Joan Robinson on Environment and Ecology

Introduction

It was once observed by Winnett (2003, p. 123) that “it seems obvious that post-Keynesian economics can provide insights which have the potential to provide a powerful critique of neoclassical environmental economics and pointers to a reformulation of this subject.” This author outlined some of these insights. In fact, the preoccupations with ecological and environmental issues among the so-called heterodox approaches, including Marxist, Sraffian, and structuralist, have been increasing, in order to cope with the orthodox interpretations and obviously with the environmental problems per se.

Joan Robinson was increasingly worried about the questions of pollution, scarcity, and externalities. Comments and analyses on ecology and environment are spread over her work and tended to intensify in her last years (see Harcourt & Kerr, 2009). Her archives reveal preoccupation and pessimism with these topics, eventually shared with her associates. In fact, she always dealt with these questions in one way or another, though not always with the same emphasis or always from a theoretical point of view. This fact could be used to argue that Robinson did not have relevant things to say about these questions. This article partially disagrees with that on two grounds: (a) these questions became more frequent over her last works; (b) her work has potential insights for a critical approach of the current ecological questions that were not fully developed.

Methodologically, this work on the history of economic thought lies between two different genres, such as once described by Mark Blaug: one that gives “an account of past thinkers’ systems in ‘their own terms’, that is, in terms, these thinkers would have accepted as a correct description of what they had done;” and another by means of which we treat “great dead thinkers of the past as contemporaries with whom we can exchange views” (Blaug, 1990, pp. 27–28). With regards to this second genre in the history of economic thought—“rational reconstruction,” in Blaug’s terms—this leads to political economy itself, as Robinson’s ideas are used to enlarge the current heterodox critiques of the neoclassical perspectives.

In this light, the article summarizes Robinson’s views, analyzes them against the existing literature, and finally evaluates, critically or not, her contributions. The work begins with the notion of (relative) economic scarcity and how it appears in Robinson’s work. Robinson’s perspectives on capital accumulation and technical progress are then discussed in light of Georgescu-Roegen’s (1971) economic interpretation of the entropy laws. Finally, Robinson’s perspectives on capital theory are tackled. The paper shows how the author did not develop some of the logical consequences of her own capital theory, in contrast to self-declared preoccupations with pollution and absolute scarcity. However, although Robinson was not able or interested in exploring fully the logical consequences of her own work for an ecological critique, her theoretical perspectives can still contribute to a wider heterodox approach to these issues. In broader terms, it is argued—in line with Winnett (2003), Holt et al. (2009), and Perry (2013)—that Robinson’s contributions can indeed contribute to a critique of the neoclassical hypotheses regarding environmental issues.

Scarcity and Prices

Until very recently several authors still could define the heterodox approaches as simple-minded, incapable of understanding and theorizing profoundly environmental issues, particularly the questions of externalities, intertemporal allocation of resources, the role of prices, or the so-called entropic processes. In fact, in the case of post-Keynesian approaches, they “remained virtually silent on environment issues until the new century, contrasting sharply with other economic paradigms” (Perry, 2013, p. 3). So, for example, the ecological economists Alier and Schülupmann (1987, p. 128) observed:

Keynes seemed to believe that wealth—and not debt—increased according to the rules of compound interest, a “fact” which he opposed to the Malthusian population “law.” [But as Soddy showed] capital could not really be stored, as it was subject to a law of continuous decrement, because, in physical terms, it was energy embodied in certain objects, subject to the law of entropy.

However, this scenario was altered, and there have been several attempts to reconsider environmental issues through the lenses of heterodox frameworks. Despite that, Robinson’s specific views on environmental questions have not been analyzed yet in a single work.

Ruth Cohen, Robinson’s close friend, observed after her death (quoted by Feiwell, 1989, p. 103):

[i]n the last few years of her life she became very depressed at the state of economic doctrine, struggling towards a theory that could create models that could grapple with history, with ecological balances in individual communities, and particularly with technological change. She was trying to evolve a different technique of thought.

Harcourt and Kerr suggested that Robinson

identified a second crisis in economic theory (the first being its inability to handle the interwar slump), the lack of a suitable framework with which to tackle the terrible problems of modern economic life: poverty, racism, urban puzzles and pollution, excessive population growth and war. (Harcourt & Kerr, 2009, p. 16)

These authors claim that Robinson even passed through an epistemological change along her last years of intellectual production (Harcourt & Kerr, 2009, p. 202). In turn, Millmow (2003) suggests that she died in depression in the context of the world political situation and the state of the economic theory. Be that as it may, Robinson’s work contains many references to the environmental issues. How did she deal with them?

