In the eighteenth and early nineteenth centuries as political economists struggled to understand how the new economic system which came to be called “capitalism” works, some focused on reasons to believe the system was “self-regulating” and capable of sustaining progress indefinitely. But many prominent political economists writing before Marx worried that the system was prone to crises, and that capitalist “development” would prove to be unsustainable in some way or other. Economics had already been labeled the “dismal science” by the time Marx came on the scene, and not because it was boring, but because so many economists predicted a dismal ending.1 However, Marx predicted dismal endings for different reasons than Thomas Malthus and David Ricardo before him, and formulated his crisis theory in Hegelian terms as “internal contradictions.”
Malthus is most famous for his theory of immiseration due to population growth, but he also worried about “general gluts” – an oversupply of output which exceeds the demand to buy it. In a series of letters between them, Ricardo famously assured Malthus that general gluts could only be fleeting because whatever income some people saved, banks like his own would quickly loan to businesses eager to invest as long as interest rates were free to rise and fall to bring saving and investment into equilibrium. In other words, Ricardo argued that as long as income saved and not spent on consumption was loaned to others who spent it on investment goods, aggregate demand should always be sufficient to buy all that is produced.2 On the other hand Ricardo’s theory of differential rent led him to conclude that as population grew and less fertile lands came under cultivation, food prices and therefore wages would rise, squeezing capitalist profits. Ricardo worried that since wages could not be reduced below some minimum, rising rents of landlords as ever less fertile land came under cultivation would inevitably squeeze the profits of capitalists, leaving them little incentive to continue to invest or ultimately even produce.
Marx dismissed the Malthusian theory of population as “a libel against the human race.” And by Marx’s time, the application of capitalist technologies to “large landed properties” had increased agricultural productivity per acre sufficiently to reduce concern over the problem that troubled Ricardo. Instead Marx wrote about what he believed were three different causes of crisis in capitalism: (1) He explained that unlike barter exchange among ordinary people, crises are quite possible when capitalists participate in a system of monetized exchange, and that “credit accelerates the violent eruptions”3 when they occur. (2) He reinforced Malthus’ concern about “under consumption” crises and railed against “Say’s law,” which was assumed by most other economists to prove that general gluts were impossible, or at least temporary and self-correcting, until Keynes convinced many economists otherwise six decades later. (3) He pointed out that during the expansionary phase of business cycles when the “ranks” of what he called the “reserve army of the unemployed” became depleted, workers might temporarily win increases in wage rates and “squeeze” profits. And finally, as explained last chapter, (4) Marx argued there was an “internal contradiction” inherent in the very process of capitalist development itself. Marx hypothesized that when individual capitalists adopt technological changes that improve productivity and lower production costs by equipping their workers with more “capital” to work with, capitalists collectively, and witlessly, generate a “tendency for the rate of profit to fall” because profits derive from exploiting labor, whose wages would come to comprise an ever diminishing portion of total capital outlays. In effect, Marx hypothesized that competition among capitalists to lower production costs by substituting capital for labor would drive them to collectively kill the goose that was laying their golden eggs. Since we began an analysis of TRPF theory last chapter, we consider this version of Marxist crisis theory first.4
As explained last chapter, all Sraffians are convinced that the Okishio theorem has adequately demonstrated that the TRPF hypothesis was misguided: Any CU-LS technical changes capitalists would ever adopt will not depress the rate of profit in and of themselves, but do just the opposite. On the other hand, many Marxists have searched for ways to hold onto the TRPF theory. Early attempts to rescue the TRPF theory of crisis from the Okishio theorem include Fine and Harris (1979) and Shaikh (1978a) who argued that while the actual rate of profit may rise when capitalists introduce cost-reducing CU-LS technical changes, as the Okishio theorem proves it must, that the “maximal rate of profit” (when w = 0) would fall. However, John Roemer demonstrated that while obviously true that the maximum rate must fall for CU-LS changes when the wage rate is zero, this is irrelevant because it can never prevent the actual rate from rising as long as the change is cost-reducing and any positive real wage remains constant (Roemer 1981: 116–117). Shaikh (1978b) also argued that in a model with fixed capital it was possible for CU-LS, cost-reducing technical changes to lower the rate of profit even if the real wage remained constant. Again, Roemer demonstrated that this was impossible as long as capitalists were profit maximizers by proving a generalization of the Okishio theorem in a model including fixed capital (theorem 5.1 in Roemer 1981: 121), and in an even more general vonNeuman model as well (theorem 5.2 in Roemer 1981: 128). Roemer also explained why an argument of Persky and Alberro (1978) to the effect that the profit rate might fall if new technological changes quickly render earlier investments in fixed capital obsolete, is an example of shifting the goalposts based on an implausible assumption of consistent, and therefore inexplicable, underestimation of the rate of technological change (Roemer 1981: 123–124).
