interactive
Systems. Princeton University Press, Princeton.
Foley, D. K. (2003). Unholy Trinity: Labor, Capital, and Land in the New
Economy. Routledge, London.
Foley, D. K. (2010). What’s wrong with the fundamental existence and welfare theorems? Journal of Economic Behavior and Organization, 75:115–
131.
Leontief, W. (1966). Input-Output Economics. Oxford University Press, New
York, NY.
Mirowski, P. (2002). Machine Dreams: Economics Becomes a Cyborg Science. Cambridge University Press, Cambridge, UK.
Schumpeter, J. A. (1942). Capitalism, Socialism, and Democracy. Harper
and Brothers, New York.
Vanek, J. (1970).
The General Theory of Labour-Managed Market
Economies. Cornell University Press, Ithaca.
Wright, E. O. (2010). Envisioning Real Utopias. Verso.
21
Socialist alternatives to capitalism I:
Marx to Hayek
by Duncan K. Foley∗
1
Utopian and scientific socialism
While philosophical speculation on the design of social, economic, and political institutions goes back at least to Plato (and perhaps further to Hammurabi), I will start this discussion at the end of the eighteenth century, in
the radical and romantic response to the French Revolution. The nascent Industrial Revolution, the eclipse if not the collapse of the religious monopoly
over discussion of the social and political “good”, the enormous energies
of mass politics unleashed by the bourgeois revolutions, and Enlightenment
theories of political economy and political theory prompted a surge of radical criticism of existing social institutions and speculation on their future
evolution.
An important and persistent strain in this revolutionary discourse viewed
the demonstrated (though not yet fully exploited) increases in productive
power unleashed by the Industrial Revolution as offering for the first time in
human history the possibility of creating a human society freed from material
poverty and even scarcity of basic goods. “Perfectibilists” such as William
Godwin (see Foley, 2006, ch 2) argued for a rational reconstruction of social
institutions to realize these possibilities, engendering an intellectual reaction
from figures such as Thomas Malthus, who proposed to demonstrate “mathematically” that population growth would doom such projects.
Leo Model Professor, Department of Economics, New School for Social Research, 6
East 16th Street, New York, NY 10003 ([email protected]) and External Professor,
Santa Fe Institute. This talk was prepared for lectures at the Havens Institute at the
University of Wisconsin, Madison, April 6–7, 2011.
∗
1
“Really existing” capitalism proved quite capable of achieving enormous
increases in social productive powers, but fickle in the distribution of the
resulting gains. Industrial entrepreneurs accumulated fortunes, a growing
“middle class” representing a significant minority of the population found
niches that provided a degree of security and comfort in the exploding division of labor, but by and large productive workers found themselves subject to
fierce competition for jobs that sharply limited their economic gains. These
disparities were dramatic enough in the few parts of the world experiencing industrialization to interest many, particularly of the middle classes, in
projects for a more rational and egalitarian organization of social production
and income distribution.
“Radical” movements demanding political reform of corrupt national systems, the universal extension of civil and political rights, free trade, republican forms of national political organization, and the like achieved major
gains in countries like France and England. On the fringe of this basically
bourgeois democratic radicalism appeared a flowering of theoretical writing
on the possibility of reorganizing the industrial division of labor on “socialist” principles of rational allocation and egalitarian distribution. In the same
period there were numerous small-scaled “experiments” in the form of local
communities seeking to organize industrial production in alternative ways.
The history of these experimental communities, some of which lasted quite a
long time, is an instructive lesson in the power of capitalist social organization to encapsulate and absorb economically deviant behavior. Neither the
theoretical nor the experimental side of this socialism achieved much headway compared to the awesome force of capitalist accumulation in fostering
proletarianization, urbanization, and globalization in line with its own logic
of the pursuit of profit.
Karl Marx’s political thought centered on the unfinished business of the
French Revolution, which had transformed the landscape of European politics, weakening the monopoly on power of ancien régime landed interests
and revealing the nascent strength of the industrial and financial bourgeoisie.
The democratic and egalitarian ideological impulses unleashed by the revolutionary moment pointed toward a more complete transformation of European
society; in class terms the growing proletariat of industrial workers, increasingly organized and united to secure economic gains, provided the political
base for such a transformational project.
Marx entered the European political scene in the eighteen-forties with a
well developed materialist theory of history, which saw class control of social
2
surplus production through exploitative social relations as the mainspring of
political and social change. His political work centered on the (somewhat
quixotic) goal of persuading the European proletariat to make change in the
class control of social surplus product the content of its program. Marx and
his allies had some success in this direction: European politics of the late
nineteenth and early twentieth centuries did become obsessed with the relation of the market and state and with mechanisms for sharing the wealth
created by industrial capitalism through redistribution. The politicians of
the European Left had less luck in transferring control of economic production from the capitalists to bureaucrats, despite some valiant attempts,
particularly in the ruins of a Europe shattered by the Second World War.
Marx’s revolutionary communism thus was grounded in a hard-headed
realistic analysis of European politics and society in his own time. The economically radical aspect of Marx’s project was the idea that the proletariat
would wrest control of the social surplus product from industrial and financial capitalists. Initially the proletariat would control the surplus product as
a class (presumably exploiting itself in the fashion outlined in the Critique of
the Gotha Programme), but with the longer-run goal of ending exploitation
altogether.
Marx branded his socialism “scientific”, because it grew out of an “objective” assessment of European political history in terms of class, and because
it was grounded in the widely accepted conclusions of classical political economy. There is some originality (or possibly incoherence) in Marx’s thinking
in this respect, since it is not obvious that social control over production is
necessarily the economic content of a working class political movement, nor
that there is anything in the logic of classical political economy that would
lead to the transcendence of private property as the keystone of economic
organization.
Marx seems to have had the “vision” that one end of this conundrum
would supply the solution to the other. As I have argued (Foley, 2006,
ch 3) Marx did see that the combination of Smithian increases in the usevalue productivity of labor due to the extension of the division of labor
and Malthusian pressures keeping wages close to subsistence would lead to
an unbounded rise in the rate of exploitation.1 This trajectory of capitalist
1
If x is the use-value productivity of labor, and w the use-value equivalent of the wage,
then the rate of exploitation is e = (x − w)/w Foley and Michl (see 1999); if w stagnates
at subsistence while x rises the rate of exploitation increases without limit.
3
accumulation does pose some purely economic problems, mainly the question
of where the aggregate demand to realize the potential product will come
from; but it raises the even more explosive political question of how the small
minority of capitalists can repress a working class which is so productive and
sharing so little in the fruits of its own productive powers.
The question of why a revolutionary European workers’ movement would
adopt a program of social control of production proved somewhat more challenging to Marx’s genius for theoretical speculation. I will address this question in section 2.
The historical evolution of capitalist accumulation resolved the problem
of the unbounded rise in the rate of exploitation in the frustratingly practical
compromise of an increase in the use-value equivalent of wages at roughly the
same rate as the increase in the use-value productivity of labor. This pattern,
with interruptions, was sustained from the mid-nineteenth century to the
nineteen-seventies; it asserted itself in a remarkably wide range of contexts
for industrial capitalist accumulation, from Europe to North America to Asia.
Since nineteen-eighty median real wages have ominously stopped rising at the
rate of increase of labor productivity in the advanced capitalist economies;
in fact, median real wages have become stagnant.
Marx realized that the general rise of wages posed a fundamental problem
for his revolutionary socialist program, since it could be argued (as Smith
had already realized and innumerable later apologists for capitalism have repeated) that capitalism was as good or better a vehicle for sharing increases
in the productivity of labor as socialism. This “trickle-down” ideology remains the bedrock of bourgeois political economy. Marx replaced the vision
of an unbounded increase in the rate of exploitation (too much surplus value)
with the more subtle and convoluted theory of the falling rate of profit (too
little surplus value relative to the value of capital), discovering the pattern of
capital accumulation Thomas Michl and I have called “Marx-biased technical
change”, which fits a wide range (but not all) capitalist economies undergoing
industrialization.
