Cycles within cycles

Nick Rogers reviews: Michael Roberts, 'The long depression: how it happened, why it happened and what happens next', Haymarket Books, 2016, pp360, £14.99

At 6.32am UK time on the morning of November 9 a notification arrived in my email inbox that Michael Roberts had made another posting to his blog. Just a couple of hours after it was confirmed that Donald Trump had been elected president of the United States, Roberts was ready with his analysis, astute as always, of the reasons why Trump had won and of the economic prospects facing the United States.1

A blogger of consistently high-quality economic analysis, Roberts has performed a valuable public service continuously for seven years (his first posting on his blog was on January 5 2010). Roberts’ perspective on the last decade is now brought together in this book - the story of the years since the crash of 2007-08, during which what began by common consent as a profound and deep recession (the title of Roberts’ 2009 book was The great recession2) has morphed into a period of sustained stagnation that Roberts thinks merits being described as nothing less than a depression.

From recession to depression

In the central chapters of his book, Roberts surveys the world economy and piles up the evidence that conditions do not represent anything like a normal recovery from a recession (the technical definition of a recession being at least two continuous quarters of negative growth).

True, most economies (putting obvious exceptions like Greece to one side) have been growing to some extent or another since some point in 2009 or 2010. So economies were actually shrinking for less than two years. Still, this was far longer than the 10-month average duration of other post-war recessions. In the United States GDP fell for 20 straight months. More to the point, the fall in economic output during that period was exceptionally sharp: the advanced economies as a whole contracted by 6% of GDP (a difference for their economies compared with a normal growth rate of 8%); industrial production fell by 13%; world trade fell by 20%; and world stock markets fell 50%.

However, what makes the period since 2007 (when the financial sectors in the US and the UK first started encountering choppy waters) different is the tepid nature of the recovery from the 2008 crash - in the UK GDP took until 2014 to reach the same level it had been at in 2008, and industrial production and manufacturing are still lower (in September 2016 the UK’s output from production was 7.9% lower and from manufacturing 5.5% lower than they had been in March 2008).

Growth rates in all economies (even China’s, which has maintained GDP growth above most other large economies) have remained well below pre-crisis trends. The economic performance of the United States (and even the UK) is a little better than that of the euro zone or Japan, but still very poor, compared with the pre-2007-08 years. The high levels of private debt that characterised the pre-2007 financial and property bubbles have been joined by high levels of public debt - bailing out the banks in the midst of a shrinking economy and rising social security demands could hardly have any other outcome. World trade has been growing more slowly than even the output of goods - that really is unprecedented behaviour for a global economy in which trade traditionally grows faster than GDP.

The productivity of workers in most economies shows barely any sign of growing - despite the boost to productivity that might be expected from worse performing companies going to the wall in the recession. Roberts gives figures for 2013 and 2014 that show the growth of workers’ hourly productivity in the US declining from 1.2% to 0.7%, in Japan from 1% to an actual fall of 0.6%, and the euro zone flatlining at 0.2% and 0.3% for those two years. The UK’s productivity figures are among the worst. Unsurprisingly, in these circumstances, wages almost everywhere have either stagnated or fallen.

And all this comes despite the extraordinary measures being carried out by central banks and governments. Interest rates have been kept at close to zero for eight years and negative interest rates of one sort or another are being applied to a quarter of the global economy - in effect, central banks are charging depositors for leaving money with them. And, of course, that mild tongue-teaser, quantitative easing, has entered common parlance. What would have been condemned as the irresponsible printing of money (although in reality the policy involves the electronic purchase of bonds), is being practised in the US, UK, euro zone and Japan with little noticeable effect on inflation - or economic activity.

The dire nature of the chronic problems afflicting global economies are broadly accepted. Take Martin Wolf, the chief economic commentator of the Financial Times, who on the day of the US elections wrote: “The next administration will take over a country with a mediocre growth of productivity, high inequality, a growing retreat from work and a declining rate of new creation of businesses and jobs.”3

Larry Summers, former US treasury secretary under Bill Clinton and a leading candidate for chair of the US federal reserve against the ultimately successful Janet Yellen, has written of this being a period of “secular stagnation”, which he defines as “the inability of the industrial world to grow at healthy rates even with loose monetary policies”.4

The Keynesian economist, Brad DeLong, like Roberts plumps for “depression” as an apt description of the present period: “At some point we will have to stop calling this thing ‘the great recession’ and start calling it ‘the greater depression’.”5

The key difference that Roberts has with mainstream economists, even those of a Keynesian inclination, is that he locates the crux of global capitalism’s current contradictions in the inner workings of capitalist production itself: Karl Marx’s theoretical insight that capitalism’s drive to increase productivity tends, paradoxically, to drive down profitability.