From an epistemological point of view, Robinson accepts logical-positivist premises and complaints about “metaphysics”—as neoclassical economics—but in her case, they are mixed with a progressive, left-wing ideology, which included an increasing commitment to the autonomous and socialist development of the underdeveloped nations (see, particularly, Robinson, 1962a, 1979a). She claimed that science should be distinguished from ideology, though maintaining that the latter was “indispensable in the world of action of social life” (Robinson, 1962a, p. 9). In this case, her “ideology” has not a “value-free” perspective regarding the way of life of modern industrial societies (including the [former] industrialized socialist nations); on the contrary, it is critical and “negative.” Though both of these theoretical frameworks tend to assume positivist approaches in methodological terms (as opposed, say, to a dialectical approach), Robinson accepted inductivism, with neoclassicals opting for deductivism (more in line with Karl Popper’s approach to scientific research; for different methodologies between the two approaches see Kaldor, 1966).

Thus, the alleged rationality of private investment decisions concerning social interests is not accepted by Robinson. Private markets do not constitute the best way of administrating environmental issues in the interests of the majority; the Coase solution of privatizing natural resources and the permission to pollute is a false solution to a deeper problem related to the qualitative question: “what is growth for?.” This point of departure led Robinson to tackle some questions concerning environmental issues, but not others. At the same time, although the majority of her positions regarding the “ecological” questions tend to overlap with the critiques of “heterodox” authors such as Georgescu-Roegen and Kenneth Boulding (particularly in her mature work), there are meaningful exceptions and absences. We shall now summarize and evaluate them, beginning with “scarcity.”

Scarcity engenders the efforts of producing goods, which after their production have “value.” “Capital yields a return not because it is productive but because it is scarce” (Robinson, 1962a, p. 73). Value is distributed between groups and social classes. In a monetary economy, this distribution is made through the price system, with goods transacted in markets through a complicated mechanism that departs from production, the rate of profits, and the distribution conditions, and, in the case of primary goods, the fertility of land, geographical position, environment, etc. Once in the markets, prices finally rely somehow on the “needs and fancies of individuals” (Robinson, 1960, p. 205). If “conditions of tranquility” do not prevail, reproduction prices, or “normal prices,” cannot exist. “All these complications destroy the simplicity of the labor theory of value as an account of the determination of the relative prices, but it remains valid as a rough generalization” (Robinson, 1956, p. 353). Basically, this is the core of Book VII of The Accumulation of Capital, entitled “Relative Prices.”

The author argues that (Marxian) labor theory of value is insufficient to understand the formation of prices in primary goods markets. In general, prices of raw materials and primary products do not have an exact relation with “labor-values,” as their prices are formed in sometimes extremely fluctuating markets, which do not match exactly the costs of capital and wages. In her words (Robinson, 1956, p. 334), “[i]nvestment in opening up land (including mines, oilfields, etc.) is different in an important respect from investment in capital goods. The value of the resources obtained bears no relation to the cost of the investment.” It is so because “over the long run the stock of capital corresponds more or less to the sum of all investments made,” but regarding natural resources, “a small investment may yield a huge increase in productive capacity,” and so in supply (Robinson, 1956, p. 334). On the other hand, “primary production, such as agriculture and mining, is strongly influenced by the vagaries of supply and demand and general swings of profitability with booms and slumps” (Robinson, 1979a, p. 35).

Robinson claims for the usefulness of the Marshallian framework to evaluate price behavior, though it would be impossible to understand prices without the existence of oligopolistic structures. “The free play of market forces creates conditions that will not be tolerated by a community that has any power to control them” (Robinson, 1962a, p. 64). So, we find in Robinson’s framework the hypothesis that, as Holt and Spash observed, “in the real world, prices reflect market power and current short-term perspectives only. They are also affected by speculation” (Holt et al., 2009, p. 15). But what about the other topic, that because reproduction prices “do not reflect scarcity [in an ecological sense], we cannot count on prices to deal with current overutilization of resources or to protect the planet from the enhanced greenhouse effect” (Holt et al., 2009, p. 15)?