2
The Grundrisse and labor money
The problem of persuading the European working class movement to adopt a
program of socialization of production was a life-time challenge to Marx. He
faced a range of committed and credible working-class political leaders whose
4
economic programs fell far short of Marx’s vision of epochal change (the
end of mankind’s prehistory), which would sweep away not only exploitative
wage-labor relations, but commodity and money forms of organization of the
division of labor as well. The ideas of Marx’s political rivals aimed in one way
or another at retaining the commodity and money forms of the organization
of the division of labor while somehow eliminating exploitation.2
One influential proposal along these lines was the idea of replacing money
with labor certificates (as put forward by the “Ricardian socialists” Bray
and Gray). Marx devotes the massive Chapter on Money in the Grundrisse
(Marx, 1973, Chapter on Money) to a detailed analysis of this idea. An
uncritical reading of Ricardo’s summary of the theory of value lends some
plausibility to the labor money proposal. Ricardo argues that competition
tends to force “natural prices” (the long-term average prices around which
short-term observed “market prices” fluctuate) to be proportional to the
labor time required for the production of commodities. Thus commodity exchange boils down on average to an exchange of labor time for labor time (in
the form of particular commodities). Ricardo regards money as a “veil” that
only serves as a medium of exchange obscuring this underlying metabolism
of the division of labor. So it is tempting to argue for short-circuiting this
process by a system that gives workers claims on the total product (labor certificates) equal to the labor time they expend. On average this should work
because the amount of labor certificates issued will just match the labor expended and the value created in the mass of commodities produced. But
by eliminating the mediation of the capitalist employer, the system would
be free of exploitation: workers would receive a full equivalent in labor certificates and claims on the social product for the labor time they actually
expended.
Marx’s evolving theory of value (an elaboration of the theories of Smith
and Ricardo) was particularly directed at ferreting out fallacies in arguments
of this type. In the discussion in the Grundrisse Marx centers his critique of
the labor-certificate scheme on the observation that under capitalist relations
of production private labor becomes social labor only through the exchange
of products as commodities. The fallacy of the labor certificate scheme lies
in its implicit assertion that private labor could be “immediately” social.
When we unpack this language, we see that Marx is raising a series of
These proposals hoped to make w equal to x, thus driving the rate of exploitation to
zero.
2
5
fundamental questions that are still fresh. One point is that commodity
exchange equalizes labor times of different workers only on average over repeated cycles of production. In any particular cycle of production market
prices and wages may diverge substantially from the natural levels that competition enforces. A second point is that the social character of privately
expended labor has to be proved through the realization of the value of the
produced commodity through its exchange for money. A third point is that
the capitalist extracts labor from labor-power: the capitalist pays the money
wage for labor-power, which may be wasted or ineffectual in the production process. This last observation underlines the fact that the wage-labor
contract, though it results in workers appropriating a “share” of the value
produced in the form of wages, is not a contract to share the value produced.
The capitalist buying labor-power pays the wage whether or not any value
is produced at all.
This line of thinking has important ramifications for the understanding of
money as what Marx later called the “socially accepted general equivalent”:
labor in actual production is always concrete and private and can express
itself as abstract and social labor only through the one-dimensional purely
quantitative medium of money. It is true enough that when the dust settles, that is, ex post, money prices of commodities average out to consistent
quantities of social labor expended, but one cannot expect to achieve this ex
post rationality ex ante through the labor certificate scheme.
The conclusion Marx draws from this analysis is that the labor-certificate
issuing “bank” would inescapably have to take charge of organizing the production process, not just at the level of individual production sites, but for
the economy as a whole.
Curiously enough, Marx more than once arrives at the end of a critique
of an argument in the posture of endorsing the argument. In this case Marx
starts by critiquing the labor-certificate scheme, but winds up concluding that
it would imply the kind of “rational” and “social” (and, it is hard to avoid
inferring, “centralized”) organization of production he himself advocates.
3
Marx as a long-period theorist
Marx spent a lot of time reading and commenting on the classical political
economists. Much of his and later readers’ attention has centered on his
critique of the treatment of surplus value and exploitation in the classical
6
political economists. As a consequence, how much Marx accepted the methods and conclusions of the classical political economists in some important
areas has not been fully appreciated. One of these areas is Marx’s adoption
of the “long-period” method of the classical political economists, an issue
which has implications for the theory of socialism.
The classical political economists’ paradigmatic question was to understand abstractly how a decentralized, competitive market process based on
free economic decisions by producers could organize production with a highly
developed division of labor. Their analysis of this question is remarkably subtle and insightful, and is the basis of Marx’s approach to the same problems.
It is important to realize in discussing this method that it is an abstract thought-experiment, not an attempt to explain the actual workings
of real-world economies directly. In reality, for example, producers are not
completely free to move between different lines of production, and markets
are not completely open to all producers. I have commented on this theory
in Foley (2006, ch 1, 2, and 3), and on Marx’s adoption of it in Foley (2011).
Marx’s most complete discussion of these issues appears in the first two Parts
of Volume III of Capital (Marx, 1981, Parts I and II).
The long-period method addresses both the problem of the allocation of
social labor time among various productive activities and the dynamics of
price determination on markets. Market prices of reproducible commodities
fluctuate both in time and space, but the long-period method argues that
they are regulated by the free movement of producers among lines of production to “gravitate” around “natural prices” (which Marx calls “prices of
production”). The basic idea is that if market prices of one type of commodity remain high enough on average for a long time to give producers of
that commodity some net perceived advantage over producers of other commodities, that producers will shift into that line of production, which will
tend to reduce average market prices there. Thus the natural prices around
which market prices gravitate must tend to equalize the perceived advantages to producers in each line of production. One simplified version of this
process assumes that the only important respect in which the advantages of
producing commodities differ is the labor time required for their production.
Under this assumption the natural prices of commodities will tend to be proportional to labor times. The mechanism that enforces these tendencies is a
constantly changing social allocation of productive labor to match production to social demand. The long-period method is simultaneously a theory
of commodity prices and a theory of the allocation of social labor. (When
7
means of production have been appropriated as capital, the movement of
producers among lines of production will be regulated by capitalists seeking
the highest rate of profit so that natural prices will tend to equalize profit
rates. In this situation the free mobility of labor continues to enforce the
equalization of the rate of surplus value or exploitation in various lines of
production (Cogliano, 2010).)
There are many notable implications of the long-period analysis of commodity production. From the point of view of the problem of socialism the
most salient is that the long-period method envisions commodity production
as a chaotic, decentralized process of social allocation of productive resources
which never reaches a stable equilibrium. In the language of modern complex
systems theory the long-period method of the classical political economists regards the production of commodities as an adaptive, self-organizing “bottomup” system. Marx understood this view very thoroughly, and in fact used it
to critique commodity-producing capitalism on the grounds of its “anarchic”
character. We can see the continuity of Marx’s views between his critique of
the labor-certificate scheme and the notes that constitute the first two Parts
of Volume III of Capital.
Implicitly and on occasion explicitly, Marx viewed socialism as an alternative to commodity-producing capitalism not just in the dimension of
exploitation through the wage-labor contract, but also in the more fundamental dimensions of commodity exchange and money. Marx’s continuing
critiques of the commodity form, ranging from his early essays on alienation
to the last part of Chapter I of Volume I of Capital on “commodity fetishism”
support this view.
Did Marx have a vision of socialism or communism as methods of organizing production? If so, was it based on what we would now call a top-down
or a bottom-up dynamic?
4
Marx’s socialism
The most complete discussion of the problems of organizing production in a
socialist society we have from Marx is the Critique of the Gotha Programme
(Marx, 1970), a commentary Marx wrote in 1875 on a draft program for a revived unified workers’ movement in the form of a German Social Democratic
Party (see Foley, 2006, ch 3 for more discussion).
In general Marx responded to questions about his views on the organiza8
tion of socialist society by saying that these questions could be solved only
by those who actually confronted them historically, that is, presumably, by
the revolutionary movements that would establish socialist institutions. This
is a prudent and in some ways sensible stance, but frustrating to a posterity
that has made notably little progress on these problems.
In the Critique of the Gotha Programme Marx envisions two “stages” to
the construction of communism. In “the first phase of communist society
as it is when it has just emerged after prolonged birth pangs from capitalist
society” Marx describes a situation strikingly like his analysis of capitalist
society: workers receive only a part of the value they produce (the rest being
reserved for social purposes like investment, education, and administration
that are financed out of surplus value in capitalist society). What seems
to be the crucial difference is that in capitalist society the surplus product
is appropriated by the capitalist (and landowning) classes in the form of
surplus value, while in the first stage of socialist society the surplus product
is appropriated through some political mechanism, perhaps the “dictatorship
of the proletariat”. Marx does not explicitly discuss the social organization of
production, but we might infer that workers go to jobs in enterprises, get paid
in the form of money, and buy products with these money revenues. This
is not dramatically different from the way capitalism works. It is interesting
that Marx presents himself in these comments as representing a hard-headed
realistic “economistic” position which acknowledges resource constraints and
the necessity of making decisions subject to social trade-offs.
It seems to me significant that Marx does not seem to raise the issue of
the mechanics of control over social production, allocation, or distribution in
these passages, despite his evident understanding that commodity production
regulates both the allocation of social labor and prices. There is in particular
no sign that Marx thought about “market socialist” institutions in which
some elements of commodity production, such as competition for sales, might
co-exist with workers’ direct or indirect control of enterprises.