By now a majority of self-declared Marxist economists have conceded that thanks to new mathematical tools unavailable to Marx which made it possible to prove the Okishio theorem, the hypothesis that CU-LS technological changes would act to reduce the rate of profit in and of itself was a red herring. The rate of profit may fall for a variety of reasons, but adoption of CU-LS technological change is simply not among the possible causes. However, there are a few Marxists (Shaikh 2016, Moseley 2017) who continue to search for empirical evidence to support the theory, and one school of obdurate Marxists who argue that a “temporal single system interpretation” (TSSI) of Marx’s value theory can rescue the TRPF at a theoretical level.5 The most outspoken representative of this small TSSI school of Marxist economists, which includes Alan Freeman and Ted McGlone, is Andrew Kliman. What he and his fellow TSSI Marxists propose is an alternative framework of analysis for interpreting what Marx was trying to communicate. In Kliman’s words:
On the standard interpretation, Marx had a simultaneist and dual-system theory: inputs and outputs are valued simultaneously, so input and output prices are necessarily equal, and there are two separate systems of values and prices. According to the temporal single-system interpretation (TSSI) of Marx’s theory, however: valuation is temporal, so input and output prices can differ, and values and prices, though quite distinct, are determined interdependently.
(Kliman 2007: 2)
Stripped of erudite sounding words, what Kliman acknowledges here is that everyone who is not of the TSSI school – those who adhere to what he calls the “standard interpretation” – assumes for purposes of analysis that whatever the going price of something is, that is the price one receives when selling it (in our context as an output), and also the price one pays when buying it (in our context as an input.) Based on this assumption, in the “standard interpretation” Marx erred in volume 3 of Capital when transforming values into prices of production – i.e. prices that yield the same rate of profit in all industries – because he inadvertently wrote down equations in which capitalists pay for inputs at one price (their “value” prices), while they sell the same goods as outputs at different prices (their “prices of production”). Although, as many Marxists who adhere to the standard interpretation have demonstrated, there is evidence that Marx himself was aware of this error but considered it to be only a technical problem, amenable to solution, which did not affect the substance of his argument. More to the point, it is easy to “correct” Marx’s equations so every good always sells for the same price in every transaction, and still calculate by how much every price of production must deviate from its labor value, as a number of Marxists who adhere to the “standard interpretation” have done.6 What Kliman asserts, however, is that Marx did not inadvertently err. He actually meant for prices of inputs, which are purchased at the beginning of the production period, to be different from prices for those same goods when they are sold as outputs at the end of the production period. We will not speculate here about whether Marx inadvertently erred, or meant for the same good to sell at a different price as an output than as an input. Instead, we will ask which assumption is more appropriate for purposes of analysis. But before doing so, notice that the TSSI conclusion that technological change can lead to a fall in the rate of profit derives immediately and solely from the assumption that the same good sells at a different price in different transactions.
Consider a capitalist who uses steel to make steel. If one assumes steel capitalists pay a higher price per ton when they buy steel as an input than they receive when they sell steel as an output, steel capitalists will obviously have a lower rate of profit than if they bought and sold steel at the same price. And if we make this price differential large enough we can make steel capitalists’ rate of profit as low as we want! More generally, if one assumes that all capitalists purchase inputs at old, pre-deflation prices, and sell outputs at new, post-deflation prices, one can make the uniform rate of profit in the economy as low as one wishes simply by assuming a sufficiently high rate of price deflation – which is all that TSSI Marxists have shown. No complicated examples and tables are necessary to illustrate this. But why should a ton of steel sell at different prices in the same analysis? Why should we reject the “law of uniform price” which has been a staple of every school of economics from time immemorial for most analytical purposes?
