Monopoly Capital had an outstanding impact on students of my generation. It was published just as the Vietnam War was heating up, when students and youth throughout the world were beginning to “contest the structures”—to use a favorite expression of that time—and were eagerly looking for analyses of these “structures.” Monopoly Capital, written jointly by two renowned Marxist economists, each of whom had already authored a classic, provided just that. It was avidly read in progressive circles around the world.
For students of economics like myself, there was an additional reason for its impact. Economic literature from both the Communist world and from Communist writers in the West tended to underplay the problem of aggregate demand. While Marx had been a trenchant critic of Say’s Law and had highlighted the demand problem, it was by this time seen at most as a problem underlying periodic crises, but not one that could affect capitalism in a secular sense, ex ante. Though written decades earlier, Nikolai Bukharin’s criticism of Rosa Luxemburg, that she was talking (in an ex ante sense) not of temporary general overproduction (which was perfectly possible) but of permanent general overproduction (which was not), captured a widely prevalent Marxist view of the time.1 Since the problem of aggregate demand affected “realization,” and since this related to the “sphere of circulation,” giving it any centrality in an analysis of capitalism amounted to privileging the “sphere of circulation,” which was anathema to traditional Marxist theory. The “sphere of production” had to be accorded centrality.
There was also a second, analytical reason why aggregate demand continued to be neglected. While Marx had discussed at length the possibility of “overproduction,” he had not analyzed what actually happened, in the sense of where exactly the economy settled, when such “overproduction” occurred. Put differently, Marx had explained why ex ante overproduction could occur, but had given no clue as to what would happen ex post when it did occur. In other words, there was for a long time in Marxian economics no study of a single period “equilibrium” (in the sense of a state of rest), let alone of multi-period dynamics, in a system subject to deficiency of aggregate demand, such as Kalecki’s or Keynes’s multiplier analysis was to provide. Having recognized “overproduction,” the discussion just moved on to describing all that might possibly follow, instead of studying how the system might actually behave from one period to the next. And even when Keynes and Kalecki had filled in this gap, many refused to accept their analysis, despite the latter being a Marxist.
This was a great pity, because, as Sweezy had shown in his 1942 Theory of Capitalist Development, Marx had already anticipated the Keynesian theory of why capitalist economies faced problems of aggregate demand—i.e., why Say’s Law was wrong—almost three-quarters of a century before Keynes. But that insight remained unused within the Marxist tradition itself, let alone beyond it.
For me and my contemporaries, this posed a problem. We considered ourselves Marxists, but the exciting debates in our discipline that were occurring at the time around the Kaleckian-Keynesian Revolution were being ignored within “our” tradition. Fortunately a group of outstanding Marxist economists like Kalecki, Steindl, Lange, and Baran and Sweezy, working in the West, wrote with remarkable rigor, incorporating the insights of the Keynesian Revolution into the Marxist theoretical tradition, while also providing deep critical analyses (of which Kalecki’s 1943 article was an obvious example) of how the Keynesian medicine would not be enough to stabilize capitalism.2 They provided, in a sense, the new impetus to Marxian economics that all of us had been waiting for. Sweezy’s Theory of Capitalist Development in the 1940s and Baran’s Political Economy of Growth in the 1950s had already set the trend. Their joint work Monopoly Capital carried that trend forward. For us it was a remarkable breath of fresh air. The book discussed an amazing range of economic and social issues; in what follows, however, I shall concentrate only on some economic issues.
Baran and Sweezy’s Monopoly Capital was founded on two basic theoretical propositions: first, that there is a tendency under monopoly capitalism for real wages not to rise in tandem with labor productivity, which in turn causes the share of surplus in output to increase over time; and second, that capitalists’ own consumption and investment have no such tendency to increase as a proportion of output, certainly not at the rate that surplus does. All other things being equal, therefore, the producible surplus at any given level of capacity utilization cannot be realized over time; and since only as much surplus would be produced as can be realized, there is an ex ante tendency for the degree of capacity utilization to decline—which, because it would further reduce investment, would unleash a cumulative tendency towards stagnation. Such a cumulative tendency towards stagnation has not been visible, Baran and Sweezy argued, because the “sales effort” and growing state expenditure, especially military expenditure—both of which serve to absorb the surplus—have increased relative to output over time. While the increase of the former is a “spontaneous tendency” under monopoly capitalism, in the sense that competition between oligopoly firms takes the form not of cutting prices but of heavy advertising and sales effort, the increase in the latter is a deliberate act of state intervention to stabilize the system.