This lacuna in the discussion of first stage of transition from capitalism
to communism is underlined by the language in the Critique of the Gotha
Programme Marx produces to discuss communism:
In a higher phase of communist society, after the enslaving
subordination of the individual to the division of labor, and therewith also the antithesis between mental and physical labor, has
vanished; after labor has become not only a means of life but life’s
9
prime want; after the productive forces have also increased with
the all-around development of the individual, and all the springs
of co-operative wealth flow more abundantly – only then can the
narrow horizon of bourgeois right be crossed in its entirety and
society inscribe on its banners: From each according to his ability,
to each according to his needs!
What is particularly interesting about this famous passage is that it focuses so much on the problems of social psychology of labor and the principles
of distribution without even a slight reference to mechanisms of social organization of production or the regulation of a division of labor. The invocation
of higher productivity as a context in which the problems of productive organization might be easier to resolve is particularly troubling: the springs
of cooperative wealth flow much more abundantly today than in 1875, without much noticeable movement to non-capitalist systems of allocation and
distribution.
The hints we can glean from Marx’s enthusiasm for the Paris Commune
(which ran a siege economy along politico-military lines) and for joint stock
corporations as precursors of the socialization of means of production do not
go very far to clear up the basic puzzle. Marx understood as well as anyone of
his generation the bottom-up logic of commodity production as a system of
allocation of productive resources and distribution, but does not even sketch
a parallel socialist solution to these problems.
Any thoughts that Marx might have had some version of decentralized
spontaneous organization of production in mind as a basis for socialism run
into his unambiguous negative comments on commodity production and
money as backward “anarchic”, “irrational”, or “contradictory”. Not for
Marx any view that the market represents a higher or transcendent form of
social interaction that contains at least the seeds of a democratic, decentralized social method of information aggregation.
5
Collectivist analysis of socialism
In the years after Marx’s death the Italian economists Wilfredo Pareto and
Enrico Barone took up the problem of the social organization of production.
Pareto and Barone wrote from a technocratic perspective, apparently in the
hope of decoupling the debate over socialism from the class struggles that
10
were threatening to tear European society apart in the first years of the twentieth century. Barone frames the issue of social control over production as a
problem of allocation and distribution, and concludes that the goal of a socialist economy in these dimensions ought to be to mimic the allocation of a
purely competitive capitalist economy. The basic idea here is to find a set of
prices, including wages, interest rates, and rents, at which individual enterprise marginal costs will faithfully represent real social marginal costs. Costminimizing production at such prices is socially efficient (Pareto-optimal). In
principle goods ought to be allocated to households or individuals to equalize these social prices to marginal rates of substitution (ratios of marginal
utilities). The simplest way to accomplish this is to give each household
or individual an income budget and let them spend it as they prefer. The
distribution of these incomes over final household consumers is a political
question, to be settled outside the allocation process.
In this setting, as Paul Samuelson later observed, it doesn’t really matter
whether capital hires labor or labor hires capital, or both are allocated by a
central planner: the important goal is the allocation of productive resources
and final products consistently with the principle of equating marginal cost
and marginal benefit. Marx’s social relations of production don’t seem to
matter from this point of view: except for distribution of income, socialism
and competitive capitalism boil down to the same principles of allocation.
The decisive rhetorical maneuver in Barone’s work is the assertion that economic allocation of resources is a technical rather than political question;
or equivalently, that the management of economic policy ought to be left to
experts except possibly for the distribution of income. This conclusion has
been particularly congenial to the mindset of engineers who tend to view
economics as a problem of optimal design and control, rather like a complex
industrial production process.
Barone’s and Pareto’s thought had only a marginal influence on practical
political economy, but an enormous impact on the development of economic
theory.
6
Really existing socialism
History, of course, was not waiting for political economy to sort out the issue
of socialism. The early decades of the twentieth century also saw the emergence of a Soviet Union dedicated, rhetorically at least, to the building of
11
a socialist economy, out of the wreckage of the Russian Empire.3 The Bolsheviks confronted two practical political economic problems in the 1920s.
Perhaps the lesser was the organization of day-to-day production and distribution. The greater was what later came to be called the problem of
“development”: how to transform the backward, largely traditional agricultural economy of the sprawling Russian empire into a viable industrialized
power. The second task was all the more pressing because the dénouement
of the First World War on the Eastern Front had revealed the military and
political vulnerability of a backward Russian economy in a world increasingly
populated by industrialized capitalist powers.
The Soviets, whose leadership contained some remarkably talented and
thoughtful figures, such as Nikolai Bukharin, arrived at a practical solution
to the problem of organization of day-to-day production fairly soon after
consolidating political power through prevailing in the Civil War with the
Whites. This took the form of the “New Economic Policy” (NEP), which
allowed, and indeed encouraged, capitalist commodity production in many
sectors of the economy, including food production. The NEP structurally
created a “mixed” economy in which a large private sector co-existed with
state-sponsored industrial enterprises. (The NEP is recognizably a precursor
of Deng Xiaoping’s “socialist market economy” which was the framework
for Chinese economic growth in recent years.) The NEP worked well to
stabilize the Soviet economy after the Civil War; under this regime output
and incomes recovered significantly from the low levels of the Civil War
period. Many Bolsheviks, including Lenin, were willing to accept the NEP
framework as an indefinitely prolonged phase of building socialism, as long
as the state held the “commanding heights” of the economy through control
of energy, transportation, heavy industry, and finance.
Two connected and predictable developments undermined the NEP. First,
as we would expect, the privatized commodity-producing sector of the economy exemplified some fundamental laws in generating large disparities in
income and wealth. The appearance of a wealthy proto-bourgeoisie in the
midst of a society governed by a one-party dictatorship committed to socialist goals threatened the narrow political base of Bolshevik power. Second,
the NEP was structurally favorable to the agricultural sector, particularly to
3
The bitter and still acrimonious political controversies over the Soviet experiment
make it very difficult to learn just what happened. The reader interested in exploring this
subject might start with Carr (1985); Nove (1989); Ellman (1989).
12
the market-oriented strata of the peasants. The NEP thus resulted in a relatively “balanced” path of economic growth between industry and agriculture.
Not surprisingly the prices of food and other agricultural products tended
to rise relative to industrially produced goods, limiting the degree to which
surplus production in agriculture could be mobilized for rapid industrial investment. Despite Bukharin’s determination to “ride to socialism behind the
peasant’s nag”, NEP policies proved explosively unstable politically, and collapsed under efforts to mobilize agricultural surpluses by military force which
ultimately led to collectivization, central planning, and the bloody political
purges of the nineteen-thirties. Stalin rather than Bukharin balanced at the
top of the “greasy pole” of Communist Party politics apparently more by
ruthless and opportunistic manipulation of coalition politics than through
consolidating a stable political base of power.
Just how the Soviet economy did operate in the 1930s, and in particular what was the practical mixture of bottom-up and top-down dynamics in
this period remains obscure. A relatively small central planning bureaucracy
had in theory enormous economic power to implement a policy of extremely
rapid industrialization centered on the expansion of heavy industry. It seems
unlikely that the fairly coarse-grained plans produced could be the whole
story of Soviet economic life under this regime. The management of new
industrial armies recruited to new urban centers involved a great deal of decentralized local improvisation and experimentation. Since the centralized
planning bureaucracy did not actually control many resources itself, local
managers facing shortages of inputs and surpluses of outputs improvised informal market-like arrangements with each other. Despite the wishful thinking of market-fundamentalists, capitalist economies in periods of extremely
rapid industrialization have often strayed significantly from the orthodoxies
of profit-maximization given market prices, making it difficult to draw hard
analytical lines by which to characterize the Soviet experiment.
7
The socialist calculation debate
While the Soviets were engaged in their furious (if in the very long run
curiously futile) attempt to change the world, western European philosophers
continued to interpret it. The Bolshevik impulse spilled over into active
revolutionary movements in Hungary and other parts of Europe. The bright
confidence of bourgeois ideology was shaken by the debacle of the First World
13
War, and the post-War political regimes in European countries were fragile
and weak-willed, and thus ready to compromise with elements of Bolshevism
and statism.
The “Austrian” school of neoclassical economics, founded by Carl Menger,
and bolstered by Eugen von Böhm-Bawerk’s critique of Marx’s theory of
value, founded itself on a belief in the existence of universal economic laws.
(From a Marxist point of view these “universal” laws look very much like the
laws of commodity production, which underly the logic of market exchange.)
The flirtation of the European political elite with the idea of fundamental
changes in economic organization along socialist lines disturbed the Austrians
deeply. Ludwig von Mises was moved to argue (von Mises, 1990) that centrally planned socialism was an impossibility. Mises’ case was that economic
rationality required determining market-clearing prices for the millions of
specific goods and services offered. This task, he argued, was beyond the capacity of any central planning mechanism on purely computational grounds.