As we know from our analysis of technical change in chapter 3, once a new cost-reducing technology has been adopted by all in an industry, and once the prices of all goods have adjusted to re-equalize the rate of profit in all industries, the new vector of relative prices, p’, will be different from the old vector of relative prices, p. So technical change will lead to changes in relative prices, eventually. And this is the reason TSSIers give for believing that not only the price of steel, but the prices of all goods will be constantly changing.
However, prices change only after financial capital has had time to move from industries where profits are lower into industries where new technologies have lowered production costs, creating temporary “super” profits. Moreover, these new prices which take time to emerge are simply new relative prices. There is no reason to believe that when capitalists adopt new technologies which increase labor productivity7 this means that we will have either deflation or inflation of prices in general. No school of economics lists technological change as among the causes of changes in the overall price level.
But for the sake of argument, assume that new technologies are adopted which increase labor productivity (although they may not), and assume this leads to price deflation (although there is no reason to believe it would). Even if all this occurred it would not mean that capitalists would now sell their outputs for lower prices than they had paid for their inputs, which is the reason the rate of profit might fall according to the TSSI school of Marxism. Even if the new prices were lower than the old prices, the old prices are the prices back before the new technologies were introduced, and the new prices are the prices after the new technologies are introduced and capital has moved among industries to re-equalize the rate of profit everywhere. Which means that the old, higher prices are the prices capitalists paid both for inputs and were paid for outputs back under the old conditions of production. The old higher prices are not the prices capitalists pay for inputs now, in the new conditions of production. Instead, the new lower prices for inputs would be the prices capitalists now not only sell their output for, they would also be the prices capitalists pay for inputs they now buy under the new conditions of production. So even if there were deflation due to technical changes, there is no reason to believe it would lower the rate of profit.
Admittedly, there are many implicit assumptions behind all notions of equilibrium in economic theorizing, including the kind of “long period analysis” which all classical economists, including Marx, used for purposes of analysis, and which modern day Sraffians use as well. And because comparative equilibrium analysis does not capture dynamic processes, there is often good reason to supplement, or refine, comparative statics with explicit dynamics. But TSSI Marxists conjure up an implicit dynamic that is completely unrealistic as well as unprecedented: The length of time it takes for the entire price structure to adjust to the introduction of new technologies is much longer than the length of time between when capitalists typically purchase their inputs and sell their outputs. Which is why everyone except TSSI Marxists assumes for purposes of analysis that whether a good is being bought as an input or sold as an output, its price is the same.8
Having returned the TRPF theory of capitalist crisis to the dustbin of history, what about Marx’s other theories about crises in capitalism? In all other cases, if interpreted as potentials for crisis – as theories about situations that can, and do sometimes, arise, which might trigger a crisis – Marx’s discussion is highly illuminating. Not only was Marx an astute observer of where problems can arise, much of his thinking about these situations has been modeled more rigorously and developed more fully by non-Marxist heterodox schools of economic thought since. In other words, Marx did a great deal to advance a realistic assessment of capitalism as a system which cannot always be relied on to self-equilibrate, and can instead easily embark on paths which merit the label “crisis.” However, when cast as inevitable tendencies, and described as internal contradictions which are inexorable and not amenable to remedy, like the TRPF they have all proved to be misleading.
It did not take Marx long to get to the subject of money in Capital. In Chapter III of Part I in Volume I he explains how monetized exchange, where actors sell one commodity, C(1), for money, M, in order to subsequently purchase a different commodity, C(2), which has a greater use-value for them, creates the possibility of a discrepancy between supply and demand in the aggregate because the act of supplying goods, C(1)−M, is distinct, separate, and prior to the act of demanding goods, M−C(2). While presumably people who initiate the first transaction generally do so only to follow rather quickly with the second transaction, i.e. C(1)−M−C(2) will proceed in orderly fashion, Marx pointed out that it is conceivable that something might prevent some of them from doing so, in which case the demand for goods in the market would fall short of the supply. Marx was certainly correct to point out that in contrast to barter exchange, monetized exchange opens the door to potential discrepancies between demand and supply in the aggregate.
