Michael Roberts on US Profit Rates: A Critique and an Alternative View

Michael Roberts has emerged as one of the leading Marxist analysts of current economic developments.  For many of us, his blog, The Next Recession, has become an indispensable and challenging resource. He has also recently published his second book, The Long Depression, a lively summary of his current research which also extends the arguments of his blog in some important and interesting ways.  For example, detailed accounts of the Long Depression of the late 19th century and the Great Depression of the 1930s.

Some of Michael’s recent posts show him at his formidable best.  To take just one of many examples, on 9 Oct 2016, a discussion of the anxieties being expressed in recent reports by bodies like the IMF and BIS about the growing problem of high debt levels.  In a brief, but notably clear and well documented account, Michael surveys some of the main vulnerabilities in the system today – the fragility of corporate balance-sheets in many emerging economies, the dangerous levels of non-performing loans which threaten leading banks in major economies such as China, Germany and Italy, the general failure of monetary policy and zero interest rates to generate adequate levels of investment and growth in production, productivity and trade.  Here also there are calculations by Michael which show that that the latest policy nostrum being advocated by the IMF and in many other quarters – i.e. big increases in state spending on infrastructure – can’t realistically be nearly large enough to revive a serious increase in overall rates of growth.

Michael believes that there is now, ‘the prospect of a new global slump on a fast approaching horizon’.  I think it is quite possible that he will be proved right about this.

One of the signal virtues of Michael’s work that he is committed to norms of scientific practice and tries wherever possible to back up his analysis and conclusions with evidence.  He is resourceful in ferreting out relevant data from official statistical sources, and in finding vivid ways of presenting his results graphically. In addition, Michael is willing to make his working spread-sheets available to other researchers.  Earlier this year when I was working on profit trends in the US economy I took up a general invitation which Michael had made in a footnote – and he immediately sent me a pile of relevant work sheets and explained detail in a covering letter. Many thanks for this.

I have been able to assess more closely a dimension of Michael’s work which I believe is open to question.  A central theme in much of his analysis is that it is a declining rate of profit which is the underlying cause of the sequence of crisis which has afflicted the major industrial economies since the late 1990s.  As the biggest by far of the industrial economies, the US has been the main battleground in the debate about this thesis.

Using the same definition of the profit rate as does Michael, I have checked his spread-sheets against the original sources in the US official statistics, and redone the calculations.  I get the same results – as did a Swedish researcher, Anders Axelsson,  who had made an earlier check of  the material.

However, there is a problem.  I do not think that Michael’s data about US profit rates actually support some of the conclusions he draws about trends over the past 20 years.

In his Post of 4 Oct 2016, Michael once again summarises the findings of the analysis of US profit rates from 1946-2014 which he published in December 2015.  Here in Figure 1 is the graph which he published in his December post in support of his conclusions.  Note that the definition of profits used in Figure 1, is the same as the one chosen by Andrew Kliman in his book-length study of US profit rates, The Failure of Capitalist Production (2011 pp.99-101). Profits of the financial as well as the non-financial sector are included in the corporate total. All figures are for the US domestic economy only – overseas investment, production and profitability of US firms are excluded. (Kliman p.75).


From the data in Figure 1, Michael draws four conclusions in his Post of 4 Oct 2016 [  ].

  • The secular decline in the US rate of profit since 1945 is confirmed … the US corporate profit rate is some 30 per cent below where it was after World War 2 and 20 per cent below the 60s’. This is clearly correct, but we should be careful about how we use the term ‘secular’. If a direct trend line is drawn between the 1950s and 2015 it certainly slopes downward.  But Michael is sometimes assumed by his less critical readers to be saying that there has been an continuous fall in US rates of profit over the past 60 years.  However the trend line which he has calculated in Table 1 shows a cyclical pattern – two periods of fall, followed by two periods of recovery, though on a dampened scale. The post-2000 recovery is especially significant given that it preceded, in 2008, the most severe crisis of the post-war period. The causality here has to be more complex and dialectical and needs to focus on cyclical patterns, not a unilinear and continuous trend.

Michael next argues that:

  • Profitability … peaked in the late 1990s after the neoliberal recovery. Since then the US rate of profit has been static or falling’.
  • Since about 2010-12, profitability has started to fall again’.
  • Finally, the fall in the rate of profit in the US economy has now given way to a fall in the mass of profits’.

To test the latter three propositions we need to look more closely at rates of profit in the past 20 years. In Figure 2 I have used Michael’s own data for this period with only two changes: adding in the 2015 figures which have since been published, and a revision upwards of his 2014 rate of profit (because the figure for Gross Value Added in that year has since been raised in the on-line data source by about $90 billion.