(An irony of history is that at the same time Mises was formulating this argument, Alan Turing and others were laying the theoretical groundwork for
the creation of electronic computers which could practically tackle problems
of this magnitude.)
Oskar Lange and Abba Lerner (Lange, 1948; Lerner, 1970) cleverly (perhaps overly cleverly) proposed to turn this argument on its head: why
couldn’t a socialist society (along the lines of Barone’s analysis) mimic a
capitalist commodity society by instructing the socialist managers of enterprises to compete on markets like capitalism firms? In this way the sacred
equalities of efficient allocation (between marginal costs and prices) could be
discovered by a socialist economy.
This was an ingenious but, as it turned out, quite perilous move. By
endorsing the idea that the comparison of socialism and capitalism could
legitimately be reduced to a comparison of allocation of resources and distribution, this argument for the feasibility of market socialism undermined
the argument for any type of socialism. This phase of the socialist calculation debate played a powerful role in shaping general equilibrium theory
by legitimating the view that the theory of resource allocation was a neutral
territory for political economy, an arena in which bourgeois and socialist scientists could coexist and carry out Popperian dialogue under a Durkheimian
blessing. This turn of events explains the peculiar structure of economic
general equilibrium theory, which abandons the substance of the neoclassical
vision in a pursuit of empty mathematical generality (see Mandler, 1999, for
14
a thoughtful exploration of these themes).
The version of economic dogma, enshrined in the “First and Second Fundamental Welfare Theorems” and “Fundamental Existence Theorem” around
which mainstream economics built its texts, curriculum, and worldview in
the last half of the twentieth century is technically flawed (see Foley, 2010, for
a detailed discussion) in ignoring the dependence of final market allocations
on the path of transactions.
The First Welfare Theorem is usually phrased in laissez-faire language
to assert that a competitive market equilibrium in the absence of externalities is a Pareto-optimum; a less-mystified statement would be that private
commodity owners free to make arbitrary voluntary exchanges, will continue
trading to some allocation close to or in the private exchange equilibrium set
in which their private reservation prices for the commodities are all equal.
(The reasoning behind this conclusion is simply that if private reservation
prices at any allocation are not equal, then transactions the agents view as
mutually advantageous are possible, and will be carried out.)4
The Second Welfare Theorem is usually presented as arguing that when
preferences and production sets exhibit diminishing marginal rates of substitution and transformation (in mathematical language when utility and
production functions are concave) any Pareto-optimal allocation can be implemented as a competitive market equilibrium with an appropriate redistribution. In this form the Second Fundamental Welfare Theorem has a political
corollary: since capitalist commodity production can in principle achieve the
same range of outcomes as any socialist alternative, the advocates of socialism might as well pursue the goal of redistribution under capitalist social
relations. In fact, from a certain point of view socialism and the redistribution of resources under capitalism amount to the same thing. The flaw
in the Second Welfare Theorem lies in its ignoring the path-dependence of
competitive equilibrium allocations on the transactions that implement it.
In general if some omnipotent authority were to reallocate ownership of resources among the members of a society in accordance with some principle of
distributional equity, the process of market trade to the exchange equilibrium
set would introduce additional horizontal inequality into the final distribution of produced goods and services. (The only way this market-generated
The private exchange equilibrium set is the Pareto-optimal locus in the absence of
externalities, because without externalities private reservation prices for commodities coincide with social marginal benefits and costs.
4
15
inequality could be avoided would be for the omnipotent authority to enforce
the final allocation of goods and services directly – shades of Marx’s critique
of the labor-certificate bank!)
The larger fallacy in the whole socialist calculation debate, however, lies
in its complacent acceptance of the idea that the only thing that matters in a
society is the production and distribution of material goods and services. In
Marx’s language this is roughly the position that it is possible to understand
the development of “forces of production” independently of “social relations
of production”. At this very abstract level, reducing the comparison of capitalism and socialism to points in an Edgeworth-Bowley box is just another
way of asserting the existence of “universal” economic “laws” independent of
the form of social organization of production. Marx hammers away at this
fallacy repeatedly: means of production become “capital” only when they
are appropriated as private property; the natural productive powers of the
natural environment become “land” only when they are appropriated; “labor” is quite a different category depending on whether we are talking about
slave labor, “free” wage labor, or the “labor of freely associated producers”.
The real import of the historical social choice between socialism and capitalism is precisely what is left out of the socialist calculation debate: the social
relations through which people organize themselves to produce.
8
Hayek changes the ground
For Friedrich Hayek Lange’s riposte to Mises’ claim of the economic impossibility of socialism was an Archimidean epiphany (summarized in von Hayek,
1948, particularly Essays IV, VII, VIII, and IX). It would be impossible
for the managers of socialist enterprises, Hayek argues, to mimic capitalist
competition because the socialist manager would not be putting his or her
own personal interests in play in making production and investment decisions. (Insofar as the socialist State tried to remedy this problem by creating
“incentive compatible” systems in which managers’ personal interests were
aligned with profit-maximization, the socialist State would be reproducing
capitalist social relations. Shades of the NEP and the Deng Xiaoping!)
The implications of Hayek’s critique, however, go far beyond the tired
dialectics of market vs state. In Hayek’s more radical view, commodity production expresses the existential content of human social existence. Each of
us inevitably understands a different aspect of the world: as a result there
16
can be no centralized representation of human knowledge in the large sense.
Only through somehow reassembling the bits of knowledge scattered through
the whole population can social production take place. But, in Hayek’s way
of thinking, people will surrender their individual knowledge to society only
as part of a deal that touches their self-interest. Marx’s story of social production without commodities and money is doomed to remain a fantasy,
because properly understood social production is commodity production. It
is only through commodity exchange that the information flows required to
sustain social production can be elicited and recombined.
Hayek’s insight is bound to be compelling to people who live with and
by computers, the internet, and the stock market. It does seem increasingly
true that economic life is an exchange of information, not the metabolism of
material production and distribution. (On the other hand these same people
do continue to eat food, live in houses, wear clothes, and drive cars. The
nutritional value of a megabyte of computer code is hard to detect.) Hayek
represents an important turning point in our conceptualization of social life,
leading to regarding human societies as complex, adaptive systems.
The blind spot in Hayek’s thinking is the complete neglect of distributional issues. This is no accident: in fact, it is a principled and consistent
implication of his social theory, in which process is privileged over outcomes.
The important thing about markets (or in the broader sense commodity exchange) for Hayek is the process of revelation of information. This process
may have as a by-product the production and consumption of material goods
and services, but that is not its main purpose. Humanity is defined by information revelation as much as by language or sociality, both of which can, of
course, be viewed from an information perspective.
In some ways Hayek’s intervention returns us to the perspective of the
classical political economists. Despite Marx’s attempt to hijack the theoretical power of classical political economy as a vehicle for revolutionary socialist
transformation, Smith and Ricardo express no discomfort with commodity
relations, though they do see commodity production as a complex system.
As far as one can tell from their writing, Smith and Ricardo see humanity as
finding its fate in the framework of commodity production, not outside of it.
But the idea of socialism is as closely intertwined with capitalist commodity production as bourgeois ideology, and is not much more likely to be
conjured away by dialectics alone.
17
References
Carr, E. H. (1985). Bolshevik Revolution, 1917-1923. W.W. Norton, New
York.
Cogliano, J. (2010). Smiths perfect liberty and Marxs equalized rate of
surplus-value. Submitted.
Ellman, M. (1989). Socialist planning. Cambridge University Press, Cambridge [England] ; New York.
Foley, D. K. (2006). Adam’s Fallacy: A Guide to Economic Theology. Harvard University Press, Cambridge, MA.
Foley, D. K. (2010). What’s wrong with the fundamental existence and welfare theorems? Journal of Economic Behavior and Organization, 75:115–
131.
Foley, D. K. (2011). The long-period method and Marx’s theory of value. In
Caspari, V., editor, The Evolution of Economic Theory: Essays in Honour
of Bertram Schefold, pages 15–38. Routledge, Abington and New York.
Foley, D. K. and Michl, T. R. (1999). Growth and Distribution. Harvard
University Press, Cambridge, MA.
Lange, O. (1948). On the economic theory of socialism. The University of
Minnesota Press, Minneapolis.
Lerner, A. P. (1970). Economics of control; principles of welfare economics.
A. M. Kelley, New York.
Mandler, M. M. (1999). Dilemmas in Economic Theory: Persisting Foundational Problems in Microeconomics. Oxford University Press.
Marx, K. (1970). Critique of the gotha programme. In Marx/Engels Selected
Works, volume 3, pages 13–30. Progress Publishers, Moscow. 1875.