There are many passages that indirectly refer to this second question. See, for example, the following statement in Essays in the Theory of Economic Growth (Robinson, 1962b): what if the economy is facing bottlenecks, maybe the basic of all environmental problems? We read the following (Robinson, 1962b, p. 75):

Natural resources. Disproportionalities. As the whole economy develops, most demands expand, and some run into bottlenecks of limited supplies of natural resources. During a period of scarcity for a particular product its price is high and its producers prosperous. Users set about to find substitutes and alternative sources of supply may be opened up. Prices fall and prosperity comes to an end. The new sources of supply, though called in existence by high prices, once developed, are not necessarily high-cost producers; and in any case they are not easily forced out of existence again. In such cases, a short-period in a seller’s market leads to a long period in a buyer’s market.

She observed in Aspects of Development and Underdevelopment (Robinson, 1979a, p. 40):

Depletion of resources. There is another problem which is of urgent importance today that cannot be well represented in terms of value. Marx assumes that the depletion of a pre-existing stock of means of production (represented by c) is continually made good by current labour. But some depletion, in particular that of minerals, can never be made good. The increase in productivity that comes about with technical change is very largely due to the use of energy to supplement human work. Mineral sources of energy—coal and oil—are now being used up in the industrial economies, socialist as well as capitalist, at a prodigious rate.

Robinson shows here awareness of the problems of “seeding with oil” and the broader depletion of natural resources, including the underdeveloped nations, but these sentences end and the author goes on discussing other aspects not related to prices. The same occurred with the fish examples in “Time in Economic Theory” (Robinson, 1980b, p. 93). The question is raised, but then the argument is conducted in terms of uncertainty, leaving aside depletion and entropic processes. Departing from that, which vignettes could have been developed on the ecological lines, which Robinson seemed to accept?

Rising prices may indicate increasing absolute scarcity. Growing scarcity will lead to prices even higher. In turn, higher prices will drive to an elevation of the supply; and so on, till the resource is exhausted. Through technical progress, a new substitute comes in. And would it be fanciful to pose the possibility that new substitutes are also not available because they also depend on some material resource? Would the economy react with a decreasing material (or energetic) intensity (“energy-saving bias” and/or “material-saving bias”)? Will the depletion of resources, together with the worsening of the general conditions of the environment as a whole, lead to a decreasing capacity of creating economic surplus, in face of an hypothetic increasing “material-using bias” or simply extensive growth? There were many opportunities for theoretical and empirical research which could make good use of Robinson’s approach regarding the theory of capital, which, however, she did not explore.

One possible logical conclusion of the above-mentioned analyses regarding the price of fish and its absolute scarcity could have been, in the words of Nove (1991, p. 6):

[c]onventional economic analysis and the normal market mechanism are not well attuned to handling physical shortage. This is no doubt one reason why central controls are usually imposed in wartime. Or take another example: fish in the North Sea. Shortage causes higher prices, which stimulates further efforts to catch fish, which makes the shortage worse, and so on, until there are no fish. A higher profit is supposed to act as a stimulus to higher output, on the implied assumption that this does not run into physical limits which make higher output impossible. This is why, in the case of fish, government regulation is necessary. Such instances could become more common in the future, as issues of environmental protection have become already.

These observations assume the ecological premises that the depletion of many of the economic resources, particularly the non-renewable ones, have practical implications not only in the long run, and that the price system is not an adequate mechanism of avoiding the absolute increasing scarcity.

In addition to that, there are at least two main “cosmic aspects”—as Solow (1974, p. 7) called them—regarding scarcity that cannot be exactly “revealed,” or expressed, through monetary prices (unless with a direct intervention of taxes, and this with clear limits): the first is that prices as reproduction prices do not express the oncoming generation’s evaluation of what society is doing today with the natural resources on which they will also have to rely someday. Robinson was aware of this intergenerational distributive conflict, but she did not interpret it in terms of the relation between scarcity and prices. She once observed in one of the models in The Accumulation of Capital:

[s]ettlers are dominated by business morality, and think of capital in terms of purchasing power rather than in kind. They have no scruple, therefore, in extracting value from the soil by mining its fertility, and if they can leave capital in terms of money to their children they think no harm in leaving deserts to posterity in general. (Robinson, 1956, p. 297)

Here the problem of the absolute scarcity and irreversible damages were assumed, but there is not a discussion of prices as eventual inadequate mechanisms of a (fair) intertemporal distribution.