The most curious aspect of Monopoly Capital from a theoretical point of view, however, was that while establishing the first proposition, it did not advance a theory of wage determination. Much has been written of depth and substance on the issue of the relationship between wages and the value of labor power under monopoly capitalism. But I have in mind something simple: if the problem of realization assumes primary importance, then the reason must lie in the fact that the wage-bill of productive workers, much of which is presumably consumed, declines over time as a proportion of the total output that they themselves produce. Why should that happen?
The argument advanced in the book went as follows. Monopoly capitalism is characterized by incessant efforts to cut costs, above all by economizing on the use of labor. Collusive pricing, however, ensures that these reductions in costs are never “passed on” in the form of lower prices, so that the tendency is for the profit margin—and hence by implication the share of surplus in aggregate output—to increase over time.
The obvious problem with this argument is that it lacked a theory of money wage determination. If money wages rose in tandem with labor productivity, then there was no reason for unit labor cost to fall, in which case, even with no change in the final price, the share of surplus would not rise. Baran and Sweezy probably had in mind, though they did not make it explicit, the Kaleckian proposition of an increase in the profit mark-up over time because of a rise in the “degree of monopoly”; but then we are really saying that monopoly capitalism entails a rise in the share of surplus not because it is monopoly capitalism per se, but because it necessarily entails a rise in the degree of monopoly (a proposition not insisted upon by Kalecki). With an increasing mark-up, arising from the changing structure of industry or from changing collusive pricing behavior, if labor is the only current input, then there would be a rise in the share of surplus in national income over time, no matter how money wages and labor productivity behave.
But when the unit prime cost over which the mark-up is applied also includes raw material costs, and when these raw material producers are largely located abroad, a rise in the degree of monopoly does not necessarily entail a rise in the share of the surplus in national income. In fact, Kalecki’s empirical data showed a secular stability in the share of wages (and hence, by implication, of surplus) in the national income in the pre-Second World War period, and he explained this by saying that the effect of the rise in the “degree of monopoly” which would, ceteris paribus, have been to depress the share of wages was offset by a decline in raw material prices relative to the unit labor cost. (This explanation accords well with the Prebisch-Singer finding that the pre-war terms of trade between primary commodities and manufactures were secularly deteriorating for the former).
But even if the share of wages did not decline, while the share of surplus went up at the expense of the raw material producers in the gross value of output (inclusive of the value of raw materials) as distinct from the national income, the tendency towards stagnation would still be there. This is because a shift from raw material producers, which include numerous peasants and petty producers, to the capitalists, would still lower the level of aggregate demand because of the higher marginal propensity to consume (to use a Keynesian term) of the former group.
Even though Baran and Sweezy left their argument somewhat understated, their postulate about a rising share of surplus and its implications for aggregate demand was generally accepted, because it was eminently plausible. Indeed, a host of other writers, including Kalecki (for whom the surplus would rise relative to the gross value of output including raw material costs) and Steindl, had argued along the same lines.
It is the second theoretical proposition of the book that was quite novel. The argument that capitalists’ consumption as a proportion of output would not rise at all, and certainly not as fast as surplus, would be readily accepted. Likewise, the argument that investment, insofar as it responds to “endogenous stimuli,” cannot possibly rise as fast as surplus when the latter itself is rising as a proportion of output, can be scarcely contested. (By “endogenous stimuli” here I mean the growth of the market that comes about because growth has been occurring and everybody anticipates past growth to continue. The stimulus for investment that arises from within the system itself—because of the operation of the multiplier-accelerator or any other similar mechanism—constitutes an example of an “endogenous stimulus.”)