Marx, K. (1973). Grundrisse: Foundations of the critique of political economy
(Rough Draft). Penguin, Harmondsworth.
Marx, K. (1981). Capital, volume 3. Penguin Books, London and New York.
[1894].
18
Nove, A. (1989). Economic history of the U.S.S.R. Penguin Books, Harmondsworth, Middlesex, England ; New York, N.Y., U.S.A.
von Hayek, F. (1948). Individualism and Economic Order. University of
Chicago Press, Chicago.
von Mises, L. (1990). Economic calculation in the socialist commonwealth.
Ludwig Von Mises Institute, Auburn University, Auburn, AL. translated
from the German by S. Adler.
19
Dilemmas of Economic Growth
by Duncan K. Foley
∗
Abstract
Geo-scientists and economists carry on discussions of central problems of economic growth, including global warming and resource exhaustion on the basis of sharply diverging assumptions about limits to
economic growth. These divergent assumptions can be traced at least
to the debate over N. Georgescu-Roegen’s work on entropy and economic production. Limited evidence indicates that economic growth
sufficient to raise world average productivity to current European standards is at the edge of feasibility in terms of renewable energy. Rosy
expectations that information technology can produce a “new” economy not subject to the resource and social limitations of traditional
capitalism confuse new methods of appropriation of surplus value as
rent with new methods of producing value. While the world economy will almost certainly transform from the small population, low
labor productivity, rich fossil-fuel based resource base of the industrial
revolution to a large population, high productivity, resource limited
pattern, the social relations accompanying this change are far from
certain or determined.
1 Controversial reviews for economic
growth theory
In the Fall of 2008 I gave a talk to a workshop made up of a group
of well-informed and concerned environmentalists, agronomists, geoDepartment of Economics, New School for Social Research, 6 East 16th, New York,
NY 10003, [email protected], External Professor, Santa Fe Institute. This is my
talk as outgoing President of the Eastern Economics Association, delivered at the EEA
meetings in Boston, Massachusetts, March 9, 2012. I am deeply indebted to Armon Rezai,
Lance Taylor, and Eric Smith for extensive conversations on the themes of the talk. I’d
like to Sanjay Reddy for calling Žižek (2012) to my attention.
∗
1
physicists, and other scientists, philosophers, and writers. The talk
summarized Rezai et al. (2011), and centered on the slide reproduced
as Figure 1.
In the overall context of the paper and talk, this slide was intended to convey two important points about the economics of global
warming.
First, following Foley (2009), the simulations reported in the slide
exemplify the theoretical point that from a welfare economics point of
view global warming is an externality, the correction of which (whether
through direct controls, carbon taxes, or a cap-and-trade system) allows a Pareto-improvement across generations. This point of view is
in sharp contrast to the presentation in many economics discussions
(for example Nordhaus, 2008; Nordhaus and Boyer, 2000) that frame
the global warming policy problem in terms of a tradeoff between the
interests of current and future generations. I thought this perspective
might be welcome to the workshop audience in substituting “gains
in efficiency and welfare” for “sacrifices of economic growth and standard of living” in the political debate over measures to mitigate global
warming.
Second, the slide simulations, based on plausible but far-fromwatertight assumptions about the costs and benefits of greenhouse gas
mitigation, and standard rational-expectations equilibrium assumptions, present a scenario in which relatively modest diversion of world
output (on the order of 1%) to investment in reducing greenhouse gas
emissions yield enormous benefits, measured in terms of long-term
per-capita world income. On an optimal path in which the global
warming externality is internalized world per capital income continues a steady rise from current levels of about $6,000 per year to a
long-run steady-state level of $20,000 per year, while on a “businessas-usual” equilibrium path where there is no effective price for greenhouse gas emissions a climate catastrophe about 100 years from the
present induces a sharp fall in world per capita income, which falls to a
steady-state level below current levels. The optimal mitigation investment succeeds in halting and reversing greenhouse gas accumulations
in the atmosphere and returning them to industrial levels, whereas
the business as usual path allows accumulation of greenhouse gases to
rise until climate damage itself stabilizes economic output.
Somewhat to my surprise, the workshop audience did not seem
very interested either in the welfare economics policy message of the
simulations, nor in their relatively optimistic scenario of greenhouse
2
Cons per capita
World Product Y
$000
$trillion!yr
20
250
10
100
5
50
2
10
20
30
40
50
decades
Carbon price
10
30
40
50
decades
Mitigation!Y
! Y spent
$!t
10 000
20
OPT
50
BAU
5000
COPT
10
1000
1
200
10
20
30
40
50
decades
CO2 Accumulation
10
20
! pot Y lost
ppmv
30
40
50
decades
Damage
50
700
10
600
1
500
0.1
400
0.01
300
10
20
30
40
50
decades
World Capital K
10
!!yr
$trillions
1000
3
500
2
250
1
100
"1
50
20
30
40
50
40
50
decades
Interest Rate
10
20
30
decades
"2
10
20
30
40
50
decades
"3
Figure 1: The optimal equilibrium path, OPT, is
plotted in green, the BAU equilibrium in red, and
the constrained equilibrium, COPT, in blue. The
carbon price on the optimal path is about $200/t,
and the damage on the optimal path is less than 1%
of potential output. The carbon price is the social
marginal value of foregoing the emissions from 1t
of carbon. For OPT and COPT, the carbon price
is effective in economic decisions, but not in BAU
(and is plotted as a dashed line). For OPT, the
mitigation percentage is the proportion of world
product devoted to mitigation. For COPT and
BAU, the mitigation percentage is the investment
called for by the imputed carbon price (and is plotted as a dashed line), while the actual mitigation
is zero.
3
gas mitigation. The strong consensus of the workshop audience was
that the projection on the optimal path of a steady-state world percapita income level of $20,000 per year with an assumed steady-state
population of 8.5 billion was grossly, indeed, to this audience, ludicrously unrealistic. In their view, even if the consequences of global
warming could be mitigated effectively, a rise in world product from
current levels of around $60 trillion a year to the $250 trillion a year envisioned in the simulation would put sufficiently catastrophic stresses
on world resources and ecologies as to guarantee a complete collapse
of human civilized life.
I was somewhat astonished myself to learn that a broad consensus
in that workshop thought a more “realistic” scenario for the world
economy would involve a reduction of world population during the
next century from its current level of about 6.5 billion to 1 billion,
a reduction the workshop consensus considered inevitable, but hoped
could be accomplished without violence or the compromise of human
rights and dignity. This prospect appeared to me just as unlikely as the
output levels in our simulation appeared to the rest of the workshop.
My co-authors and I had regarded our specification of output and
productivity growth as reaching an asymptote as rather conservative
compared to many economic growth models that unapologetically imply much higher levels of world production than our scenario.
This experience suggested to me that there is an urgent need for
dialogue between economists and other disciplines to narrow the gap
between the futures the two groups envision as feasible and desirable.
2 Nicolas Georgescu-Roegen and entropy
Political economy and economics have grappled with the problems of
limits to economic growth in land and resource scarcity from their earliest incarnations. The immediate background to our current preoccupation with the dilemmas of economic growth, however, is the debate
in the nineteen-sixties and seventies prompted by Nicolas GeorgescuRoegen’s work connecting thermodynamics with economic production
theory (Georgescu-Roegen, 1986, 1999). (An important collection of
papers on this subject was published by the Eastern Economics Journal in 1986.)
Georgescu-Roegen, who made important early contributions to the
4
economic theory of production, raised the question of how economic
production interacted with physical theories, particularly thermodynamics. Production functions and sets typically deal with economically defined categories, such as labor, land, produced inputs, and
the like. Real-world production, however, involves the transformation
of matter, such as mineral, coal, petroleum, and natural gas deposits
and the like. Production functions generally abstract from material
inputs of this kind. Georgescu-Roegen called attention to the rather
subtle point that the expenditure of energy in productive activities
(through the generation of electric or steam power, or the refinement
of mineral ores, for example) does not actually destroy any energy or
matter, because of the law of conservation of matter and energy. From
a thermodynamic point of view the key insight is that in the process
of organizing energy and matter into products useful to human beings, economic production necessarily raises overall entropy and thus
reduces the “available” or “useful” energy and matter available in a
closed system like a planet. In raising these issues Georgescu-Roegen
was contributing to a widespread interest among physical scientists
in the middle decades of the twentieth century in the complex relations between physical and chemical laws and the phenomena of life,
and particularly human life in society. One of the insights of this discussion was that life processes based on various metabolisms locally
violate the Second Law of Thermodynamics in the process of maintaining highly organized and therefore low entropy systems like living
organisms. The laws of thermodynamics require that this local reduction in entropy be more than compensated by a rise in environmental
entropy through the organism’s generation of heat and waste products. As far as I know, no one has disputed the theoretical correctness
of Georgescu-Roegen’s observation that economic production, as an
aspect of human social and individual reproduction, must also raise
the entropy of its physical environment, in this case, the planet we
live on. Georgescu-Roegen drew the conclusion that ultimately this
entropic limit would assert itself in the form of scarce resources and
energy.1
Clearly then, the entropy law is the root of economic scarcity
The entropy relevant to Georgescu-Roegen’s argument is thermodynamic entropy. In
other contexts I have discussed the informational entropy associated with economic allocations, but I’d like to emphasize that physical thermodynamic entropy and economic
informational entropy, though they share the same terminology, refer to quite different
concepts.