Neoclassicals assume unlimited possibilities of substitution between “natural” capital and “human capital”/“reproducible capital.” Societies can use and even exhaust natural resources, but in the end “substitution,” “Coasian property rights” or even the price system as such will avoid absolute scarcity. We saw how the price mechanism cannot, eventually, avoid absolute scarcity; but, even if it could, all irreversible and intangible damages caused by industrialized societies to the air, water, soil, animal life, etc., cannot be easily priced, on ecological grounds. The buying of a “permission to pollute” or related taxes will only lower the “rate of discount” regarding the future, but will not eliminate externalities coming from production and consumption as such. In fact, it would be difficult to translate these damages in the form of rights to pollute, taxes, and so prices. Robinson also does not tackle directly these questions even in her wide-ranging Economic Philosophy (Robinson, 1962a) or Freedom and Necessity (Robinson, 1970). Here the opportunities for the theoretical discussion, including economic philosophy, on the basis of Robinson’s premises were (and are) also many, with further practical consequences for a reorientation of the economic policies and the current ways of life (Arestis & Sawyer, 2015).

Economic Surplus and the Nature of Capital

One of Robinson’s last mentions of the question of natural resources and depletion was made in Aspects of Development and Underdevelopment (Robinson, 1979a), wherein she advocates the surplus approach of Marx and Baran to deal with development questions, claiming that “it is better to return to the classics to see how their conceptions can be adapted to deal with our contemporary problems” (Robinson, 1979a, p. 18). She accepts the classical definition of a “flow of production and consumption of material goods as a self-reproducing system yielding a surplus over necessary costs” (Robinson, 1979a, p. 18). Then she enlarges it to include manufacturing and services, discussing Marx and Sraffa, who defined surplus as “the more than the minimum necessary for replacement of capital” (Sraffa, 1960, p. 7). Thus, with corn, there is tangible and physical notion of surplus: “corn was sought as a measure of absolute value as there could be a ‘material rate’ of surplus in it, in the sense that the same product appeared both as net output and input” (Bharadwaj, 1971, p. 187).

Robinson accepted somehow the concept of irreversible entropic degradation: “natural resources are not indefinitely expansible, and land may set a permanent limit to economic growth” (Robinson, 1979b, p. 21). At the same time, she implicitly assumes, in line with the ecological critique, that this is not a problem of the “long run.” It has relevant consequences for the present societies and those who will come: “pollution and the irreparable loss of resources, both for production and pleasure, provide the most obvious and notorious objections to the doctrine that the free play of market forces in a regime of laissez-faire leads to beneficial results for society as whole” (Robinson & Eatwell, 1973, p. 310). But in entropic terms—if this notion is to be openly accepted—this definition is incomplete.

The question of the entropy law applied to economics—that is, “all production involves transforming low-entropy (usable) energy and material into high-entropy (unusable) forms” (Perry, 2013, p. 4)—is one of the most controversial in the history of sciences, including economics. Paul Samuelson, Georgescu’s personal friend (and to whom Georgescu dedicated one of his heterodox books, Georgescu-Roegen, 1976), conceded that in the long run there will be “non-sustainable equilibrium” because of the entropy law. In Samuelson’s (1999, p. 17) words:

[w]hat underlies this inevitability [is] Georgescu-Roegen’s Law of inevitable Dissipation of Useful Concentrated Matter. This is a good-sense certainty, not some esoteric principle probability of physics Boltzmann and Clausius: only if all useful copper and gold and helium could be 100 recycled would a perpetuum mobile be possible. If pigs could fly, if my Aunt Sally were a stagecoach, if…, and if…

However, Samuelson, Nordhaus, Solow, and neoclassicals deny that this conclusion has important consequences from the point of view of teaching and theorizing economics with traditional neoclassical premises; for example, there is no mention of Georgescu-Roegen’s ideas in the many editions of his Economics, including the last (Samuelson & Nordhaus, 2010). As Solow (1974) observes, this “long run” has nothing, or little, to do with the current generations or the next to come; in the short (and medium) run, technical solutions will fix the problems. These authors, as the majority of the neoclassicals (or some Marxists), opted for not denying the entropic process per se but argued that it is irrelevant for practical purposes, or that it can be compensated.