But the general view among those economists who incorporated the issue of aggregate demand into growth theory was that in addition to endogenous stimuli, capitalism also has exogenous stimuli. Of these, external markets, or encroachments into pre-capitalist markets, were an obvious example, which Rosa Luxemburg had highlighted; state expenditure was another. But in addition to these, it was contended that “innovations” constitute an exogenous stimulus that is “spontaneously” available to the system. A steady stream of innovations gives rise to investment over and above what endogenous stimuli alone would generate, i.e., over and above what would occur if investment was governed only by the anticipated growth of the market (the acceleration principle).
Now, if such a “spontaneous” exogenous stimulus is available, then there is no reason why the consequent investment may not rise significantly enough in conditions of monopoly capitalism to offset the stagnationist implications of a rise in the share of economic surplus. It may not of course happen, but Baran and Sweezy’s conclusion then cannot necessarily be drawn. It becomes essential for the authors to look at the impact of the exogenous stimulus provided by innovations on the level of investment in conditions of monopoly capitalism.
Basing themselves on earlier work by Joan Robinson, Josef Steindl, and Paolo Sylos-Labini, they put forward the novel proposition that, leaving aside “epoch-making innovations” like railways and automobiles, the normal run of innovations do not constitute an exogenous stimulus at all in conditions of monopoly capitalism.3 Whatever investment would have occurred anyway because of the endogenous stimulus takes the form of new processes and new products, but their availability does not stimulate any higher investment over and above what the endogenous stimulus generates.
Steindl had put forward this idea in his Maturity and Stagnation in American Capitalism, but had not specifically linked it to monopoly capitalism. Baran and Sweezy did so. They explained it carefully and provided substantial empirical support for this proposition. Their argument, if I can restate it in my own way, was simple: any additional investment that a new process, say, generates beyond what the anticipated growth of the market dictates, would necessarily entail snatching a bit of the rivals’ market. But each of the rival oligopolists is powerful, and neither price-cutting (which would be ruinous for all, as others would follow suit and a price war would ensue) nor extra sales effort (which can be matched by others, including those who do not have access to the new process) can snatch markets away from rivals in the short run. Even those with access to the new process would not undertake any extra investment over and above what the anticipated growth of their own market would warrant—i.e., what the endogenous stimulus alone would dictate. What oligopolists with access to a new process will do, therefore, is incorporate this new process into the investment that they were planning to make anyway.
It may be thought that there is an additional channel through which a new process would cause some extra investment, namely displacing some of the existing capital stock of the oligopolist firm that embodies an old process. Such a substitution would entail an increase in gross investment but no increase in net investment. It would, however, generate some additional demand for the system as a whole, because new capital stock would have to be produced within the system to enable such extra gross investment to occur.
This argument, however, lacks substance. For it to hold, the difference in unit prime costs between the two processes must be so large that the rate of profit on the new capital stock (installed to replace some existing stock) arising for this reason alone is sufficient to cover the interest cost and risk premium associated with such installation. This being a tall order, the oligopolist would normally continue using the existing capital stock for its full physical life (or, more accurately, until the cost of maintenance and repair, which depends on age and not on technological progress, and which increases suddenly rather than steadily, becomes exorbitant).
While the oligopolist firm will therefore be content with introducing the new process only by incorporating it into the investment that would have been undertaken in each period anyway, and hence also continue the old processes, it would at the same time not let go of the new process for use by anyone else. Baran and Sweezy drew attention to a remarkable feature of the U.S. economy that arises because of this: on the one hand, U.S. corporations spend enormous amounts on research and development; but on the other hand, old processes continue to be used in production.
What this meant from the macroeconomic point of view was that the effect of the “spontaneous” exogenous stimulus on investment, far from being greater under monopoly capitalism than earlier and thereby making it possible to offset the stagnationist consequences of a rise in the share of the surplus, was actually less. Even as the share of surplus was rising, the effect of the exogenous stimulus in generating investment was waning under monopoly capitalism, thus compounding the stagnationist tendency.
Epoch-making innovations, according to Baran and Sweezy, belonged to a different category altogether, since in their case the nature of the innovation itself was such that a new market for the product would be created, at the expense of rivals’ older products, even without any price cuts or extra sales effort, and this would encourage investment in such innovations over and above what would otherwise have been forthcoming. Such innovations constituted, in other words, a genuine exogenous stimulus.
