Harvey versus Marx on Capitalism’s Crises Part 2: Getting Profitability Wrong

by Andrew Kliman

The evidence is clear: from the end of WWII to the Great Recession U.S. corporations' overall rate of profit fell, and this fall is almost entirely accounted for by Marx's law of the tendential fall in the rate of profit.

First published: 12 March, 2015 | Category: Economy, Philosophy and Theory

The first part of this article focused on David Harvey’s interpretation (Harvey 2014) of Marx’s law of the tendential fall in the rate of profit (LTFRP) as capitalist production increases—a law Marx identified as ‘the most important law of political economy’. It remains to address Harvey’s belief that the LTFRP has not in fact been operative since the 1980s. 

The labour-force data discussed in part 1 constitute Harvey’s only ‘evidence’ that the LTFRP has not been operative; he provides no direct evidence regarding the rate of profit (i.e. the amount of profit as a percentage of the volume of invested capital). However, he does challenge the evidence that has been put forward by myself and others which indicates that the rate of profit fell throughout the 1980s and 1990s (see Figure 1).1

Figure 1. U.S. Corporations' Rate of Profit

U.S. Corporations' Rate of Profit

Harvey says that ‘some serious questions have to be asked’ about this evidence. He is quite right, and he asks crucially important questions. The problem is that his discussion proceeds as if his questions are ones that we have never heard nor taken into account. They are actually long-standing, standard questions. I for one have taken all of them into account when gathering and interpreting my data. There is therefore no need to respond to them; my analyses and interpretations of the data have already anticipated and dealt with them. I simply need to make clear how they have done so.2

What Harvey calls his ‘most important objection’ to ‘much of the falling-rate-of-profit literature’––once again, he is vague about the object of his critique––is the fact that ‘[p]rofit rates can fall for any number of reasons’. Therefore ‘[d]ata that show a falling rate of profit do not necessarily confirm the existence of the specific mechanism to which Marx appealed’ (labour-saving technical change). 

This is exactly right. Thus, when I considered the trajectory of U.S. corporations’ rate of profit from the end of World War II to the Great Recession and I concluded that ‘Marx’s law of the tendential fall in the rate of profit fits the facts remarkably well’ in this case, (Kliman 2012, p. 137) my conclusion was not based on the mere fact that the corporations’ rate of profit fell. It was based on a ‘decomposition analysis’ that separates out (decomposes) various potential causes of the fall and measures the effect that each one had on the rate of profit. Moreover, because the standard way of decomposing the rate of profit is not particularly appropriate when conducting a causal analysis (for a reason stressed by Harvey), I decomposed it in a different manner. 

Traditionally, the rate of profit has been decomposed into the rate of surplus-value (or ratio of profit to employee compensation) and a function of the value composition of capital (or ratio of the constant to the variable components of the capital-value advanced (invested)). This is fine in some contexts, but the nominal value composition of capital that researchers construct is different from the value composition to which Marx refers. It is affected not only by the relative amounts of value invested to acquire means of production and hire workers, but also by changes in the rate at which commodities’ money prices rise in relation to the commodities’ actual values. Because two different factors affect it, movements in the nominal value composition have no clear-cut, unambiguous meaning. For example, when the nominal value composition remains constant, as it did in the U.S. during the 1960s and 1970s, we cannot conclude that the relative amounts of value invested to acquire means of production and hire workers also remained constant. It is possible that relatively more value went to acquire means of production, which tends to raise the value composition, but that this effect was offset by accelerating inflation.3 As Harvey correctly emphasises, this is a ‘major problem’. 

My alternative decomposition dealt with this problem by separating the two determinants. I decomposed the overall movements in the rate of profit into 

  1. movements caused by changes in the rate at which commodities’ money prices rise in relation to the commodities’ actual values;
  2. movements caused by changes in the ratio of profit to employee compensation, and
  3. movements caused by ‘everything else’. 

I found that, although the first two causes had been important during certain shorter periods, neither of them had a substantial effect on the rate of profit in the long term—that is, when we consider the postwar era as a whole. Almost the entire long-term fall in the rate of profit was therefore caused by changes in ‘everything else’. 