The rate of profit is the central theme of the third volume of Marx’s Capital - the definition of the rate of profit, how it is formed, how the distribution of profit between the different categories of capitalist impacts on it. And it is in discussing the “law of the tendential fall in the rate of profit” that Marx most clearly sets out an analysis of capitalist boom and slump. Yet for much of the last century it has barely featured in the thinking of many (probably most) Marxist economists. Even today, phenomena such as financial speculation and rising inequality play a greater role in the analyses by the majority of Marxist academics of capitalist crisis.

Michael Roberts is one of a small band of Marxist economists who seek to place profitability at the heart of their analysis of capitalism. For a system based on production for profit, it would be fairly astonishing if expectations of the return on their investments played as little a role in the thinking of capitalists as it does in the thinking of many of those inhabiting Marxist academia.

Measures of rates of profit both across history and from around the world are integrated into the discussion throughout The long depression. For instance, in applying the term ‘depression’ to the global economy, Roberts is explicitly comparing the economic performance of the last decade with two other major economic events in history: the long depression of 1873 to 1897 and the great depression of 1929 to 1939, when deep initial slumps were followed by long periods of economic stagnation.

Roberts outlines the controversy in the economic literature about whether the two decades or more after 1873 were a depression. He leans heavily on the work of Arthur Lewis6 in concluding there is “definitive proof there was a long depression”. However, he takes issue with Lewis’s profit-squeeze theory explanation (increasing competition from rising economies acting to reduce British profit margins) and provides alternative evidence that the dive in the profitability of British capital from the early 1870s to the mid-1890s was caused by rising capital intensity. As British capitalists increased investment in equipment and machinery in order to raise the productivity of their workers, the ratio between the quantity of capital goods and workers (ie, the organic composition of capital7) rose. Only workers produce value, so the output of value per unit of investment (and with it the rate of profit) fell. As their return on investment worsened, capitalists became more reluctant to invest and the economy tended to grow more slowly.

Roberts maintains that essentially the same factors were behind the great depression - there is no controversy about the severity of the economic crisis of 1929 onwards. According to Roberts, profitability started falling from 1924 and failed to recover during the 1930s and that goes a long way to explain the origins and course of this cataclysmic event in economic history.

Similarly, the stagnation of the last decade reflects lower profitability and a collapse in investment. World War II served to boost profits and, along with high levels of investment, launched the ‘golden age’ of the post-war era. From the mid-1960s profitability began to fall, reflecting yet again the rise in the organic composition of capital (Andrew Glyn and Bob Sutcliffe formulated an alternative profits-squeeze explanation based on the heightened militancy of workers successfully raising wages at the expense of profits8). The fall in profitability continued through to the early 1980s, leading to the recessions of 1974-75 and 1980-82. The political response was neoliberalism - a determined effort to raise the profitability of capital. Partly by increasing the exploitation of workers (the rate of surplus value), but also with the cheapening of the price of capital goods (constant capital) caused by the high-tech revolution, the rate of profit was partially restored - but never to the same levels as during the post-war boom. From 1997 profitability began to fall again - leading, via a credit-fuelled financial and property binge, to the crash of 2008.

Roberts does not discount the role of debt - both private and public - in undermining economic growth. He devotes a whole chapter to this phenomenon. Lowered profitability makes reducing (deleveraging) debt more difficult, which in turn holds back investment. This is the explanation for the rise in corporate holdings of cash (hoarding). Roberts surveys alternative explanations for the poor investment record. He locates confirmation of the importance of profitability in some unlikely mainstream analyses, including a paper from the Bank for International Settlements that finds a “seemingly more plausible explanation for slow growth in capital formation” to be “a lack of profitable investment opportunities”. It concludes that “even if they are relatively confident about future demand conditions, firms may be reluctant to invest if they believe the returns on additional capital will be low”.9

Roberts’ theory of economic depressions - of which the period since 2007-08 is in his estimation the third in the history of capitalism - is not entirely dependent on measures of profitability. He outlines a rather complex model, in which several economic cycles of varying length operate to more or less defined timescales.

A four-year cycle based on the building up by capitalists of stocks of inventories (identified by Joseph Kitchen) operates within the better known business (or Juglar) cycle lasting eight to 10 years. At double the length of the business cycle is the 18-year Kuznets cycle of construction and infrastructure investment. Then we get to the profits cycle, which measures between 32 and 36 years from trough to trough. At roughly double the length of the profits cycle, the 50 to 70-year (and relatively renowned) Kondratiev cycle reflects very long-term changes in production and commodity prices and depends on “global demographic and resource factors”.