1
5
in a much stronger sense than simple finiteness. To wit, Ricardian land (i.e., terrestrial space) is finite but only at a
given moment; over time it is not, since we can use it over
and over again without diminishing its amount. By contrast, since low entropy of energy or matter can be used
only once, the scarcity of these elements steadily increases.
This is the most important object lesson of thermodynamics for the modern economist.
While I thus insisted ... that the economic process is entropic in all its material fibers, I hastened to add that it
cannot be reduced to the degradation of low entropy; that
would be to look at it as a physicist. However, the true
“output” of the economic process is not a material flow of
waste, but a unique flux, the flux of the enjoyment of life.
Without including this flux as well as many specifically human propensities into our analytical armamentarium we are
not in the economic world.... (Georgescu-Roegen, 1986)
Georgescu-Roegen’s intervention coincided with the publication of
other widely read books, particularly the “Club of Rome” study, Limits to Growth and Paul Ehrlich’s The Population Bomb (Meadows
et al., 1976; Ehrlich, 1971) that sounded alarms concerning the sustainability of world patterns of economic growth. These were also
years of cold war propaganda battles concerning the relative effectiveness of “free enterprise” and “socialist” systems of economic production in fostering growth of output. One latent theme in the promotion of growth economics was the hope that economic growth, by
providing continually increasing standards of living to workers, would
mitigate or overcome the class conflicts of capitalism. In the United
States economists associated with John F. Kennedy’s administration
had made the promotion of economic growth a central issue in partisan political economic debate. The strong suggestion in GeorgescuRoegen’s and related work that there were inherent contradictions in
this program of growth economics posed a potentially strong challenge
to it. Robert M. Solow, who had already become famous as one of
the formulators of the seminal and immensely influential Solow-Swann
model of economic growth, took up the defense of the growth program
with his characteristic brilliance and rhetorical charm (Solow, 1973,
1974).
Some of Solow’s defense of economic growth doctrine is directed
6
at the methodologically untidy Club of Rome work, which offends
economists’ sensibilities particularly by neglecting to include any price
signals in its bewildering and somewhat arbitrary array of feedback
loops. Georgescu-Roegen, however, does not fall into this trap. Solow’s
response to Georgescu-Roegen takes the form of a burst of “elasticity
optimism”, in this case optimism about the substitutability of inputs
to production, and the historical resourcefulness of capitalist technological change in finding ways around looming bottlenecks in production. (This estimation of the long-run tendencies of capital accumulation goes back to Karl Marx and Adam Smith.)2
As you would expect, the degree of substitutability is also
a key factor. If it is very easy to substitute other factors
for natural resources, then there is in principle no “problem.” The world can, in effect, get along without natural
resources, so exhaustion is just an event, not a catastrophe. Nordhaus’s notion of a “back-stop technology” is just
a dramatic way of putting this case; at some finite cost,
production can be freed of dependence on exhaustible resources altogether.
If, on the other hand, real output per unit of resources is
effectively bounded – cannot exceed some upper limit of
productivity which is in turn not too far from where we are
now – then catastrophe is unavoidable. In-between there is
a wide range of cases in which the problem is real, interesting, and not foreclosed. Fortunately, what little evidence
there is suggests that there is quite a lot of substitutability between exhaustible resources and renewable or reproducible resources, though this is an empirical question that
could absorb a lot more work than it has had so far. (Solow,
1974)
Many economists felt vindicated in their decision to neglect GeorgescuRoegen’s warnings as a result of the widely-reported 1980 wager beSolow has expressed more cautious second thoughts on these questions. “There is no
reason at all why capitalism could not survive without slow or even no growth. I think
its perfectly possible that economic growth cannot go on at its current rate forever.... It
is possible that the United States and Europe will find that, as the decades go by, either
continued growth will be too destructive to the environment and they are too dependent
on scarce natural resources, or that they would rather use increasing productivity in the
form of leisure.” (Quoted in Stoll, 2008)
2
7
tween Julian Simon and Paul Ehrlich on the price trend of five potentially scarce metals over the next decade. (Ehrlich, who bet that prices
would rise, lost.) Exactly what this particular data point might have
to say about the long-run limits to economic growth is not easy to explain, but Ehrlich’s high profile as the author of best-selling books on
the supposed threat of world population growth predisposed the public
to see the wager as a test of the limits to growth hypothesis. Economic
growth theory barely wavered from its long-standing traditions of assuming Cobb-Douglas production functions (which imply asymptotically zero requirements of any particular input to achieve economically
usable output) and unlimited labor-saving Harrod-neutral technical
change at a constant rate, which imply growth paths with unbounded
growth of total and per-capita output. The considerable literature
that stemmed from the “New” growth theory starting in the 1980s
centered on understanding possible economic mechanisms for the generation of technical change rather than on entropic or resource limits
to growth.
The economist who took Georgescu-Roegen’s line of thinking most
seriously was Herman Daly (who would have been the favorite economic speaker at the workshop I described above), who has consistently questioned the attachment of economists to growth model assumptions that are essentially free of resource constraints to production (Daly, 2008). Daly satirizes the implications of Cobb-Douglas
production functions as implying that it is possible to bake a cake
without eggs or flour as long as the cook whisks the empty bowl faster
and faster.
3
Some numbers
We can make a little progress (though it will turn out to be limited)
in evaluating the Georgescu-Roegen episode by considering some facts
about world energy use and availability.
Solow and other elasticity of substitution optimists argues that
bottlenecks in production arising from scarcity of specific raw materials can almost always be surmounted by technologies provided there
is enough usable energy. This conclusion rests on the protean powers of chemistry to transform the useful properties of matter through
chemical reactions. Since these transformations almost by definition
are entropy-reducing (the output we seek to produce are harder to
8
come by than the inputs) the Second Law of Thermodynamics, as
Georgescu-Roegen points out, dooms these processes to create more
than offsetting entropy in heat and other wastes. It would be reckless of economists to put too much confidence in this argument in
any specific circumstance, because some bottlenecks prove very recalcitrant to economically viable technological solutions, even with very
cheap and abundant energy. Nonetheless, taking the long view, we
can provisionally reduce Daly’s issue of outputs without inputs to the
question of the minimum usable energy it is reasonable to suppose
civilized human life requires.
Units are a potentially confusing (but irrelevant) issue in this discussion. Energy is a stock variable. In standard metric units energy is
measured in the rather unfamiliar unit of joules, in the more familiar
units of kilowatt-hours, or in the exotic units of tons of oil equivalent
and the like. (A joule is a watt-second, the amount of energy accumulated at a power level of one watt during one second.) If we think in
what seem to be intuitively appealing units of, for example, kilowatthours/year it turns out we are really talking about power (that is
energy used or generated per unit time). In fact kilowatt-hours/year
is just kilowatts multiplied by the irrelevant and potentially confusing
constant 8760 hours/year. I will use watts and kilowatts (power) to
measure energy use and generation per unit time.
The human body runs at a power level of about 100 watts. Huntergatherer societies may add a few more watts from campfires and the
like. Wikipedia tells us that the per-capita power rate for the whole
world in 2008 was about 2.5 kilowatts per person. This works out
to a total power rate of around 15 terawatts. (A terawatt is 1012
watts, or, as Americans reckon, one trillion watts.) About 90% of
this power came from burning fossil fuels (U.S. Energy Information
Administration, 2011).
The global average, of course, was composed of very different rates
in different countries and regions. The United States ran at about
10 Kw per capita, four times the global average, European countries
about half that rate, two times the global average. Countries like
India and China where per-capita real GDP is growing rapidly from a
low level, had relatively low per-capita power rates that were growing
rapidly.
World population is currently around 6.5 billion. Projections based
on different methods tend to predict stabilization of the world population (primarily as the result of rapidly falling fertility) in the range of
9
8–10 billion people. (I tried my hand at this exercise using some ideas
from classical political economy in Foley (2000) and came up with an
estimate of stabilization at around 8.5 billion.) For the sake of erring
on the conservative side and making it easy to multiply, let’s work on
the assumption that the world population will stabilize at 10 billion
people.