But once (1) and (2) have been set aside, it follows mathematically that ‘everything else’ is just the ratio of employment to the amount of capital invested in fixed assets, as measured in terms of labour time. Almost all of the long-term fall in the rate of profit is attributable to the decline in this ratio. In other words, it is attributable to the fact that employment consistently grew less rapidly than capital accumulated. This is precisely how Marx’s law explains the long-term tendency of the rate of profit to fall. Thus, the law accounts for almost all of the fall in U.S. corporations’ rate of profit. 

The ratio of profit to employee compensation had little effect on the rate of profit because it changed very little. (It fell a bit during the early part of the postwar period, but had no upward or downward trend between 1970 and the Great Recession.) It is important to emphasise that the long-term stability of this ratio is not a statistical mirage caused by the fact that the U.S. government classifies the pay of CEOs and other top corporate executives as employee compensation rather than as profit. Recent estimates of mine (see Kliman 2014b) indicate that re-classification of top executives’ pay as profit makes very little difference. Yes, their pay skyrocketed in recent decades, but there were simply too few top executives for this to have had much effect on the numbers. Between 1979 and 2005, the rising share of the product received by managers in ‘the 0.1%’ and ‘the 1%’ (the top 0.1 percent and top 1 percent of the income distribution) depressed other business-sector employees’ share by only 0.4 and 0.6 percentage points, respectively, according to my estimates. 

Another of Harvey’s objections to the falling-rate-of-profit evidence is that ‘[t]here is a gap between where profit (value [sic]) is produced and where it may be realised. … The patterns of … flows of capital and revenues are intricate and it is not clear that data collected at one point in the system accurately represent the movements in their totality’. He is correct once again. It would be wrong to conclude from the data discussed above––which pertain to the profitability of domestic capital investment––that there was a decline in U.S. corporations’ overall rate of profit, on foreign as well as domestic investment. My conclusion that the rate of profit fell was instead based upon consideration of both the foreign and the domestic accounts. Government data on U.S. corporations’ capital investments abroad and their profits from investment abroad are available from the start of 1983 onward. They indicate clearly that U.S. multinational corporations’ rate of profit on foreign investment trended downward substantially between the start of this period and the Great Recession (see Figure 2)4. Because the denominators of the domestic and foreign rates of profit measure somewhat different things, the two data sets cannot properly be combined, and thus we cannot ascertain the exact extent to which U.S. corporations’ overall rate of profit fell. However, the fact that both the foreign and domestic rates of profit declined does mean that we can be confident that the overall rate of profit did indeed fall.

Figure 2. U.S. Multinational Corporations' Rate of Profit Abroad

(after-tax income from foreign direct investment as a percentage of accumulated foreign direct investment)

U.S. Multinational Corporations' Rate of Profit Abroad

Harvey also notes correctly that U.S. multinationals use ‘transfer pricing’ to shift profits generated in one country onto the books of a subsidiary in a different country where they are not taxed or are taxed at a lower rate. He could have added the fact that multinationals’ foreign profit and investment data are attributed to the countries in which their foreign subsidiaries are incorporated, which frequently differ from the countries where production takes place and the products are sold. As a result, it is difficult if not impossible to know what the multinationals’ rate of profit in any particular country really was. But this does not matter insofar as the total picture is concerned. Transfer pricing schemes allow corporations to shift around profits and titles to investments, but they do not affect the total volume of profit or investment. Harvey claims that transfer pricing allows profits to be ‘disguise[d]’, but he provides no evidence and I know of no such evidence. Shielding profits from the reach of tax authorities is not the same thing as disguising them. 

The evidence discussed above pertains to U.S. corporations only. Harvey objects that ‘it cannot be taken as evidence of what is happening to global capital’. Indeed it cannot, but what makes this fact a legitimate objection in this context? The topic of his paper is the fall in the rate of profit as a potential cause of economic crisis, and he is well aware that the latest crisis began in the U.S. before it spread throughout the world ‘through contagions in a global financial system’. Since the U.S. was the epicentre of the crisis, and its subsequent spread elsewhere has a straightforward financial explanation, what we need to focus on is whether and how a fall in the U.S. rate of profit, not the global rate, was an underlying cause of the crisis. 