Apart from these, Figures 2, 3 and 4 are based on Michael’s own spreadsheet (No. 9)


Source: Profit = net GVA = Gross Value Added MINUS Annual Depreciation MINUS Employee Compensation.

Corporate sector (Financial and Non-financial Companies. Domestic Economy only)

GVA Domestic Corporate Business – BEA NIPA Table 1.14, line 1

Employee compensation – BEA NIPA Table 1.14, line 4.

Fixed asset annual depreciation (Historical Cost)  – BEA Fixed Assets Table 4.6, line 17.


On my reading, the data in Figure 2 do not support conclusions (2) and (3).

  • Since the late 1990s the rate of profit has not been ‘static or falling’. The rate was higher in 2005 and 2006 than in the 1997-8 peak. It seems unlikely that that the financial crisis which started in 2007, and went critical in 2008  was caused by a profit rate in 2007 at the same level as the late 1990s. The 2008 crisis was far more severe in its impact on jobs, wages, growth and trade than the dotcom downturn of around 2000.  Yet the impact on profitability was considerably briefer and more limited.  Profitability fell in 2008 and after as a result of the financial crisis rather than as its cause.
  • Profitability did not ‘start to fall after 2010-12’. After rising sharply from the 2010 low point, profit rates stayed level in 2013, rose fractionally in 2014, and only in 2015 was there the beginnings of a fall.
  • Michael also holds that a drop in the rate of profit is normally followed by a rise in the mass of profit which is only a temporary phase. Figure 3 shows that this was not the case in the recent period.


Source: As for Table 2.  Gross Domestic Income – BEA NIPA Table 1.10, line 1.

In Figure 3 the mass of profit is measured as a proportion of Gross Domestic Income to eliminate the effect of inflation. When a comparison is made with Figure 2, it is evident that the two downturns in the rate of profit were not followed by a rise in the mass of profit which was only temporary.  The mass of profit recovered more quickly from the downturn of 2008, and that rise was sustained right through to 2014.  The rate and mass of profit track each other quite closely until 2006 when both rates were exceptionally high.  But the recovery after 2008 was more rapid for the mass than for the rate, and was not temporary, but continued through until the 2015 drop.


An alternative explanation

That the mass of profit has been relatively high in the last 10 years gives some support to the broader argument I have been developing that in this period the system has been contending, not with falling rates of realised profit, but rather with an excess of profit relative to the levels of investment which have been lagging.

I have spelt out the arguments in an article which is in the latest issue of HM (24.3).  Also I have added further evidence in a series of posts on my website. See especially the posts of May 15 and April 26 2016.

The system is being wracked and distorted by the malignant consequences of the effectiveness of neoliberal profit raising counter-tendencies. Profit rates have been driven up, and investment constricted, by a potent combination of market forces, aggressive campaigns by capital to raise the rate of exploitation, financialisation, state policies, and a deep change in the mode of regulation of the corporate sector (shareholder value etc.).  The operation of these forces has generated a global surplus of capital in the money form which is too large to be completely recycled back into productive investment.  Thus what we have is not a crisis of Keynesian lack of consumer demand, nor a Monthly Review crisis of monopoly profits.  But instead, a crisis of a particular sort of disproportionality – between available accumulations of money capital and the capacity of the system to absorb them.

Official statistics are not the only source of support for this thesis.  See, for example, an authoritative study by the Toronto based McKinsey Global Institute Global Competition (2015). The data in this research covered about 28,000 large firms from 42 countries  (i.e. companies with turnover equivalent to over $200 million per annum). MGI found that: since 1980 corporate cash holdings have ballooned to 10 per cent of GDP in the US, 22 per cent in Western Europe, 34 per cent in South Korea and 47 per cent in Japan.

Corporate accumulation of cash reserves is only one source of an overall excess of liquid capital in the system.  Other major channels are: (1) the investible capital piling up in the global economy as the numbers and average wealth of the ultra-rich continue to rise; (2) international current account imbalances. In the pre-2008 period the US current-account deficit was a huge 5.6 per cent of GDP in 2006; China’s current account surplus was 10 per cent of GDP in 2007 – the counterpart flows of capital from China for lending and investment in the US were thus enormous.

The tendency of the rate of profit to fall, which Marx correctly identified, has been reversed in the recent period by the strength of a range of countertendencies. The rate of exploitation has been driven up, the turnover of capital has accelerated, the expansion of labour-intensive service sectors has slowed the rise in the organic composition of capital.