According to Georgescu-Roegen, the entropy law is the “most economic of all physical laws” (Georgescu-Roegen, 1971, p. 280): the available energy and matter are constant, but their total disorder irrevocably increases. It covers “not only energy but also material structures” (Georgescu-Roegen, 1971, p. 142; see also Georgescu-Roegen, 1976), and it cannot be understood with traditional logics and mechanistic physics. Despite many similarities between their works, particularly by assuming the practical consequences of irreversible natural damages and increasing (absolute) scarcity, Robinson’s did not take Georgescu-Roegen as an interlocutor, and his contributions to a critique of the equilibrium theory, the substitution hypothesis, and the entropy law, which today are being accepted as “scientifically” valid (Holt et al., 2009, particularly chapter 10). She never refers to entropic processes, though apparently accepting its practical validity in the form of the impossibility of complete recycling and substitution. Regarding the relation between the levels of economic activity and the utilization of energy, Robinson (1979a, p. 40) observed:

Professor Reddy has pointed out that, comparing countries, there is a clear correlation between GNP per capita and the consumption of energy per capita. Public opinion in the Wealthy countries has begun to recognize that their life style and their technology are excessive wasteful, though no one has much idea what to do about it. For Third World countries to overcome poverty by industrializing on the Western pattern is out of the question. Certainly they need accumulation but they need to direct it into forms sustainable to their own situation.

Here, Robinson seems to agree with the hypothesis of the (material and energetic) “limits to growth”—explained, in a more formalistic perspective, by the entropy law. So, if the entropy law is to be openly assumed, it means that: (a) total recycling is impossible; (b) the exhaustion of resources is a relevant phenomenon not only in the very long run; (c) the natural capacity of Earth to provide the physical conditions to the creation of surplus will decrease with time, because of the above-mentioned process of irreversible entropic degradation of the resources (arable land, mines, clean water, etc.). From an entropic perspective, what would indeed be the logical conclusions of her own implicit premises?

If mitigation of the already polluted areas and already existing environmental damages, and the abandoning of hazardous forms of production and consumption, are all included in the relevant relations between investment and consumption, the consumption levels per head will tend to fall. In this case, it would be necessary to develop a better definition of economic surplus, according to the following aspects: the economic surplus is the physical difference between production and consumption made possible by a previous existing stock of low-entropy resources, which are degraded over time; the progressive depletion of the low-entropy stocks (or funds) makes more difficult in economic terms to create the economic surplus; mitigation of negative externalities and the necessity of recycling both reduce the available economic surplus and so the standard of life. Models of accumulation should then have adopted “recycling” and “mitigation departments” to see how they interact with “investment” and “consumption” sectors, and with an economy undergoing progressive depletion of natural resources. “Mining” and alike should be excluded from the “productive” sectors, as they in fact do not produce, but only extract.

These new “departments” should be responsible for partially reversing the increasing depletion of resources and other general damages to environment, gaining increasing importance, side by side with agriculture, which should take increasing advantage of the sunlight. Regarding this, Robinson’s (1979a) arguments for the case of land reforms in developing nations were correct in ecological (we could say, entropic) terms. In this case, incorporating environmental categories could have been an opportunity to enhance the course of Robinson’s theoretical treatment, given the similarities between the surplus approach and the entropy itself in terms of their material or physical conceptions.

In broader terms, if the entropy process is to be taken as relevant for our present and future, an explicit assumption of it (entropy) may lead the Marxist, Sraffian, and structuralist approaches to a more precise definition of what is economic surplus, making it more all-encompassing. Robinson’s answer “What is growth for?” would then assume an even more disturbing and relevant aspect regarding the contemporary urbanized and industrialized societies—a perception that may have reinforced her pessimism (Millmow, 2003). The categories of environmental economics could be used to make a more rational critique, but they would undermine the search for far-fetched “full employment” policies, particularly in the urban areas.

Growth, Technical Progress, and Equilibrium

Externalities and depletion are related to numbers, or growth itself. Any community has some negative impact on environment, increasing scarcity, and damaging natural life. Though some partial mitigation is also possible, the growth of material and energetic consumption per head intensifies depletion and externalities, being more exacerbated if the population is growing. Neoclassical theory cannot deny that a rich individual has a higher “ecological footprint” than a poor one. Robinson seemed to accept these generalizations. She wrote in 1951 that if the (wrong) predictions of a population of four million in 2035 in England were correct, there would be “an agreeable picture of England’s green and pleasant land, exporting the product of a highly mechanized agriculture, running two trains a day over an electrified railway system, and obtaining coal entirely by surface mining” (Robinson, 1951, p. 130). Despite this scenario being “too fanciful to be of serious value,” it is clear that the growth in numbers caused, per se, more pollution, more consumption and—why not?—a deterioration of the aesthetic aspects of daily life.