If we want those people to enjoy an average standard of living
something like that of Europe today, for example, we are thinking of a
world power rate of 50 terawatts, almost four times our current power
rate. Even if we were very clever at conserving energy in production,
transportation, and agriculture, and somehow managed to squeeze a
European life-style out of 3 Kw, we would be thinking of a world power
rate of 30 terawatts. My colleague Lance Taylor argues (Taylor, 2008)
that historically there has been a close positive relation between labor
productivity gains as measured by real GDP indexes and energy inputs
to production, so that even reducing energy input to 3 Kw per person
might prove difficult. Economics suggests that this kind of reduction
is likely only at a considerably higher price of energy than the world
economy has experienced over recent decades.
Expanding world power generation by a factor of two or four might
seem well within possibility, but there are some aspects of this scenario
that are quite breathtaking.
One is greenhouse gas emissions. If we are thinking of expanding
fossil fuel energy production by these factors, we have to envision either a very effective scheme of mitigation of greenhouse gas emissions,
or a very rapid increase in atmospheric concentrations of carbon dioxide. If effective mitigation is put in place, or it turns out that greenhouse gases pose no threat to economic production through mediating
climate change, then we could shrug our shoulders and let coal- or
natural gas-burning power plants multiply. But these are big “ifs” in
themselves.
Another issue is the presumably finite stock of fossil fuels. These
are in fact large, particularly in the case of coal, and have tended to
expand regularly due to exploration and reduction in extraction costs,
but it is not reasonable to suppose that fossil fuel reserves are unbounded. If we looked at this situation from a strictly “sustainable”
point of view (the point of view Georgescu-Roegen explicitly championed) we are talking about replacing almost the whole current 15
terawatts of energy generation based on fossil fuel burning with some
kind of renewable technology such as solar or nuclear, and then doing
10
the same thing again or three times more.
While uranium is abundant in the earth’s crust, current nuclear
technology to generate usable power (basically by generating heat in
slow-neutron nuclear reactors) suffers mightily from the generation of
high-entropy waste that is hard to handle and dispose of. GeorgescuRoegen argues that the only “promethean” energy technology on the
horizon, that is a technology that promises to produce more usable
fuel than it uses, is nuclear breeder technology. Nonetheless, his cautions about the safety of nuclear technology are, if anything, more
compelling today after Three-Mile Island, Chernobyl, and Fukushima
than when he wrote.
But we should not ignore that the ordinary nuclear reactor by itself is not a Promethean recipe; it only replaces
the fossil fuels as a source of heat. The “breeder” reactor,
however, is a Promethean recipe: it performs a qualitative
conversion, of fertile into fissionable nuclear material, and
just like the heat engine, it produces more fuel than it consumes. But the use of nuclear energy in any type of reactor
raises issues about the safety of all life on this planet, issues that are far from even moving toward a settlement.
(Georgescu-Roegen, 1986)
In any case, nuclear power generation has notoriously staggering
capital costs and long times-to build, and the prospect of dotting the
earth-scape with enough breeder reactors to produce 50 or even 30
terawatts generating capacity is far from appealing.
What about solar energy? The solar radiation reaching the earth’s
surface is on the order of 87,000 terawatts (87 petawatts), which at first
glance seems, in accord with Solow’s point of view, to dwarf a measly
30 or 50 or even 100 terawatts (which is about the power equivalent
of the whole biomass fixation of energy through photosynthesis). But
as Georgescu-Roegen points out:
. . . there are several feasible recipes for using direct solar
energy, but a Promethean recipe does not exist yet. The direct use of solar energy does not fulfill the minimal strictly
necessary condition of a Promethean recipe, which is that
some solar collectors could be reproduced only with the
aid of the energy they can harness. But in spite of the loud
din about the solution of the energy crisis by the ’cheap
and renewable’ solar energy, none of the recipes tried out
11
proved that any could be Promethean. The main obstacle
is the extremely weak radiation of the solar energy reaching
the soil. The obvious upshot is that we need a disproportionate amount of matter to harness solar energy in some
appreciable amount. (Georgescu-Roegen, 1986)
Some tentative conclusions can be drawn even from this very sketchy
evidence. Economists would be well-advised, if they believe that “real”
output has a material component, to introduce some minimal energy
input requirement into models of production, along with asymptotically stationary world population. In the medium run (in this context
a century or two) the main difficulty with depending on fossil fuels
for the energy needs of economic production is controlling greenhouse
has emissions. From a longer-run point of view, the problem of providing enough sustainable energy to support civilized standards of
living to a foreseeable world population is not ruled out by thermodynamic limits, but does appear to be on the boundary of what even
very advanced technology might hope to achieve. Figure 2 attempts
to visualize the tradeoffs and limits involved in the long-run between
sustainable human power levels and population.
The problem with Georgescu-Roegen’s intervention is that, while
he makes an indisputable observation that economic production is subject to the physical laws of thermodynamics, and must in that perspective be viewed as involving an irreversible degradation of entropy,
the quantitative importance of these limits even at current levels of
world population and production is not easily established.
4
Cake without flour
Some growth economists might regard the considerations we have just
reviewed as rather quaintly anachronistic in putting so much emphasis
on the material nature of economic production. Well-established patterns of economic growth show that as incomes rise, the proportion of
output as measured by such indexes as real GDP consisting of material goods steadily declines. The major sources of growth in incomes
(and, given the way we measure GDP, in indexes of output) shift to
the tertiary sector, particularly services. The chief input to services
is human intelligence, and at least in some accounts, intelligence is an
unlimited resource. So why couldn’t real GDP, measured to include
the use-value of services, continue to grow without limit?
12
86000 TW
173000 TW
10 Billion Population
100
KW�Person
30 TW
10
5
50 TW
Now
1
15 TW
Preindustrial
0.1
1
5
100 TW
10
20
Population, Billions
Figure 2: World population in billions is on the
horizontal axis, and per-capita power usage in Kw
on the vertical axis. Iso-aggregate power level loci
are drawn for planetary solar flux (174,000 Terawatts), surface solar flux (87000 Terawatts), biosphere fixation of energy (100 Terawatts), 50 and
30 Terawatts, corresponding to a population of 10
billion with current European power usage and
60% of current European power usage, and the
15 Terawatt level corresponding to current world
power usage, mostly generated by burning fossil
fuels. The two points represent preindustrial and
current population and power levels.
13
There are some immediate problems with this conception. Strictly
speaking the production of almost all services does require material
and energy inputs, as the gigantic server farms required for information technology are a concrete reminder. Maintaining the human
capital to provide a glittering array of intellectual services requires material and energy inputs, and these very likely increase as the quality
of intellectual output rises.
But this vision of endless growth without material or energy inputs requires some re-examination of just what it is that we regard
as output and try to measure in indexes like real GDP. Some rapidly
growing service industries, such as finance, seem to be able to produce increasing measured output without much input increase, even
of human employment, at all (Basu and Foley, 2011; Foley, 2011). An
examination of the issues raised by the growing significance of service
industries which have no measurable output raises some deep questions about the conception of economics.
The paradigmatic economic interaction for economic theories rooted
in the marginalist revolution such as neoclassical economics and its
various descendants is a transaction in which one good moves from the
possession of an agent who subjectively values it less to the possession
of another agent who values it more, in exchange for another good (in
many transactions money). As the familiar Edgeworth-Bowley box
construction illustrates, this type of transaction puts both agents on
a higher (or at least no lower) indifference curve, and thus achieves a
Pareto-improvement in the allocation of existing resources. Many financial transactions are of this type, for example initial public offerings
to take companies public, real estate brokerage, insurance contracts,
and other more exotic forms of financial arbitrage. It is important
to remember, however, that the transfer of existing goods or assets
in these transactions is not production. When financial intermediaries appropriate some part of the economic surpluses generated in
these transactions as revenue, however, economic statisticians have
felt compelled to regard the resulting incomes as part of national income, and to invent an imaginary product, financial services, to put
on the product side of the accounts as a counterpart.
It is hard to imagine limits to the magnitude of subjective economic
surpluses that could be realized through transactions of this type. If,
for example, policies or the historical evolution of the division of labor
increase economic insecurity by eroding the institutions of traditional
societies, one can easily imagine an unlimited expansion of insurance
14
transactions as a result.
But from the point of view of classical political economy, it is the
increase in material productivity of labor, not the increase in economic
insecurity associated with the expansion of the division of labor that
is the source of improvements in economic welfare. This point of view
is deeply embedded in the methods of national income accounting,
for example, in the fundamental rule that transactions involving the
transfer of existing assets do not constitute production of goods and
services, no matter how much economic surplus they may represent.