Investment levels in the major economies have lagged in money terms, as the value and price of investment goods has fallen in relative terms.  Corporate strategies in the productive and financial sectors have shifted over the past three decades to defensive and aggressive operations in the mergers and acquisitions market for corporate control, and to the maintenance of high share-prices.  These objectives require large war-chests of money capital, and a careful rationing of investment expenditure. Corporate tax evasion has soared, based on the holding of profits in money form, and laundered through tax havens rather than reinvested in production. Generous payments to executives and shareholders have also been a priority in surplus-value allocation.

From these various sources, from around 2000 a  rising surge of surplus loanable capital was being transferred into the banks and financial markets to be lent out and invested by them. The consequences were contradictory.  Large profits, initially, in the financial sector, but then increasing difficulty in finding a large enough supply of safe assets and reliable borrowers.  Disaster hit in 2007 after more than $1 trillion dollars had been lent in unsustainable subprime mortgages in the US, and securities based on these mortgages had been sold on a huge scale to banks in the US and Europe.  Severe strains also arose in Europe because banks in the Northern countries had lent lavishly to finance unsustainable booms in the peripheral economies of the EU.

And since then, a seemingly intractable combination of ultra-low interest rates, stressed banking systems, demand deficiency, faltering growth in key sectors of the world economy – and, since mid-2015, indications that profit rates may be starting to fall. Governments and markets have been testing new ways of coping with the problem of excess money capital.  The  patterns of stress in the system have altered.  Large sums of money have been absorbed by the major banks in the reconstruction of their balance-sheets. Leverage ratios have been driven down and their reserves in the central banks hugely increased.  In the short run this is a safer (if much less profitable) way of dealing with surplus liquidity in the financial circuits than handing it out in subprime mortgages or as loans to Spanish property developers.  But the crisis then takes the form of stagnating growth in investment and trade.

Since 2008 one of the three channels through which excess money capital is being transferred into the financial system has diminished. The US current-account deficit fell from 5.8 per cent of GDP in 2006 to 2.7 per cent in 2014  However, the numbers and net worth of the global wealthy keep rising.  And the continuing build-up of corporate cash piles remains a further and potent source of excess money capital in the system.

We have to remember the scale of the trends being summarised in the US in Figures 1-3 above.  For example, from its low point in 2008, the mass of profit in the US (using Michael’s definition) rose from about $2 trillion to over $3 trillion. This is a very sizable increase, given that Gross Domestic Income in the US was just over $18 trillion in 2015.  Since investment levels were relatively static in this period the result was a rapid rate of accumulation of corporate cash reserves. Notoriously, much of this flow is booked through tax havens to evade US taxation of profits.

The question of tax evasion leads us back to the definition of profits which Michael and Andrew Kliman use in their work. Here I stress that none of the ways in which Marxist researchers define profit rates and use official statistics are entirely satisfactory. Mine own included.  However if we want to do science rather than peddle myths we have to use the data available, but, obviously, with critical care and caution about their limitations. In all the sciences, research at the edge of knowledge is endemically plagued with problems and controversies about the meaning and validity of the data being used.

Marx defined the rate of profit as follows: s/ C + v.  I.e. surplus-value divided by capital advanced (constant capital + wages). But this can be construed in two ways:  either (1) as the capital advanced and surplus value extracted by companies – the corporate rate of profit; or, (2) a social rate of profit, which Michael calls a whole economy rate of profit.  The latter is a measure of the surplus generated within the whole economy in a given year, after deducting, (a) the amount of productive capital consumed in the private capitalist sector and, (b) total wages of all employees (not just those employed in the capitalist sector).

In the work which is summarised in Figures 1 to 3 above, and in common with Kliman, Michael used the corporate definition of profit.  He now prefers, he says, a whole economy rate of profit and this is what he employs in his latest book, The Long Depression.  I’ll discuss this, and the problems it poses, another time. But Figure 4 shows much the same pattern as the corporate profit rates in Figures 2 and 3 – a higher rate in 2005-6 than in the late 1990s, a sustained, if unspectacular, recovery after 2008, and a small drop in 2015.


Gross Domestic Income NIPA, Table 1.10, line 1.

Employee Compensation  NIPA Table 1.10, line 2.

Consumption of Fixed Capital  Fixed Assets, Table  1.10, line 23.

Private Non-Residential Fixed Assets  –  Table  4.3, line 1. (hist cost).

These differing definitions raise questions about Michael’s claim that the cause of a fall in the rate of profit is a rise in the organic composition of capital which is faster than any rise which has taken place in the rate of exploitation.  This is certainly a crucial mechanism.  But it does seem unlikely that in the cyclical variations in recent profit rates, a rise in the organic composition of capital plays a significant role.  Marx himself saw the organic composition of capital as changing over longer  periods of time, not as the cause of short-run movements in the business cycle.

