A note on Goldman Sachs and the rate of profit

Issue: 124
Posted: 1 October 09

Joseph Choonara

A recent paper produced by Goldman Sachs (GS) claims to shed new light on the trajectory of the global economy in the run-up to the current crisis.1 As well as discussing the “global savings glut”,2 the paper presents new research on the “return on physical capital”. The latter is of interest to Marxists because it bears some resemblance to the “rate of profit”, and Marx famously placed the tendency of this rate to fall at the centre of his analysis of capitalism and crisis.3 For Marx this ratio between profits and investment is crucial to the dynamic of capitalism because it governs the potential rate of expansion of capital.

The GS paper claims that “far from declining…the global return on capital…has trended up over the past decade or so. Even in 2008, by which stage the financial crisis had begun to hit profits materially, the global [return on capital] remained above its long-term average.”

The authors of the GS paper in fact provide two measures of the return on capital. The first, which they call the “yield on capital”, is calculated using a method similar to that of the Marxist theorist Robert Brenner, who argues that capitalism has faced a long-term crisis of profitability.4 However, the GS paper’s authors would not agree with Brenner’s conclusion. They argue that the genuine return on capital has to be formed by adding a second quantity, the “capital gain”, to the yield on capital. So, we have:

Return on capital = yield on capital + capital gain

The capital gain is measured by the “fractional changes in real capital prices” over a given period, relative to the changes in prices of goods consumed by households.5 They seek to justify this through the argument that we must “consider a representative household…facing a choice of consuming” a quantity of capital “or investing it”, and that we must look at the total return the household receives through “forgoing consumption”.

But capital stock is not capital derived by households abstaining from consumption.6 In the real world capitalists tend to expand investment by using profits obtained from previous production, and ultimately from the exploitation of workers, or by borrowing money gathered together by banks. In neither case is the cost of capital goods relative to household consumption goods especially relevant. The essential question is whether the profits made on past investment can sustain future accumulation. In this context a rise in the price of capital stock is a bane as much as a blessing because capitalists have to purchase these goods in order to accumulate.

The yield on capital, though far from perfect, is a better proxy for the potential rate of expansion of the system. If this is understood, three interesting results emerge from the GS paper:

(1) The yield on capital for the US economy matches quite closely the pattern found in Robert Brenner’s work (see figure 1). But the GS paper misses out the crucial long-term trend because it only goes back to the early 1980s. If we track the yield figures back to the start of the post-war period, using Brenner’s method, we see that profit rates tended to decline up to the 1980s. After that time they stabilised, rising and falling through successive business cycles, probably trending upwards slightly. Crucially, we remain caught in a period of problematic profitability. In other words, the GS paper does not undermine Brenner’s analysis or that developed by this journal.7

Figure 1: Two methods of calculating the US profit rate

figure 1

(2) The global yield measured by GS follows a roughly similar pattern to that of the US yield over recent decades (see figure 2). This is important because most major discussions of profitability among Marxists have focused on the US profit rate. The GS research ought to give us more confidence that global patterns broadly follow US patterns.

Figure 2: Global and US yield on physical capital calculated by GS.

figure 2

(3) The gap between GS’s “return on capital” and their “yield on capital” is of interest (see figure 3). Recall that the former differs from the latter only in that it takes into account capital gains or losses. Taking the US figures, we see that the price of capital stock falls in the recessions of the early 1980s, 1990s and 2000s—examples of the depreciation of capital through crisis that Marx described. As the yield recovers from these recessions, the return on capital generally draws close to the yield, occasionally rising slightly above it. The exception, in which the return on capital soars above the yield, is the period 2004-7.

Figure 3: US yield and return on physical capital as calculated by GS

figure 3

A similar, but less extreme, trend emerges from their analysis of the EU5 economies8 and the global economy as a whole, though not China, where the boom was based on a genuinely high yield, with the return on capital lagging slightly below. This suggests that from 2004 there was, in the US and Western Europe at least, something of an investment frenzy, even in the non-financial sector, creating a temporary surge in capital costs without producing a sustained rise in profitability. And by 2008 the return on capital had, unsurprisingly, collapsed back below the yield.


1: Daly and Broadbent, 2009. There are many problems with the paper, but here I look only at the one of most interest from a Marxist perspective. I have based my criticisms on the rather brief and unsatisfactory account of the methodology provided in the paper itself. Thanks to Graham Turner for his input on some of the technical issues and for providing me with data on the breakdown of the capital stock depreciator.

2: A concept that will be familiar to readers of this journal. See, for instance, Callinicos, 2009, and Harman, 2008, for our analysis.

3: Harman, 2007. Note, however, that Marx’s rate of profit does not feature in economic statistics. Various proxies that might be expected to show a similar pattern are used by Marxists.

4: See Choonara, 2009, for a brief discussion of Brenner’s work.

5: Technically, the capital stock deflator deflated by the household consumption deflator. This is the ratio of the price indices of capital goods and household consumption goods.

6: About 170 years ago Marx castigated Nassau Senior for just such an “abstinence theory”.

7: Although there remains debate between Marxists on how best to measure profit rates. For instance, Brenner’s method, like that of GS, values capital stock at its current cost rather than its historical cost. Andrew Kliman has argued, for instance in his article in this journal, that the historical cost (ie the amount actually required to purchase the capital stock) should be used.

8: Germany, UK, France, Italy and Spain.


Callinicos, Alex, 2009, “An Apologist with Insights”, International Socialism 122 (spring 2009), www.isj.org.uk/?id=532

Choonara, Joseph, 2009, “Marxist Accounts of the Current Crisis”, International Socialism 123 (summer 2009), www.isj.org.uk/?id=557

Daly, Kevin, and Ben Broadbent, 2009, “The Savings Glut, the Return on Capital and the Rise in Risk Aversion”, Global Economics Paper No 185, Goldman Sachs Global Economics, Commodities and Strategy Research.

Harman, Chris, 2007, “The Rate of Profit and the World Today”, International Socialism 115 (summer 2007), www.isj.org.uk/?id=340

Harman, Chris, 2008, “From the Credit Crunch to the Spectre of Global Crisis”, International Socialism 118 (spring 2008), www.isj.org.uk/?id=421