While this argument is theoretically persuasive, there is an intriguing issue here. In one of the most exciting chapters of the book, “On the History of Monopoly Capitalism,” Baran and Sweezy discuss the effect of the spread of automobiles in the United States, but what is not clear is why, during the Great Depression, the spread of automobiles, and hence of investment expenditure in the automobile sector, could not revive the economy. The United States, as is well-known, sank once more into a recession in 1937 after a brief recovery and finally came out of the Great Depression only with the increase in military expenditure in preparation for the Second World War. Why did the impact of the automobile revolution not lift the economy out of the Great Depression? Is it the case, perhaps, that even the impact of epoch-making innovations is not independent of the state of demand in the economy, in which case even such innovations cease to be effective against the tendency toward stagnation? Baran and Sweezy, in other words, might well have overestimated the effect of epoch-making innovations, which too, instead of being authentic exogenous stimuli, should perhaps be considered only quasi-endogenous at best.
In any case, because for these reasons capitalist consumption and investment could not increase as a proportion of national income while the magnitude of economic surplus did, a problem of realization and an associated ex ante tendency towards stagnation characterized monopoly capitalism. It did not manifest itself ex post because of increasing expenditure on the “sales effort” and because of increasing state expenditure. And Monopoly Capital contained an interesting Statistical Appendix by Joseph D. Phillips estimating the economic surplus as a proportion of gross domestic product of the United States and the mode of its realization.
A criticism has often been levelled against the book that even if the share of surplus is shown statistically (i.e., ex post) to be increasing, this fact per se does not establish the causation advanced by the authors. For instance, if the state taxes workers and spends the proceeds, then we would have an increase in the share of the surplus in output matched by an increase in the share of state expenditure in output. But from this ex post observation it would be wrong to deduce any ex ante tendency toward overproduction and stagnation. In other words, the Statistical Appendix, conforming to the theory, does not in any way establish its validity.
This is no doubt true. From ex post data we cannot infer ex ante tendencies. Ex ante tendencies have to be independently established. But from an independently established theoretical tendency for a rise over time in the “degree of monopoly” under monopoly capitalism, we can certainly place greater confidence in the validity of their theory because of the statistical estimate of a rise in the share of economic surplus in output.
The problem with the argument of Monopoly Capital lay not in what it said, but in what it did not say. While establishing that monopoly capitalism, afflicted by a tendency towards a rising economic surplus relative to output, could offset this tendency through growing state expenditure, especially military expenditure, the book did not discuss the contradictions in this arrangement that could disrupt the functioning of the system and therefore bring about a change in it. The book shifted the critique of capitalism onto a moral-ethical plane, away from any economic contradictions that might afflict it, thereby implicitly suggesting that the system had successfully manipulated its economic contradictions to a point where they were no longer threatening to its stability. The system no doubt had overcome its economic contradictions in a manner that was inhumane, dangerous, and threatening to humankind, but it had done so nevertheless. And what is more, because capitalism had successfully manipulated its economic contradictions, challenges to it could now be expected only in the “outlying regions” like Vietnam, where its militarism had inflicted a horrendous war upon a hapless people. By shifting the critique of capitalism from an economic to a moral plane, Monopoly Capital also shifted the location of the challenge to the hegemony of capitalism to the periphery.
But this was the overwhelming perception at the time. Philosopher Herbert Marcuse had written along the same lines. With the Vietnam War on one side and the relative economic stability achieved in metropolitan countries on the other, the period of what has been called the “golden age of capitalism” contributed to this perception. Monopoly Capital showed that underlying this so-called golden age was massive U.S. military expenditure (later writers have often talked of “military Keynesianism”). What it did not do, no matter how correct its position on the location of the theatre of revolutionary action might have been at the time, was to see the contradictions immanent in this very manipulation of domestic contradictions by monopoly capitalism.