In Part II of Capital, which follows immediately and is far more quoted, Marx goes on to explain a second profound truth: Capitalists engage in exchange for an entirely different reason than increasing the use-value of commodities in their possession. Capitalists begin with money, M, with which they purchase commodities, C, only in hopes of selling commodities for an amount of money, M’, which exceeds M, i.e. M−C−M’ > M represents the logic of capitalist participation in exchange. And if capitalists have reason to believe that M’ may not exceed M sufficiently, they may well no longer purchase C, waiting for conditions to improve, and thereby disrupt the orderly process of production and consumption. Pointing out that the likelihood of self-reinforcing, disequilibrating forces in market economies is greatly increased by the motive which guides capitalists’ participation in monetized exchange, and that the more the credit system is extended the greater the danger of crisis becomes, is an important antidote to presumptions of automatic self-equilibration by financial markets.9
However, having made this important point, Marx quickly moved on to a subject more dear to his heart. After explaining in Chapter V of Part II why aggregate profits cannot be explained by capitalists “buying cheap and selling dear” from one another, he proceeded to argue in Chapter VI that in order for M’ to exceed M capitalists must find some special commodity which has the peculiar property that its use will generate a greater value than the value it contains in order for M’ to exceed M on a regular basis: {M−C(i) → C(o)−M’} where → indicates that a set of commodities capitalists purchase as inputs, C(i), are transformed during production into a different set of commodities which capitalists sell as outputs, C(o). In Marx’s words: “Our friend, Moneybags, must be so lucky as to find, within the sphere of circulation, in the market, a commodity, whose use-value possesses the peculiar property of being a source of value, whose actual consumption, therefore, is itself an embodiment of labour, and, consequently, a creation of value. The possessor of money does find on the market such a special commodity in the capacity for labour or labour-power” (Marx 1967a: 167). In other words, as explained in our previous chapter on profits, Marx believed that among all of the inputs capitalists purchase, C(i), only the input labor-power is capable of increasing the value of C(o) above the value of C(i) on a systematic basis, and therefore causing M’ to exceed M on average.
There is no reason to repeat arguments from previous chapters why Marx’s explanation of the origins of profit – based on what proves to be only a temporary, “working” assumption that all goods, including labor-power, are bought and sold according to their “exchange values” – is less compelling than the Sraffian explanation of where profits come from based on our ability to identify and measure the magnitude of a physical surplus of goods that emerges from production. But there is also no reason to fail to appreciate Marx’s argument in the first two chapters of Part II of Volume I of Capital that once exchange is monetized, once capitalists motivated by pursuit of profits participate in exchange as major actors, and once money evolves into an elaborate credit system, it would be surprising if financial crises did not erupt.
Rosa Luxemburg expounded on the under consumption version of Marxist crisis theory early in the twentieth century, interpreting imperialism as an attempt to secure more buyers through colonization to stave off stagnation from under consumption in advanced capitalist home economies (Luxemburg 1951). Later in the century Paul Sweezy became the foremost Marxist proponent of an under consumption theory of crisis. Sweezy was thoroughly conversant with previous debates about under consumption crises among Marxists, which he summarized and evaluated in Theory of Capitalist Development: Principles of Marxian Political Economy first published in 1942. In 1966, much to the dismay of many Marxists because they did not use the concept “surplus value,” but talked instead simply of an “economic surplus,” Sweezy and Paul Baran published Monopoly Capital: An Essay on the American Economic and Social Order in which they argued that modern US “monopoly capitalism” was intrinsically prone to stagnation for lack of sufficient aggregate demand, and that a great deal of US government behavior, in particular high military spending accelerating the Cold War, could be understood as attempts to counteract this tendency toward ever greater stagnation.