It needs emphasise that both the corporate and the whole economy rate of profit in US official statistics have one large limitation. They cover the domestic economy only. The source for corporate value added figure is: NIPA Table 1.14 Gross Value Added of Domestic Corporate Business; for the whole economy rate of profit the source is NIPA Table 1.10 Gross Domestic Income by type of income. See for more discussion, my Post of April 13.

The Methodological Handbook for the US National Accounts explains that:

Domestic profits include all profits made by companies from operations in the United States as a geographical territory, irrespective of the nationality of the company. (i.e. where its headquarters are located)  Crucially, as the Handbook explains, “The profits component of domestic income excludes the income earned abroad by US corporations”. (Section 13 – 5).

We need not assume that all profits booked through tax havens are necessarily missing from the profits figures in the National Accounts: there are many loopholes which allow profits to escape taxation and to be reported in the corporate tax returns and company accounts which the Washington statisticians rely on.  But the balance of probability is that the exclusion of foreign-sourced profits from the National Accounts must mean a large underestimate of actual profit rates in the current period.  Thus it is by no means certain that US profit rates have in fact been 20 per cent lower than in the 1960s. For example, in the business press, in recent years, it is usually assumed by journalists, without any question, that profit rates in countries like the US have been at their highest in the post-war period, both before the downturn of 2008, and in the recovery since 2010.

There are of course other problems with the concept of domestic profits as used in the National Accounts of the US and other high-income economies.  As John Smith and Tony Norfield have explained in their recent and valuable books, much of the surplus-value which appears to be generated in the domestic economy is derived from the exploitation of labour in the low-wage economies. Here there are fundamental questions to be further explored.

 Some conclusions

Michael Roberts’ overall argument has many dimensions.  He acknowledges that different crises are sparked by different triggers. Due recognition is given, for example, to financialisation, and the instabilities generated by increasingly levels of debt in the major economies.  There are interesting sections on these aspects of the crisis in his book on The Long Depression  But in his work there is an constant theme – namely that the crucial underlying cause of the crises of the post-2000 period is that the rate of profit peaked in 1997 and has not recovered since.  Behind this is a logically questionable assumption, that if crises are recurrent (even though different in form) there must be a single and common cause.

I have shown above, that as Michael’s own empirical work makes evident, there has not been, over the past 20 years, a simple linear fall in the US rate of profit.  Rather what we see are cyclical patterns of oscillation.  Falling rate of profit tendencies are battling it out against counter-tendencies, with complex results which have to be explained dialectically and not by looking for a single unilnear cause.

We should always be searching for causality of course.  But capitalism is a complex adaptive system.  The contradictions as they evolve (‘find room to move’ in Marx’s phrase) change the immediate configuration of the system. The tendency of profit rates to fall is not in itself a contradiction.  Michael’s own work on cycles (see the chapter in his Long Depression book on cycles within cycles) has taken promising directions, and resumes themes explored by some of the great Marxist economists of the interwar period (e.g. Preobrazhensky, Maksakovsky) If the research programme of cycles within cycles is to advance, as I hope it will, its creative implementation will require the sort of exceptional statistical and analytical skills which Michael possesses.


3 thoughts on “Michael Roberts on US Profit Rates: A Critique and an Alternative View”

  1. Just a few points.
    First, there is very serious problem with the valuation of the fixed capital stock, discussed at some length by Anwar Shaikh in his book. Basically due to the influence of neo-classical economics the value of the fixed capital stock is its expected revenue or interest rate, so as profit rates rise so does the value of the fixed capital stock. Consequently the rate of profit as measured by neo-classical national income estimates never actually changes. This is probably the most realistic explanation in my opinion of why it has been more or less flat for the last forty years. This is exacerbated by the shift to chained measures or the geometric mean rather than the Laspreyes or Paashe indices traditionally used, it means that it is virtually impossible to correct for the assumptions of the statistical agencies.

    Second the is no estimate of turnover by Michael Roberts so his estimate of the rate of surplus value is wildly inaccurate.

    Third his estimates of the world rate of profit are even worse. In China for example he estimates a rate of profit for 1978 when China was a centrally planned economy with no value production. His later estimates are little better as they use the PPP estimates of the Penn world tables, these are physical not value series.

    Fourth as you point out $2 trillion of profits are held abroad by US MNCs so even if none of the above applied, this is enough on its own to totally skew the figures.

    Fifth put all this together and its clear that the estimates of the ROP can be classified on a scale between arbitrary and pointless. Its why they have virtually no relationship to the actual movement of the economy of class struggle.

    Six if there was a capitalist crisis why is there no class struggle anywhere in the world?

    Seven he says they’ll be a recession every year. He’s bound to be right sooner or later.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s