Confusingly, Roberts occasionally mentions a cycle in clusters of innovation (as identified by Joseph Schumpeter) and seems to distinguish this from the Kondratiev cycle, suggesting it takes place over an even longer timescale. Roberts also touches on a stock market and a credit or financial cycle that run in tandem with profit (inversely in the case of the credit/financial cycle)10.

Roberts’ thesis is that a depression occurs in the world economy when all these economic cycles are synchronised in their downward phase:

This conjunction does not happen very often. Indeed, given the duration of the long Kondratiev cycle, it can only happen once every 50 to 70 years. If this is right, then it explains why the start of the 19th century depression in 1873 was only repeated 56 years later with the start of the great depression in 1929 and with the start of the current long depression in 2008, some 79 years after that.11

His thesis also explains why not all economic downturns lead to depressions:

The up wave in the profit cycle from 1946 to 1965 coincided with the up wave in the Kondratiev cycle. Thus the troughs in the Juglar and Kuznets cycles in the mid-1950s did not produce a very deep recession or downturn in economic growth and employment. Because the Kitchin cycle troughed also in 1958, the ‘pause’ was longer than in 1954. But high and rising profitability in an environment of a Kondratiev up wave was generally good news for capitalism.

From 1965 to 1982, the rate of profit fell. The Kondratiev cycle was still in an upswing of prices, though. What we got was successively worse economic slumps (1970, 1974, and 1980-82) alongside rising prices - in other words ‘stagflation’. In 1974, the Kuznets, Juglar, and Kitchin cycles troughed together. In an environment of falling profitability, world capitalism suffered its first post-war simultaneous economic slump. The 1980-82 recession was deep and long-lasting because profitability reached lows and the Kondratiev prices cycle peaked. But the real estate Kuznets cycle was also at a peak, so output and employment fell while prices stayed up - the ultimate stagflation crisis (pp232-33).


The long depression is highly readable and brim-full of valuable analysis and facts. It also makes an intriguing and challenging contribution to Marxist economic theory, which I think is worth engaging with. In this section of my review I want to touch on a few problems - not just with Michael Roberts’ theoretical approach, but with the approach of the broader profitability school of Marxism: the interpretation of Marx’s political economy with which I largely agree.

First, Roberts’ ‘cycles within cycles’ thesis strikes me as somewhat over-elaborate and more than a little eclectic. He seems to have taken note of every economic cycle ever identified by an economist (the Kitchin, Juglar, Kuznets and Kondratiev cycles are all named after the economists who first brought them to public attention) and woven it into his schema. Now when Roberts speaks of a cycle he correctly means an economic process which self-generates its ups and downs: ie, the basis for the slump is set during the boom, and the slump creates the conditions for the boom to come. The impulse for switching between one phase of the cycle and the next (to use the phraseology of theorists of cycles) is endogenous (internal) rather than exogenous (external) to the cycle. And, what is more, there is a determinism to the cycle: it happens inevitably and over a fixed time period, unless exogenous events intervene.

To be fair, Roberts says that the thesis is “really a series of propositions that are not fully confirmed by the evidence” (p217). What is not clear is precisely what is driving any of these cycles to operate in as regular a way as the thesis demands. Do the rhythms of the global capitalist economy really fit such a neat pattern? Even when it comes to the profits cycle, while Roberts is correct to say that the rate of profit trends up or down - depending on the balance between the tendency for production to become more capital-intensive (thus reducing the rate of profit) and the counteracting factors that either increase the rate of surplus value or reduce the value of the capital good inputs (thus increasing the rate of profit) - I just do not see any theoretical explanation for why this should be constrained to a regular 32- to 36-year cycle.12

Nor does Roberts explain the extent to which profitability drives either the shorter or longer-term cycles, with which it interacts. If it does not and there are a range of other factors driving the cycles - some in the realm of circulation rather than production (which seems to be the case) - then how far is profitability really the principal driver of economic crises? Is profit not left as just one of many causes of capitalism’s inherent instability?

Second, how far do recessions or slumps prepare the ground for the next upturn, particularly in the context of depressions? Roberts quotes Yanis Varoufakis’s criticism of Marx on capitalist crisis:

Marx told the story of redemptive recessions … However, there was nothing redemptive about the great depression. The 1930s slump was just that: a slump that behaved very much like a static equilibrium - a state of the economy that seemed perfectly capable of perpetuating itself …13

I think Varoufakis is basically correct about the great depression. During the period 1929-33 there was almost four years of continuous economic contraction. It was the longest recession in capitalist economic history. Economic assets were pulverised, a huge proportion of the population left without work and prices of all sort of commodities driven down in an unprecedented manner. Yet this did not create the conditions for a proper recovery. The rest of the 1930s resembled our era of stagnation and woefully below-trend growth, but with much higher unemployment and social destitution. Clearly the massive devaluation of capital between 1929 and 1933 was insufficient to trigger a new upturn.