On the other hand, the higher the wealth per head, the higher the burden on nature. This word, burden, means here irreversible loss, in line with the critique of the notion of equilibrium and reversible processes that constitute premises of the neoclassical analysis. The idea of a small and/or decreasing population with equal opportunities, mechanized industry, and agriculture, and a more or less moderate consumption per capita, seems more plausible to Robinson than a scenario of an overpopulated nation, developed or not. The multiplication of patterns of high standard consumption destroys the “utility” theory. In Robinson’s (1962a, p. 109) words,

The most noticeable effect of a growth in numbers, however, when it occurs at a high standard of life, is the way human beings destroy amenities for each other through cluttering up the country with their bodies, their houses and their motor-cars. The external diseconomies of consumption are then so marked as to leave utility theory completely in ruins.

So, according to the author, there is a contradiction between a theory that supposes individual’s freedom as a necessary condition for maximizing general “utility” and “efficiency” and the fact that an individual who is maximizing his/her utility by means of consumption is maximizing the general levels of depletion. In the end, according to this, the maximization of consumption will lead to the maximization of externalities and depletion. We could add that, if the surplus approach assumes entropy as a relevant process, the higher the consumption levels, the higher the necessity of recycling and mitigation, and the lower the levels of consumption per head with time.

Growth leads to scarcity in relation to natural resources, particularly land. Robinson did not develop models with absolute scarcity regarding other natural resources than land, but she claims that the “endowment of natural resources enjoyed by an economy is obviously of the highest importance” (Robinson, 1956, p. 67). In the long run, only the peasant’s morality is sustainable Robinson, 1956, p. 35):

[i]ncome consists in the kindly fruits of the Earth, and capital in the fertility of the soil. Animal and human economies can flourish for a time by mining the soil, creating deserts, or by preying upon other economies, but to be viable over a long run, in peaceful conditions, an economy must be impregnated with the peasant’s morality.

Robinson does not see these questions as if they were in a social void. Differently from Samuelson or even from Georgescu-Roegen, class divisions and their different impacts on resources are highlighted (for a class-based critique of Georgescu-Roegen, see Burkett, 2006). In the review of Richard Wilkinson’s Poverty and Progress (which was published in 1973), “Formalistic Marxism and Ecology without Classes,” Robinson (1980a) criticized the idea that the laws of the economic history could be explained by the hypothesis of an “ecological equilibrium” and a “community without classes.” Robinson (1980a, p. 246) poses the question thus:

[Wilkinson] traces the story of how shortage of wood led to coal mining and coal mining led to steam pumps and so to all the rest. But how much was the despoiling of woodland due to growing numbers in the countryside (as happened in India) and how much to rising middle-class consumption in the towns?

That is, the “despoiling of woodland”—nature, consequently—can be due both to (a) growing numbers and (b) high-income consumption. The consumption per head of a rich individual is obviously higher (“ecological footprint” in ecological terminology), but it is not possible to disregard the former as another way of increasing Earth’s losses. Here Robinson’s almost resembles or overlaps with the “degrowth” approach, but regarding this specific issue, her approach comes in fact to be superior, because class and national subdivisions and their different impacts on nature are not excluded. Aesthetics, fruition, enjoyment of depopulated cities and countryside, etc., are an additional part of the argument. But Robinson’s point is still based on a purely empirical statement, which relates directly numbers, consumption, and externalities with irreversible natural losses—denied by the neoclassical substitutability hypothesis, but supported by ecological economics. Here lies one of the main divisions between both of these theoretical frameworks: the centrality of, say, “quantity.” With regard to this point, Robinson was clearly already on the side of the ecological economics.

According to this view, the increasing absolute scarcity of natural resources, caused by economic growth, will engender disequilibrium. Departing from (implicit) entropic lines, increasing scarcity means disequilibrium. As we can read in the “Land” section in The Accumulation of Capital (1956), scarcity of land or raw materials per se can alter the rate of economic growth. Scarcity of land or primary goods can lead to depression as “the change over from geographical expansion to accumulation in a given space involves changes in techniques,” and the “entrepreneurs may be disconcerted;” then “if they fail to adapt themselves and merely wring their hands over the scarcity of land, investment falls off and stagnation sets in” (Robinson, 1956, p. 332).