Among Marxist economists debates often rage over how to interpret Marx on any subject, which aspects of Marx’s many theories deserve more or less emphasis, and of relevance here, which of Marx’s different theories of capitalist crisis is most compelling in some particular historical period. Sweezy and Baran are widely considered to be the most important Marxists who argue that stagnation, or what was initially referred to as a crisis of over-production or under consumption, is the most insightful and relevant of Marx’s crisis theories for modern capitalism.
Much of Baran and Sweezy’s analysis of post-WWII US capitalism was indeed insightful. What was not helpful was the impression they left readers that stagnation is inevitable and unavoidable as capitalism advances. What more careful modeling has since demonstrated is that no matter how much wage growth may lag increases in labor productivity, and no matter how much capitalists’ propensity to consume out of a growing share of national income may fall, the sub-optimal growth path this yields can continue indefinitely. In other words, the presumption of an inevitable breakdown is unwarranted. As discussed below, more rigorous modeling by economists beginning with Steve Marglin (1981) demonstrated that there are a number of possible growth trajectories depending on values of key variables, and as undesirable and unnecessary as “profit-led” regimes like those projected by Baran and Sweezy may be, they do not lead to the kind of total breakdown under consumptionist Marxists project. They simply lead to lower rates of capital utilization, and consequently lower levels of production and income than is technically possible.10
Marx initially presented this argument as a theory of how profits might be squeezed by rising wages during the expansionary phase of business cycles when unemployment was reduced and worker bargaining power was temporarily increased. He also cited it as a reason why capitalists might favor capital-using, labor-saving technical change to replenish the ranks of the reserve army of the unemployed, relieve pressure to increase wages, and revive profits. In the 1970s a number of radical economists (including me) scoured data over a number of business cycles in the US for empirical evidence that a profit squeeze, or an under consumption, or a TRPF theory of crisis was more consistent with the historical record.11 Both Marxian and Sraffian economists later developed a long-run version of a profit squeeze theory of crisis as well. And indeed this is one area where there is a great deal of agreement between Marxian and Sraffian economists today: While the expansionary phase of a business cycle can reduce unemployment and increase labor’s bargaining power in the short-run, it is also possible that the bargaining power of labor might increase in the long-run as well: Labor-friendly governments, higher levels of unionization, more ample social welfare programs, or simply an increase in worker consciousness might squeeze profits independent of short-run fluctuations in unemployment. A staple of Sraffian theory is that all points on an economy’s wage-profit frontier are technically possible; and that where on the curve any economy will end up is largely determined by precisely the kind of social/political conditions analysts study as potential causes of a squeeze on profits that last longer than the expansion of a typical business cycle. Some Sraffians and Marxians have extended this logic to how an increase in the rents of natural resource owners might squeeze profits just as a secular rise in wages can. Which brings us to our final school of Marxian crisis theory.
Terrence McDonough describes Social Structure of Accumulation (SSA) theory as follows:
Social Structure of Accumulation (SSA) theory is a theory of stages of capitalism. Capitalist stage theory focuses on periods intermediate in length between a short-run business cycle and overall capitalist history. These periods consist of a long period of relatively stable capitalist accumulation followed by a relatively long period of crisis and breakdown. Each of the periods of accumulation is underpinned by a set of institutions designated as an SSA.
(McDonough, chapter 34: 370 in McDonough et al. 2014)
The SSA school arose in the United States in the 1970s as what was sometimes called the post-WWII “golden age of capitalism” came to an end, and the “stagflation crisis” emerged in many of the advanced economies.12 The original authors, who self-identified at the time as Marxists, included Sam Bowles, David Gordon, Thomas Weisskopf, Michael Reich, and Rick Edwards. They were later joined by others including David Kotz, Terrence McDonough, Martin Wolfson, and Phillip O’Hara. Over the past four decades some of the original authors ceased to self-identify as Marxist, and some of the newcomers self-identified as institutionalist or post-Keynesian. Nonetheless, SSA came to be known as a new Marxian approach to crisis theory that differed from older Marxian schools of thought, as explained by McDonough:
Marxist theories of capitalist crisis had tended to locate crisis in fundamental tendencies of the capitalist economy which were always potentially present. These tendencies included the tendency of the rate of profit to fall, disproportionalities among economic sectors and a tendency for either overproduction or underconsumption. Thus the emergence of crisis would be the present expression of these long-run secular tendencies. The new theories that arose in the 1970s and early 1980s in the wake of the stagflationary crisis did not share the same emphasis on these tendencies. Crises could arise due to the breakdown of the institutional framework which conditioned the previous period of capitalist expansion…. This argument defined recurring crisis periods as more serious than downward fluctuations of the ordinary business cycle, but not as the expression of an ultimate crisis of capitalism.