The contrast between Monopoly Capital and Paul Baran’s The Political Economy of Growth in this respect is quite striking. Having put forward a similar argument about U.S. military expenditure underlying high levels of activity both in the United States and elsewhere in the earlier book, Baran had gone on to contend that such expenditure, financed by fiscal deficits, created a liquidity overhang in the economy that could come crashing down at any time in the form of inflation. For this he had been taken to task by Robinson, who accused him of dragging in the Quantity Theory of Money. But Baran was not arguing that a larger money supply (or larger liquidity) caused higher prices, i.e., he was not, in Robinson’s language, reading the quantity equation (MV=PQ) “from the left to the right,” which is what the Keynesian Revolution had opposed; he was talking about an increase in private wealth, in the form of claims on the government, which could change form and be held as commodities (or claims on commodities) at the hint of an inflation, and hence exacerbate any inflationary episode that might arise. His focus, in short, was on asset preference, and he could not be accused of slipping into monetarism. Indeed, the Keynesian objection to the monetarists’ reading the quantity equation “from the left to the right” had precisely been that they did not consider asset preference.
Curiously, however, in Monopoly Capital, no argument of this kind figured at all. This may be because the “model” underlying the book presumed that state expenditure was financed by taxes on surplus, and hence had an expansionary impact through the “balanced budget multiplier.” Quite possibly, the authors were induced to presume this in response to Nicholas Kaldor’s earlier criticism in his review of Baran’s book that the observed share of post-tax profits had not gone up under monopoly capitalism, and that this questioned the basic proposition about rising surplus.4
But the point remains that spending on the maintenance of a string of military bases around the globe had inter alia given rise to a current account deficit for the United States, which had only widened, along with the fiscal deficit, during the Vietnam War. Under the Bretton Woods system, where the dollar was ordained to be “as good as gold” (with $35 the price of an ounce of gold), this spending could be financed by printing dollars, and therefore caused no immediate concern. But it brought three problems in its train.
First, the huge outpouring of dollars contributed to the creation of the Eurodollar market, and so greatly enhanced international liquidity that the regime of capital controls that had existed under the Bretton Woods system had to be abandoned, under pressure from finance capital wishing to go “global,” in search of financial investment opportunities all over the world. The genesis of the regime of globalization that came later, where fiscal deficits were to be curtailed in accordance with the caprices of finance capital—including even in the United States, which has no “fiscal responsibility” legislation—and hence Keynesian “demand management” could no longer be undertaken, lay in this massive outpouring of dollars from the United States.
Second, the inflation that Baran had feared actually materialized, as increased expenditure on the Vietnam War put excess demand pressure on prices. True, the real inflationary upsurge that occurred in 1968 was caused by a worldwide wage explosion—that is, by cost-push factors—but underlying this sudden increase in wage-demands was the fact that inflation caused by excess demand had been eroding real wages for some time. Perhaps even in the absence of any such erosion, demands for higher wages would have surfaced at some point, since capitalism cannot do without a substantial reserve army of labor to keep money wage demands in check. But the point here is that Monopoly Capital did not enter into the issue of inflation, i.e., whether the overcoming of ex ante overproduction would bring inflation into the picture, unlike what Baran had argued earlier.
Third, with the outpouring of dollars and the inflation that ensued, wealth-holders around the world ceased to be content to hold dollars. Its “good as gold” status could no longer be sustained. While President De Gaulle of France took the lead in this shift from dollars to gold, his doing so was not some gratuitous act of vindictiveness; it reflected ground realities. President Nixon snapped the gold-dollar link in 1971, and the Bretton Woods system collapsed soon thereafter.
This entire series of developments, which only showed that the system had not manipulated all its contradictions, was unanticipated in Monopoly Capital. This is not to blame the authors, for no book can cover everything. It is just to point out the limits to the scope of the book. Interestingly, Magdoff and Sweezy were to write some of the most incisive pieces on the “financialization” that occurred in the period following the publication of Monopoly Capital, pieces that were collected together in their book Stagnation and the Financial Explosion.5 But this entire set of issues did not figure in Monopoly Capital.
Likewise, because the emphasis in Monopoly Capital was on the question of the realization of a growing surplus, the international setting, within which the U.S. economy (that was their focus of attention) is located, is discussed primarily from this perspective. Harry Magdoff was to write The Age of Imperialism as a sort of complementary work to Monopoly Capital, but the latter work itself does not foreground the major economic ramifications of imperialism. And Magdoff’s own analysis, though the core of his argument, relating to metropolitan capitalism’s abiding need for raw materials, remains valid, predates the onset of globalization and the changes it effected.