Harvey notes that profitability has rebounded substantially in recent years. Of course, post-recession trends have no bearing on whether a prior fall in the rate of profit was among the causes of the Great Recession. His point is rather that, since a measure of the rate of profit which fails to capture the post-recession rebound is suspect, what that measure tells us about a prior decline in profitability is suspect as well. I could not agree more, but all of the rates of profit I have computed (using more or less inclusive definitions of profit) do indeed capture the post-recession rebound. They all fell during the Great Recession—bottoming out 24 percent to 38 percent below their peak values of 2006––but by 2013, all of them had rebounded to levels close to or greater than those of 2006. The main cause of the rebound in profitability has been a sharp post-recession decline in workers’ share of the product, which in turn has been caused by companies producing more without increasing their workforces. It has not been caused by ‘wage repression’. Even after adjustment for inflation, employees’ hourly compensation has risen.

Right the first time (on Marx’s ‘underconsumptionism’)

Why does Harvey repeatedly stress that there are ‘conflicting forces’ and ‘multiple contradictions and crisis tendencies’? Why are we presented with the mono-causal LTFRP strawman? Notice what these two things imply when we take them together: since the law can hold true only if other causes of crisis and counteracting factors are assumed not to exist, we must jettison the law once we recognise that they do exist. Thus, as I noted in part 1, I suspect the talk of multi-causality is masking Harvey’s desire for an apousa-causal crisis theory. He is clearly not happy with the specific multi-causal theory of crisis that emerges, when all is said and done, from volume 3 of Capital––a theory in which the LTFRP remains intact and other determinants such as the financial system are linked to it and mediate the way in which it appears. 

In particular, Harvey seems to want to impute to Marx an underconsumptionist theory of crisis––that is, a theory in which the lack of ‘effective demand’ is not the mediated result of the operation of the LTFRP and intermediate links such as businesses’ investment decisions and financial disturbances, but an independent, unrelated phenomenon produced by the masses’ restricted consumption. He attributes to Marx the notion that ‘if wages are too low[,] then lack of effective demand will pose a problem’. As evidence, he quotes a sentence in volume 3 of Capital (and a similar footnote in volume 2) in which Marx states that ‘the ultimate reason for all real crises always remains the poverty and restricted consumption of the masses’. (Marx 1991a, p. 615) 

In the time-honoured fashion of underconsumptionists everywhere, Harvey strips away the context in which this sentence appears. When read in context, the sentence has nothing to do with periods in which low wages supposedly lead to inadequate demand, nor does the masses’ restricted consumption seem to be a ‘cause’ of crises in the modern sense of the word ‘cause’ (which Aristotle called ‘efficient cause’). It is merely the condition that makes crises possible (an Aristotelian ‘formal cause’), not something that turns this possibility into a reality.5

Only a few years ago, Harvey had a much clearer understanding of the sentence and the passage in which it appears, and he took care to analyse it in context. After asking where the extra demand comes from that enables the surplus-value that has been produced to be realised in money form, Harvey (2012, p. 25) noted that ‘Marx’s answer is as surprising as it is ruthlessly honest. In a two-class closed society comprised of capitalists and labourers, there can be only one source of the extra demand and that is from capital, since exploited labour could never furnish it’. In other words, it is capitalist firms’ demand for additional means of production––investment demand––and capitalist households’ demand for consumer goods that allows the portion of output that contains the surplus-value to be sold. Harvey then quoted from and summarised much of the passage in question, in order to make clear that the shortfall in demand that characterises economic crises is not due to the restricted consumption of ‘the masses’ or ‘exploited labour’, since their consumption is always restricted––crisis or no crisis. Thus, blaming the crisis on the masses’ restricted consumption is like blaming an airplane crash on gravity (which always exists, crash or no crash). 

The shortfall in demand is instead caused by the fact that the extra demand that needs to come ‘from capital’ has temporarily stopped coming from capital. The key problem is that capitalism requires what Harvey (2012, p. 26) calls ‘continuous capital-accumulation’, i.e. additional investment in production, but a shortfall in demand occurs when and because the volume of additional productive investment is less than what is needed.  

Since Harvey knows (or at least used to know) all this, why has he suddenly taken the ‘restricted consumption’ sentence out of context and pressed it into the service of an underconsumptionist ‘wage repression’ theory of crisis that is alien to Marx’s actual and ‘ruthlessly honest’ account of the demand problem? Perhaps the answer is that the ‘ruthlessly honest’ account takes us straight back to the fall in the rate of profit. Once we understand that a lack of demand is almost always a matter of inadequate investment demand, we are led to ask why investment is inadequate, and this question leads to two further ones:  Has the volume of profit (surplus-value) that has been generated large enough to fund an adequate level of investment demand? And is the expected rate of profit on the new investments of today high enough to bring forth investment in the volume that is needed?