(McDonough, chapter 34: 371 in McDonough et al. 2014)
Thus, SSA theory is less deterministic than earlier Marxist crisis theories, and in that regard shares more in common with post-Keynesian, neo-Kaleckian, and structuralist heterodox schools of macroeconomics, as discussed below. Many of its authors have provided insightful analyses of different periods of economic history identifying problems which became more severe over decades, and eventually gave rise to new institutional structures, or SSAs, which once again facilitated a revival of profits and accumulation.13 However, one can question if beyond a great deal of insightful historical analysis – in this case written by economists with greater expertise in economic theory than historians often have – there is an underlying theory in the usual sense of the word. Two key “assertions” which can be interpreted as the basis of SSA “theory” are: (1) If anything comes to threaten the ability of capitalists to receive an acceptable rate of profit, there will eventually be crisis, and a search for some institutional way to revive profits will be launched. And (2) if anything threatens a healthy rate of accumulation (presumably of capital stock), there will be crisis, and a search for some way to revive accumulation will ensue.14
In the case of a low rate of profit it is obvious why a crisis would eventually arise. Moreover, we can identify an important economic agent – capitalists – who would press for changes, or a “new SSA” to rectify what is clearly a problem for them. The second proposition may at first seem equally obvious, and seems to be considered equally obvious by most SSA theorists. But is this really the case? What if a high rate of profit were somehow maintained, but accumulation stagnated?15 Would this trigger a crisis? And who would propel the search for a new SSA to revive accumulation? One can speculate that when accumulation stagnates output may stagnate as well, which means little improvement in living standards for workers, who might press for changes. Or, one can speculate that if accumulation stagnates in one capitalist country that country will lose out in competition – either economic or military – with other capitalist countries where accumulation is more robust. In which case nationalist or military forces might press for a new SSA to revive accumulation. But unlike the case of low profits, the connection between stagnant accumulation and both a crisis trigger and some powerful agents to propel a search for a new SSA is less obvious.
When Marx wrote economists made no distinction between microeconomic and macroeconomic theory. But in the aftermath of the Great Depression and the “Keynesian revolution,” economists agreed to a rough division of labor: Microeconomic theory was to focus on how relative prices of different goods and services, and the real values of various distributive variables are determined in private enterprise, market economies. Whereas macroeconomic theory was to focus on determinants of overall production and growth of output, overall employment, and price inflation or deflation – including how the monetary system might affect real, and not merely nominal, values of variables. Microeconomists were also far more likely to conduct their analysis under the assumption that actors had “perfect information” and markets were successfully equilibrate, whereas some macroeconomists focused on disequilibrating forces that might prove self-reinforcing and the problem of unknowable uncertainties about the future. In this context, post-WWII crisis theory became part of macroeconomics. And in this context it would seem that Sraffian economics, which is firmly rooted in heterodox microeconomic theory, has no counterpart to Marxian theories of crisis.16
However, this is not entirely the case. If we consider economists such as Michal Kalecki, Joan Robinson, Nicholas Kaldor, Don Harris, Steve Marglin, Lance Taylor, Paul Davidson, Hyman Minsky, Amitava Dutt, Tom Palley, Wynne Godley, Sydney Weintraub, Michael Hudson, Jan Kregel, Geoff Harcourt, Marc Lavoie, Alfred Eichner, Randall Wray, Paul Auerbach, Peter Skott, and Robert Blecker, we have a heterodox school of macroeconomics intellectually allied and linked with Sraffian microeconomics which we can compare to Marxian crisis theory. The important link between Sraffian micro theory and these heterodox macroeconomists, who call themselves post-Keynesian, neo-Kaleckian,17 and “structuralists,” is that they both focus on production of a physical surplus and its distribution, and seek to elaborate theoretical models where definitive conclusions can be deduced from assumptions about the values of key parameters. The division of labor between these heterodox macro theorists and Sraffian micro theorists is based on (1) whether output is broken down into different goods and services or aggregated, i.e. whether analysis is conducted in a multi-good or single good framework, (2) whether analysis focuses on a long-run steady state, or on various possible growth paths, (3) whether uncertainty is a major consideration, and (4) whether money is treated as only a numeraire, or the credit system plays an important role in the analysis. Since this book is about Marx and Sraffa, only a brief description of various schools of heterodox macroeconomics which are allied with Sraffian heterodox microeconomic theory is offered here.