This brings us to the third problem that I think remains to be addressed by the profitability school of Marxist economics. Most economists are agreed that in the case of the great depression it was World War II that played a major role in setting the stage for the post-war boom. But how?

Often too easy a parallel is drawn between the destruction of value in a slump and the destruction of use-values in a war. Take Roberts:

War adds a new dimension to ‘creative destruction’. Physical destruction of the stock of capital accompanies value destruction. This produces a dramatic fall in the cost of capital. War is an exogenous event that can sharply interfere in these endogenous profit cycles (p225).

But the ‘creative destruction’ of war simply does not have the same impact on the cost of capital as an economic slump. A slump, it is true, will tend to reduce the price that the owners of an asset can secure for it, while leaving the asset (at least initially) intact and in a reusable state. During or after a slump it is possible for new investors to come along and take advantage of the plight of capitalists who are struggling by purchasing their assets at knock-down prices That, of course, greatly increases the chance of the new investors making a decent profit, once the economy picks up. It is just that in the case of the 1929-33 slump that was not sufficient to kick-start capitalist economies.

War, on the other hand, destroys physical assets. A factory that has been bombed out no longer exists. There is nothing for investors to come along and buy at any price. On the contrary, they have to resort to the output of whatever productive facilities have survived in a situation generally of short supply and rationing. They are going to have to pay more than they would hope for capital goods and other inputs into production.

So a slump tends to push the prices of capital goods and other supplies below their value; a war will tend to force their price above their value. The destruction of value and the destruction of use-values are in no way comparable. The impact of World War II on the global economy must be sought elsewhere.14

If Roberts’ analysis of how capitalism extracted itself from the great depression is flawed, then his prediction (p269) that the current economic downturn is likely to shift towards a period of more vibrant economic behaviour from about 2018 (after another sharp recession has devalued even more capital), may be a misreading of the course of economic events. Could we be facing an immediate future more closely aligned with the 25-year, late 19th century long depression than with the shorter, decade-long great depression? Japan, which has already suffered economic stagnation for a quarter of a century (since its banking crisis of the early 1990s), may be a warning of the fate awaiting other advanced capitalist economies.

Despite my reservations (and these are issues for which no-one has really provided a satisfactory explanation), The long depression is a fascinating book. Above all, it clearly identifies the extraordinary nature of the economic period through which we are living - economic conditions that have given rise to political phenomena such as the Brexit vote in the UK and the election of Trump in the US, and no doubt will present us with more surprises in the months and years to come.


1. M Roberts, ‘Donald Trump and the poisoned chalice of the US economy’’: https://thenextrecession.wordpress.com/2016/11/09/donald-trump-and-the-poisoned-chalice-of-the-us-economy.

2. M Roberts The great recession 2009, available at https://libcom.org/files/The%20Great%20Recession%20-%20Profit%20cycles,%20economic%20crisis.pdf.

3. Financial Times November 8 2016.

4. Financial Times October 8 2015.

5. Quoted by the author on p119.

6. WA Lewis, ‘The deceleration of British growth, 1871-1913’, discussion paper No3, Development Research Project, Princeton University, November 1967.

7. Since Roberts does not distinguish between the different definitions of the composition of capital that Marx provides, I assume he is equating the value and organic compositions of capital. Some other Marxist economists - eg, Andrew Kliman - define the organic composition of capital as broadly the same as its technical composition.

8. A Glyn and B Sutcliffe British capitalism, workers and the profits squeeze London 1972.

9. R Banerjee, J Kearns, M Lombardi, ‘(Why) is investment weak?’, Bank for International Settlements, 2015: www.bis.org/publ/qtrpdf/r_qt1503g.pdf.

10. See pp5, 217-18 and 225-33 for a detailed discussion of the different kinds of cycles.

11. p218.

12. I have set out an alternative, less deterministic thesis of capitalist economic crisis in ‘“Revolutions in value” and capitalist crisis’ Weekly Worker November 19 2015; and ‘Profitability and the dynamics of capitalism’ Weekly Worker November 26 2015, where I engage with the work of Andrew Kliman and others.

13. Y Varoufakis, ‘Confessions of an erratic Marxist in the midst of a repugnant European crisis’, December 10 2013 (https://yanisvaroufakis.eu/2013/12/10/confessions-of-an-erratic-marxist-in-the-midst-of-a-repugnant-european-crisis), quoted by Roberts in footnote 20 to chapter 3, p306.

14. In ‘Profitability and the dynamics of capitalism’ (Weekly Worker November 26 2015) I have suggested that there was a sharp reduction in the composition of capital during World War II as a result of a number of developments, but including tight state control of research and development and production.