The strength of Monopoly Capital lay in its incorporation of the insights of the Keynesian Revolution into a Marxist understanding. Its weakness lay in the fact that it did not go further. It did not anticipate a post-Keynesian era, but confined itself to providing a strong, though essentially moral, critique of the manner in which Keynesianism was being implemented.
Paradoxically, however, the basic argument of Monopoly Capital acquires greater relevance in the era of globalization; and, what is more, the ability of the system, now seen as a global system, to manipulate its contradictions, is now absent, which is why the current capitalist crisis has been continuing for so long. Precisely because capital—not just capital-as-finance but also capital-in-production—is globally mobile, the world is no longer segmented, and workers in advanced countries are drawn into competition with lower-paid workers of the third world countries. Their wages, now subject to the drag exercised by the massive third world labor reserves, do not increase, even as labor productivity increases. The wages of workers in the third world likewise do not increase, notwithstanding the shift of several manufacturing and service sector activities from the advanced countries to their soil, because their massive labor reserves remain unexhausted. Hence the vector of world real wages does not increase, even as the vector of world labor productivities does, giving rise to the phenomenon of an increasing share of surplus in world output.
What Monopoly Capital had visualized is coming true with a vengeance in the era of globalization for the world economy as a whole, and also within particular national economies, and for reasons that are so obvious and compelling that no complex theory of wage determination is required. And what is more, no state intervention to offset this ex ante stagnationist tendency owing to the rising proportion of surplus in world output is possible in today’s world, unlike what the authors of Monopoly Capital had visualized.
This is because, first, no world-state exists that can possibly counter this stagnationist tendency at the global level. Second, individual nation-states cannot act to pull their own economies out of crisis, because in the era of globalization they cannot run fiscal deficits or tax the surplus without risking capital flight (unless they impose capital controls, which would mean leaving the vortex of globalized finance altogether, and hence would require an alternative class alliance seizing power). Third, the United States, which can undertake larger state expenditure to revive activity in the world economy as a whole—i.e., which can act as a surrogate world-state, because the dollar still enjoys a status as the most favored currency among the world’s wealth-holders—is hamstrung by the fact that it is still only a nation-state. Since the demand-generating effects of any increase in U.S. state expenditure are likely substantially to leak out abroad in the absence of U.S. protectionism (anathema, because it would undermine neoliberalism), such an increase in expenditure would create employment abroad while increasing U.S. debt. This is unacceptable to the United States as a nation-state. And finally, as to the possibility of a group of advanced nations coming together to undertake a coordinated fiscal expansion, as Keynes and many others had suggested during the 1930s Depression: not only are the logistics of such coordination difficult, but finance capital would also oppose it. Such opposition, as Charles Kindleberger has suggested, was the reason the proposal was not seriously considered in the 1930s; today the opposition of international finance capital would be even more powerful.6
This is why, unless some new “bubble” arises, of which there are no signs, and which too would inevitably collapse, precipitating a new crisis, the world economy will continue to be mired in stagnation. The basic framework of Monopoly Capital helps us to understand this predicament better than any other book written since.
- ↩See Nikolai Bukharin, “Imperialism and the Accumulation of Capital,” in Kenneth J. Tarbuck, ed., Imperialism and the Accumulation of Capital (London: Allen Lane, 1972).
- ↩Michał Kalecki, “Political Aspects of Full Employment,” originally written in 1943 and reprinted in Selected Essays on the Dynamics of the Capitalist Economy 1933–1970 (Cambridge, UK: Cambridge University Press, 1971).
- ↩Joan Robinson, The Accumulation of Capital (London: Palgrave Macmillan, 2013); Josef Steindl, Maturity and Stagnation in American Capitalism (New York: Monthly Review Press, 1976); Paolo Sylos-Labini, Oligopoly and Technical Progress (Cambridge, MA: Harvard University Press, 1962).
- ↩Nicholas Kaldor, review of Paul Baran, The Political Economy of Growth, American Economic Review 48, no.1 (1958): 164–70.
- ↩Harry Magdoff and Paul M. Sweezy, Stagnation and the Financial Explosion (New York: Monthly Review Press, 1987).
- ↩Charles P. Kindleberger, The World in Depression 1929-1939 (Berkeley, CA: University of California Press, 1986).