The most important modern elaboration of Marx’s early warnings that money and credit contain the seeds of crisis is Hyman Minsky’s “financial instability hypothesis” (Minsky 1986, 1992). Minsky provides a set of plausible hypotheses about the likely behavior of banks and borrowers and explains how they logically lead to a step by step increase in systemic financial fragility. While he makes no attempt to predict when a “Minsky moment” will arrive, he emphasizes that competent regulation of the financial sector is necessary to avoid such an outcome. In short, post-Keynesian monetary theory and theories of financial crisis are very much in tune with Marx’s early warnings, while providing more rigorous analysis.
According to “Say’s Law” in the aggregate production, or supply, creates its own demand. According to what we might call Keynes’ “macro law of supply and demand” in the aggregate production, or supply, will follow demand if it can. These two “opposite” visions, perspectives, or hypotheses if you will, give rise to immense differences between competing macroeconomic theories today, and in particular between post-Keynesian and “new classical” theories – the name given to rational expectations, neoclassical, macroeconomic theories. In particular heterodox macro theorists insist on the efficacy of fiscal and monetary stimulus to increase aggregate demand temporarily when necessary, while new classical theorists, like their pre-Keynesian classical counterparts, once again claim, against all evidence, that such efforts are pointless. But as mentioned above, Marx provided a perfectly sound rebuttal to Say’s Law long before Keynes was born. So once again, we find that Marx, Keynes, and post-Keynesians are on the same page regarding the possibility of crises due to a temporary lack of sufficient aggregate demand. The difference is that Keynesians explain how fiscal and monetary policy can, and should, be used to address short-run deficiencies of aggregate demand, and post-Keynesians demonstrate how wage-led regimes are both possible and preferable to profit-led regimes in the long-run. In effect, Keynesians and post-Keynesians demonstrate how it is possible to save capitalism from itself, while as we have seen, under consumption Marxists insist that there is a long-run stagnation problem that will eventually prove immune to stop-gap measures to ward it off.
Which leads us to the most important difference between Marxian and post-Keynesian, neo-Kaleckian, and structuralist theories: They come to different conclusions about what economic growth can look like in the long-run. Non-Marxist theories of “distribution and growth”: (1) provide more analytically rigorous explanations for different possible growth paths than Marx and his followers; (2) make clear that depending on the values of key variables, both “wage-led” and “profit-led” regimes are possible; (3) emphasize that while failure to achieve full-capacity growth is possible, this need not lead to a total breakdown; (4) establish that there are no “inherent, internal contradictions” that will inevitably bring any capitalist system crashing to an end; and finally, and perhaps most importantly, (5) demonstrate how, when crises do occur, there are various government policy options which can, if used appropriately, “save the day.” In contrast, Marxist crisis theories have traditionally projected inevitable collapse, and even social structure of accumulation theory projects the necessity of major institutional changes when any historic structure of accumulation reaches a crisis point.
We have seen how every attempt to salvage TRPF crisis theory from the dustbin of history has proved futile. What has been suggested here – although admittedly not demonstrated – is that theoretically rigorous neo-Kaleckian models of distribution and growth now demonstrate that while stagnation due to low capacity utilization is one possibility, (1) it is not the only possibility, and (2) even this trajectory is economically “sustainable” and need not lead capitalism to “self-destruct.” And what has also been suggested is that one of the two key assumptions made by SSA theory, namely that if accumulation slows sufficiently something must give, has been lacking sufficient supporting argument. None of which is to say that completely understandable and justifiable dissatisfaction and disgust with a profit-led, low-capacity utilization, under performing capitalist regime may not lead workers to replace it with a far superior wage-led, full-capacity, capitalist regime – or better yet, with an entirely different and superior system of participatory, ecological socialism.