Inadequate profitability was a main cause of the long-term slowdown in U.S. corporations’ investment demand for productive fixed assets. Between 1948 and 2007, their rate of accumulation of fixed assets fell by 41%, while their after-tax rate of profit on fixed-asset investment fell by 43%. The only other factor that can influence the rate of accumulation is the share of profits that is re-invested in production; it actually rose a bit (3%). The entire decline in the rate of productive capital accumulation is therefore attributable to the decline in the rate of profit (see Kliman and Williams 2014 for further discussion). 

Expectations that the future profitability of productive investment will be inadequate also seems to have been a major problem for some time now, as well as a crucial determinant of why recovery from the Great Recession has taken so long and has been so weak. In order to explain the feebleness of the recovery, mainstream economists and economics writers such as Paul Krugman, Martin Wolf, and former U.S. Treasury Secretary Lawrence Summers suggest that the U.S. economy entered a period of ‘secular stagnation’ some time before the recession, perhaps as early as the mid-1980s—that is, a period in which adequate demand is no longer sustainable unless real (i.e., inflation-adjusted) short-term interest rates are ridiculously low, maybe as low as –2% or –3%. This means that borrowers pay back less than they borrowed, once inflation is taken into account. If the only way to induce companies to undertake sufficient investment is to provide them with money that they don’t have to pay back, the expected rate of profit on new investments must be terribly low indeed! (See Kliman 2014a for further discussion.)

I am not ‘proclaiming that it is all a consequence of some hidden tendency for the rate of profit to fall’, as Harvey puts it. That would be wrong for two reasons. First, all manner of intermediate links and complicating factors have also been at work. (The Great Recession has also weakened businesses’ confidence in the future, to take just one example.) Second, the tendency for the rate of profit to fall is not ‘hidden’. As Hegel said, essence must appear, or shine forth, in the observable world. I think it has done so. 

Andrew Kliman is the author of The Failure of Capitalist Production: Underlying Causes of the Great Recession (Pluto Press, 2011) and Reclaiming Marx’s ‘Capital’: A Refutation of the Myth of Inconsistency (Lexington Books, 2007).  A professor emeritus of economics at Pace University (New York), he works politically with the Marxist-Humanist Initiative.





1 The data used to construct Figure 1 come from the U.S. Bureau of Economic Analysis: National Income and Products Accounts Table 1.14, lines 1, 4, 7, 9, 10, and 12; Fixed Asset Table 6.3, line 2; and Fixed Asset Table 6.6, line 2. Net operating surplus and after-tax profit are measures of profit. The denominator of both rates is accumulated investment in fixed assets, net of depreciation. Depreciation is valued at historical cost.

2 I shall limit these remarks to a discussion of my own analyses, since I am less knowledgeable about others’.

3 As Harvey notes, even the ‘real’ value composition of capital––the one to which Marx refers rather than the nominal one––is not purely an index of labour-saving technical change. In this respect, it differs from the ‘technical’ and ‘organic’ compositions. However, my estimates indicate that U.S. corporations’ real value composition tracked the technical and organic compositions quite closely. Between 1947 and 2007, the real value composition increased by about 120% while the technical and organic compositions increased by about 160%. Throughout almost all of this period, the relation between the different compositions of capital was even stronger than these numbers suggest. The difference in growth rates is largely due to brief periods (the latter parts of the 1960s and 1990s) in which exceptionally rapid wage growth temporarily depressed the real value composition.

4 The data used to construct Figure 2 come from the U.S. Bureau of Economic Analysis’ ‘Balance of payments and direct investment position data’ tables. The numerator of the rate of profit is ‘Direct Investment Income Without Current-Cost Adjustment’; the denominator is ‘U.S. Direct Investment Position Abroad on a Historical-Cost Basis’. The data are for ‘all countries’.

5 For further analysis of the passage, see pp. 165–67 of Kliman (2012). For discussion of the logical and empirical flaws of the underconsumptionist theory of crisis, see chap. 8 of the book.

All comments are moderated, and should be respectful of other voices in the discussion. Comments may be edited or deleted at the moderator's discretion.

Remember my personal information

Notify me of follow-up comments?