The classical political economists and Marx addressed these issues
through the concepts of “productive” and “unproductive” labor. In
the version of this distinction Marx distilled from his critical review of
Adam Smith, productive labor (whether it produces material goods or
services, since providing haircuts is hard to distinguish from making
hats) returns the costs of production with a profit, while the cost of
unproductive labor is paid out of revenues without any recovery or
return. This classical-Marxian line of thinking puts the origin of the
incomes from the production of “services” like finance in a different
perspective.
This perspective is perhaps most clearly articulated in Marx’s analysis of wage labor and the origins of surplus value. Productive labor is
responsible for the whole value added in production, but receives only
a fraction of the value added in the form of the wage. The resulting
surplus value constitutes a pool of potential revenue for which capitalist producers, landowners, intellectual property owners, financial
firms controlling money capital, and the state compete. The implications of this analysis, which, unfortunately, is for the most part
systematically excluded from the modern economics curriculum, are
far-reaching. No particular capitalist firm, no matter how large in
revenue and employment, can have much direct effect in increasing
the pool of surplus value. Thus “money-making” in capitalist society is proximately based on taking surplus value away from others.
In an economy where resources and intellectual property command
enormous rents there may be a vanishingly small connection between
the revenue of any entity and its actual contribution to production of
useful output.
Many people today are dazzled by the apparently magical ability
of innovators to appropriate enormous revenues on the basis of ideas
and their manipulation alone. This phenomenon has understandably
spawned theories of a “new” economy supposedly based on new prin-
15
ciples of the creation of value. Classical-Marxist political economy, in
contrast, locates incomes to innovation not in new principles of the
creation of value but in new (or newly important, since most of these
“business models” have actually been around for a long time) modes of
appropriation of surplus value. As Slavoj Žižek vividly points out, increasing returns in the appropriation of rents for intellectual property
simultaneously obscure the origin of the resulting enormous incomes
in the pool of surplus value appropriated from productive labor and
mystify the factors behind the increasing inequality in the distribution
of these revenues (Žižek, 2012). The origin of the rent of a particularly exploitable resource like a waterfall or a petroleum deposit is
hard enough to understand, but at least the owner of a waterfall cannot allow an unlimited number of cotton mills to exploit the resulting
usable energy. By contrast the owner of the rights to distribute a
piece of software that, due to network externalities, becomes a technical standard, can allow an effectively unlimited number of users to
install the software and charge each of them a fee.
It would be, however, a peculiar political economy that convinced
itself that the increasing returns in the rents to artificially created
assets like systems software were a remedy for thermodynamically
imposed decreasing returns to resource use in material production.
5
What is “economic growth”?
In a crisis-shaped political-economic environment of debate in which
consensus seems almost impossible to achieve, we hear calls from all
sectors of the political spectrum for a return to “growth”. The right
believes that the private capitalist economy will inherently produce
growth in incomes, output and employment if it is not fettered by state
intervention, borrowing, expenditure, and taxation; the left seeks an
increase in growth through direct expansion of aggregate demand precisely through state borrowing and expenditure. But can we seriously
be satisfied with any path that returns us to pre-crisis patterns of economic growth without addressing some of its increasingly intractable
contradictions?
One side of economic growth is an expansion of incomes and expenditure, which generally seem (if they are reasonably distributed)
to be a good thing in so far as they allow people access to material
goods and services and economic security. The other side of economic
16
growth is an expansion of production, employment, resource use, and
environmental damage, which generally seem to be bad things in so
far as they require onerous effort and have negative consequences on
the health and well-being of current and future generations. In all
traditions of political economic discourse, however, incomes and production are tightly intertwined. We think of the basic way a person
can appropriate an income is to produce something of value (or employ
someone else to produce something of value). Thus it is very difficult
for us to separate the problem of creating and distributing incomes as
claims on the good things of life from the problem of minimizing the
costs of producing what is necessary for a good life.
Political economy emerged in the seventeenth and eighteenth centuries in societies where populations were relatively small, productivity was desperately low, and advances in technology promised the
release of age-old resource and energy constraints through the exploitation of big fossil fuel reserves. In this context we can see why
the main preoccupation of political economy was to promote production, increases in labor productivity, and population through shaping
social institutions such as capitalism to pursue these goals.
We are not by any means there yet, but we can see the outlines
of a very different world economic situation in the not-very-distant
future. In this world a rather large population with very high labor productivity confronts binding resource and energy constraints
on material production. Because labor productivity is very high, a
small proportion of the population’s potential labor suffices to supply
it with a high material standard of living. This small proportion may
comprise a small number of productive workers, or a small portion of
each person’s life-cycle devoted to productive labor. In either case,
further increases in labor productivity promise little improvement in
social life. The technological frontier switches from squeezing more
output out of each hour or work to reducing the entropic resource
and energy costs of material production to an absolute minimum. In
this emerging situation a major question is how people can secure the
incomes that give them a claim to a share of the material production.
Inevitably in this world the link between income and production will
be much more tenuous than in the early or late industrial periods.
The capitalist institutions the world inherited from the industrial
revolution predispose us to continue business-as-usual. In this model
a small full-time productive labor force will be heavily exploited in
the Marxian sense, though perhaps rather well-compensated in terms
17
of absolute real income, to produce an enormous surplus value. An
elite of financial and intellectual property controllers (supplemented
perhaps by some media figures and sports stars) will appropriate the
lion’s share of this surplus value, and the rest of us will scramble to
get income in any way we can. There are, however, other possibilities,
many of which we can see in developments as diverse as medical expertise and open-source software development. Productive labor might
be spread widely but thinly in a pattern in which workers received
large life-time claims to income from short periods of engagement in
productive labor. Thus much of the incomes capitalist institutions
realize as various private forms of surplus value to be competed for
would be converted to life-cycle benefits. Society might find that limiting the scope for increasing returns to intellectual property so as
to forego some excessive and unequally distributed rents would not
sacrifice very much of the incentive for people to innovate.
Georgescu-Roegen may have failed to get the time-line of the process right, but it is hard to argue that he misread the qualitative pattern of change in store for the complex of phenomena we call economic
growth.
References
Basu, D. and Foley, D. K. (2011). Dynamics of output and employment
in the U.S. economy. NSSR SCEPA PERI Working Paper Series.
Daly, H. E. (2008). Ecological Economics and Sustainable Development, Selected Essays of Herman Daly. Edward Elgar Pub.
Ehrlich, P. R. (1971). The Population Bomb. Macmillan, rev ed
edition.
Foley, D. K. (2000). Stabilization of human population through economic increasing returns. Economic Letters, 68(3):309–317.
Foley, D. K. (2009). The economic fundamentals of global warming.
In Harris, J. M. and Goodwin, N. R., editors, TWENTY-FIRST
CENTURY MACROECONOMICS:RESPONDING TO THE CLIMATE CHALLENGE. Edward Elgar Publishing, Cheltenham UK
and Northampton MA.
18
Foley, D. K. (2011). The political economy of U.S. output and employment 2001–2010. NSSR SCEPA Working Paper Series.
Georgescu-Roegen, N. (1986). The entropy law and the economic
process in retrospect. Eastern Economic Journal.
Georgescu-Roegen, N. (1999). The Entropy Law and the Economic
Process. Luniverse Inc.
Meadows, D. H., Meadows, D. L., Randers, J., Behrens III, W. W.,
and Watts, W. (1976). The Limits to Growth: A Report for the
Club of Rome’s Project on the Predicament of Mankind (A Potomac
Associates Book). Universe Books, 2nd edition.
Nordhaus, W. D. (2008). A Question of Balance: Weighing the Options on Global Warming Policies. Yale University Press, New
Haven.
Nordhaus, W. D. and Boyer, J. (2000). Warming the World. MIT
Press, Cambridge, MA.
Rezai, A., Foley, D., and Taylor, L. (2011). Global warming and
economic externalities. Economic Theory, 49(2):329–351. DOI:
10.1007/s00199-010-0592-4.
Solow, R. (1974). The economics of resources or the resources of
economics. The American Economic Review.
Solow, R. M. (1973). Is the end of the world at hand? Challenge,
16(1):39–50.
Stoll, S. (2008). Fear of fallowing: The specter of a no-growth world.
Harper’s Magazine. http://harpers.org/archive/2008/03/0081958.
Taylor, L. (2008). Energy productivity, labor productivity, and global
warming. In Harris, J. M. and Goodwin, N. R., editors, TwentyFirst Century Macroeconomics: Responding to the Climate Challenge. Edward Elgar, Northhampton MA.
U.S. Energy Information Administration (2011). International Energy Outlook: 2011. Technical Report Report Number: DOE/EIA0484(2011), http://205.254.135.7/forecasts/ieo/.
Žižek, S. (2012). The Revolt of the Salaried Bourgeoisie: The New
Proletariat. LRB, (26 January 2012):1–7.
19