1 Scholars attribute the phrase to Thomas Carlyle who described political economy as “dreary, desolate, and indeed quite abject… what we might call… the dismal science.” Irrespective of what “dismal” result Carlyle was actually complaining about in this remark, the phrase was soon interpreted as referring to the fact that many political economists predicted dire outcomes.
2 What Ricardo and other devotees of what came to be known as “Say’s Law” failed to consider is the possibility that even if all savings are successfully loaned out, if they are used to purchase assets rather than newly produced investment goods, supply may well fail to “create its own demand” in the aggregate.
3 Marx 1967c: 441. See Chapter XXVII, “The Role of Credit in Capitalist Production.”
4 See Weisskopf (1992) for an excellent exposition of Marx’s different theories of crisis and their modern champions.
5 See Kliman 1996, 1997, 2001, and 2007, Kliman and McGlone 1988 and 1999, and Kliman and Potts 2015.
6 It is even possible to demonstrate that if one interprets Marx’s own transformation (Marx 1867c: 163–164) as simply the first step in an iterative process of updating input prices to conform to the previous output prices until there is no longer any discrepancy, one can achieve a “correct” transformation of values into prices of production. See Shaikh 1977.
7 And as we have seen, just because a new technology is adopted does not mean it necessarily increases labor productivity. When r > 0 capitalists sometimes commit “sins of commission” and adopt new technologies which lower labor productivity.
8 It is also possible that TSSI Marxists simply do not understand the role that the law of uniform price plays in economic analysis, and therefore do not understand when it is appropriate. Of course in the real world the same good often sells for different prices in different transactions which take place at the same time. But what it means to have a market for a good, or say that a market is “well ordered,” is that arbitrage will reduce real world price discrepancies between different transactions for a given good which take place very close to the same time. In any case, many have published compelling critiques of the TSSI school. For examples see Laibman 2000, Skillman 2001, Mongiovi 2002, Mohun 2003, and Veneziani 2004.
9 When Capital was published the vast majority of economists were still convinced by Ricardo’s arguments that Say’s Law holds. Having failed to heed Marx’s argument to the contrary, it took the economics profession another sixty years – and countless recessions and several major depressions – to finally discover the error of their ways with the help of Keynes’ General Theory of Employment, Interest and Money and the Great Depression of the 1930s.
10 To see that these are the predictable consequences of repressing wages and increasing capitalists’ propensity to save in a formal political economy growth model see Hahnel 2014a: 251–259. Theoretical advances in growth theory by heterodox neo-Kaleckians and structuralists during the last decades of the twentieth century, as discussed below, made it possible to determine what the logical consequences of tendencies like those above really are.
11 See Glyn and Sutcliff 1972, Boddy and Crotty 1974, Weisskopf 1979, and Hahnel and Sherman 1982.
12 At roughly the same time a related school of Marxism arose in France. The founding document of the “French Regulation school” was Michel Aglietta’s A Theory of Capitalist Regulation published in 1979.
13 Classic SSA publications include: Bowles et al. 1983, 1986, and 1989, Gordon et al. 1982, Kotz et al. 1994, McDonough et al. 2010 and 2014, and Kotz 2015.
14 McDonough, chapter 34: 372 in McDonough et al. 2014.
15 If one responds that this is impossible – that high profits cannot be maintained absent robust accumulation – then SSA theory has only one proposition: healthy rates of profit must be maintained, or a search for a new SSA to revive them will ensue. But one might argue that this is obvious to pretty much anyone, and hardly qualifies as a “theory.”
16 Although Sraffa and Luigi Pasinetti are frequently mentioned as among the founders of the “European” school of post-Keynesian economics.
17 While Keynes and his General Theory was the inspiration for many who call themselves post-Keynesians, Michal Kalecki and his Theory of Economic Dynamics was the inspiration for other heterodox macroeconomists often called neo-Kaleckians.