When the frontier population “ceases to grow, the expansion of external demand comes to an end,” and the capitalist economy may also face depression (Robinson, 1956, p. 333). If immigration fails to recruit population and increase of population suddenly falls, accumulation at the old rate can continue only by raising the degree of mechanization, but at the same time depressing the rate of profit (Robinson, 1956, p. 333). In the long run, the scarcity of land or primary goods will rise. Accordingly, the relation between the scarcity of land, rising rents, and then falling rates of profit, such as originally discussed by Ricardo, becomes important again: “considering the rate at which the population of the world is growing, all aspiring to attain the per capita level of destruction of natural resources now prevailing in the United States, it seems as though Ricardo’s problem may well become actual before long” (Robinson, 1962a, p. 95)—a statement that would be endorsed by ecologists (Alier & Jusmet, 2003, p. 320).

Scarcity of labor, land, and resources stimulates technical progress. But this interrelation is not a simple process. The contradiction between technical progress, the increasing supply that in the short run it makes possible, and the overall scarcity in the long, was defined by the literature as the “Jevons Paradox:” “as technological improvements increase the efficiency with which a resource is used, total consumption of that resource may increase rather than decrease” (Tainter, 2008, p. 10). In this case, if the validity of this assertion is accepted, it would not be completely exact to claim that “the difficulties are political and economic rather than technical, [the main difficulty arising] from competition” (Robinson & Eatwell, 1973, p. 310). Supporters of the Jevons Paradox claim that the very technical progress is contradictory and to some degree an uncontrollable process, even in the absence of competition. Even for an isolated socialist economy, it seems odd—looking from this point of view—to suppose that technical progress should be stopped given long-run assumptions concerning scarcity, or even increasing short-run scarcity or natural damages.

Anyway, it would be hard, from a heterodox perspective to accept “well-behaved” production functions relating smooth substitutions of “natural capital” and “human” or “reproducible capital” that could maintain “forever” the levels of consumption of a given community. Changes in technical progress lead to decreasing costs, but decreasing costs lead to higher consumption, and so towards increasing scarcity of non-renewable resources. Here dialectics may be a more comprehensive epistemological tool to understand what the positivist literature is probably mistakenly calling a “paradox.”

Robinson seemed aware of the consequences of the “capital critique” applied to the neoclassical hypothesis of a perfect technical substitution between “capital” and “natural resources.” She observed (Robinson, 1956, p. 311):

Looking at the matter in philosophical light, the reason why there is no meaning to be attached to the marginal product of “capital” is that, from a long-run point of view, labour and natural resources are the factors of production in the economy as a whole, while capital goods and the time pattern of production are the means by which the factors are deployed.

There was here another opportunity of extending the Sraffian theoretical principle to an ecological critique, as the literature lately has demonstrated (Perry, 2013; Winnett, 2003): not only capital but “nature” itself (exhaustible raw materials, clean air, and water, etc.) are not “jelly” and cannot always be technically “substituted” (see also Harcourt, 1972). There are resources that just cannot be substituted at all; an individual can ride a bike instead of driving a car, but she cannot drink or breathe irreversibly polluted water or air (in the case of nuclear disasters, for example). At the same time, all interrelated substitutes are made in the last instance of some tangible resource. As Mayumi and Gowdy observe (1999), entropy is a condition to all substitutes, imposing a limit to substitution, particularly of “reproducible capital.” Entropy law does not deny the possibility of an intense process of accumulation of capital and material abundance (nor the existence of processes that partially reverse or delay the rise of entropy), but states that these kinds of events are not “sustainable” in the long run, because the higher the consumption rate, the higher is the speeding up of the degree of entropy on Earth. That is, entropy is accelerated by the economic growth and technical progress. So, in dialectical terms, if the entropy law is to be accepted, it could be argued that relative scarcity gives place to absolute scarcity over time. This is so because, in concrete terms, even if we specify a “quantity of capital” as a “number of dollars or a stock of productive equipment,” or “a number of tons of putty,” “they are not out of the wood.” In Robinson’s (1980b, p. 87) own words:

Two points on the isoquant represent two different techniques of production, one with a higher ratio of putty to men employed than the other. A movement from one to the other would involve augmenting the stock of putty or dismissing some workers. Before we can go on with the story, we want to know which.

Then the concrete (material) differences between different energy supplies cannot be ignored by an inductive-historical methodology as Robinson’s profess, for whom “we should have a long-run theory based on past experience” (Robinson, 1980b, p. 94). Being that correct, the uncertainties surrounding all kinds of alleged “suitable adjustments”—in Harcourt’s (1972, p. 106) expression, are so great that forecasting “ergodic” or “balanced” processes of substitution, mitigation, “fair” intertemporal allocation of resources, and so on, seems far-fetched and against the very common sense.

In fact, another of Robinson’s critiques of the equilibrium concept could be adapted here to contest the hypothesis of perfect substitutions and complete mitigation that could maintain “forever” the same level of consumption per head, with no distributive conflicts. In strict economic terms, the promise of a smooth technical solution to depletion and absolute scarcity means leaving aside the Keynesian uncertainty principle, with “historical events [being] introduced in a timeless picture” (Robinson, 1980b, p. 87). The fact of having exhausted the majority of the low-entropy resources, and the several uncertainties involving the possibility of maintaining the current levels of consumption per head in the long run, seem to validate Robinson’s observation that “it is only by interpreting history, including the present in history, that economics can aspire to be a serious subject” (Robinson, 1980b, p. 90).

In light of the entropy laws, the conclusion is that industrialization—which was originally a successful struggle against (relative) scarcity—is creating absolute scarcity in the long run; here lies the importance of agriculture, particularly in the underdeveloped nations, given that agriculture produces economic goods with an infinite source of energy (sunlight). In this respect, we reach again the epistemological difference between the economic theories which are based on induction and those which assume deductive “Newtonian” mechanistic principles. Georgescu-Roegen and Joan Robinson, despite their different epistemological assumptions, would agree that there is only one history, with its own particular and irreversible past, and an unpredictable future. “Evolutionary laws are not amenable to analytical expressions” (Georgescu-Roegen, 1976, p. 209). In fact, as we can see, the notion of time is at the core both of Robinson’s positivist approach based on inductivism, and the dialectics behind Georgescu’s economic interpretation of thermodynamics. Thus, a common ground between these two different epistemological (and even ontological) perspectives can be found in order to build a multifarious ecological critique.

Conclusion

This article aimed at summarizing Robinson’s views on the relation between environment and economics, particularly when concerning the role of prices, economic surplus, capital theory of even economic philosophy. We saw that Robinson, already in the 1950s, was aware of the problems related to Western patterns of consumption in an assumed finite world, and of the problem of the intertemporal allocation of resources. Robinson’s remarks were, in general, implicitly in line with the heterodox critiques of the neoclassical assumptions such as those developed particularly by Georgescu-Roegen. When emphasizing the disconnection of investment and consumption decisions in capitalist markets from major concerns regarding the next generations, Robinson showed herself aware of the problems of intergenerational equity and valuing the environment. She was also aware of the likely decreasing returns in agriculture and mining, though it was too early to see all the potentialities and threats of, say, genetically modified organisms. She also saw the generalization of the Western pattern of industrialization as materially impossible, and other aspects.

However, the passages quoted here appear in works whose main purpose was not the discussion of environmental problems per se. They came in general as by-products of other economic debates, and it explains partially Robinson’s limits when approaching these issues. Robinson’s references regarding environment and ecology are diluted in a work whose main preoccupations were to extend the post-Keynesian theory to the long run, to attack the equilibrium principle and also the theory of the “marginal productivity of capital.” She departed from the Keynesian aggregate demand principle and then from the Sraffian critique, along with her own intellectual achievements (see also Harcourt, 1972). She also did not take into account many fields of environmental economics, such as environmental accounting, the Coase theorem, and the practical implementation of environmental economic policies. Despite worrying about the ecological issues, she avoided mentioning, quoting, or even debating with the leading figures such as Kenneth Boulding or Georgescu-Roegen, or with neoclassicals such as Robert Solow or Paul Samuelson.

This position is at least partially responsible for the fact that Robinson never became a reference among the economic literature on environment, or that post-Keynesians and other heterodox theorists did not emphasize environmental questions until very recently. It also expressed—may be—an excess of confidence in the economist’s profession, ignoring interdisciplinary approaches. Despite Robinson’s apparent or alleged epistemological breakthrough, she continued to be identified with the struggle for (“capitalist” or “socialist”) full employment as an economic right of the working class. This is ironic because Robinson implicitly accepted many of the hypotheses advanced by Georgescu-Roegen and other heterodox approaches of the relation between economics and environment. However, her theoretical perspectives can still contribute to a critique of the neoclassical framework on environment, a fact that the heterodox economic literature is currently showing, which includes a reconsideration and an emphasis on the role of agriculture for the development process on ecological grounds.