Business cycles, unit roots and animal spirits

The Golden Age of capitalism, when the major economies grew at over 4% real GDP a year and there was relatively moderate inflation and no significant fluctuation in employment i.e slumps, lasted just a short time – from about the late 1950s to the early 1970s.

After that, the major capitalist economies experienced a series of regular and recurrent slumps starting with the first simultaneous post-war international recession in 1974-5, the deep ‘double-dip’ recession of 1980-2, the industrial slump of 1990-2, the mild but global recession of 2001 and finally the Great Recession of 2008-9, the deepest and longest lasting slump since the 1930s Great Depression.

Mainstream macroeconomics did not see these recessions coming and even after they arrived, economists failed to consider their causes or even accept that what mainstream economics used to call ‘business cycles’ were back.

The Great Recession has forced the mainstream to consider causes and explanations more carefully. Keynesians continue to revive the view that slumps are due to sudden collapses in ‘effective demand’ and/or changes in ‘animal spirits’ (the psychological temper or confidence of entrepreneurs about the future). As the Great Recession has morphed into a Long Depression, where there is no recovery to previous trend growth in output, investment or incomes, Keynesian theory has dredged up the ideas of the pre-war Keynesian Alvin Hansen who proposed that the immediate post-war capitalist economies would enter ‘secular stagnation’ due to a slowdown in population growth and chronic weak demand (he was wrong).

The doyens of modern Keynesian economics, Paul Krugman and Larry Summers, now hold that the major economies (or at least the US) are in a permanent liquidity trap, where even with interest rates near zero, business investment won’t pick up enough to restore full employment (not that capitalism has ever achieved that except maybe in those few ‘golden’ years of the 1960s). This stagnation can only be broken by government intervention and/or investment.

The still dominant neoclassical school of economics denies that such stagnation exists or is endogenous to the capitalist economic system. For this school, the Great Recession was a particularly large ‘shock’ to an otherwise steady development of output, investment and employment. But it was temporary – Ben Bernanke in his new blog is determined to provide the reasons why it is temporary (see my post, These economists reckon the issue is due to either bad monetary policy (Bernanke) or the lack of control over banks and credit, causing financial crises (Rogoff, – not a problem of the lack of demand or ‘secular stagnation’. So the answer is not more government investment and government borrowing but financial stability and solid monetary policy.

As Brad de Long put it in a recent post (, “Summers and Krugman now believe that more expansionary fiscal policies could accomplish a great deal of good. In contrast, Rogoff still believes that attempting to cure an overhang of bad underwater private debt via issuing mountains of government debt currently judged safe is too dangerous–for when the private debt was issued it too was regarded as safe.”

The concept of the nature of modern capitalist economies as ones that have an equilibrium growth path that sometimes is knocked off kilter by random events and then returns to equilibrium has led to a whole research programme on ‘business cycles’ based on dynamic stochastic general equilibrium (DSGE) models that purport to consider the impact of various ‘shocks’ on a model capitalist economy – shocks like changes in the attitudes of investors or consumers and policies of governments (‘representative agents’).

Unfortunately, DSGE models have signally failed to offer any clear explanation of what is happening in modern capitalist economies, let along provide a guide to predicting future downturns in the ‘business cycle’ – see my post,

Recently, two top-line economists, Roger Farmer (Keynesian) and John Cochrane (neoclassical) have tried to feel their way to a compromise position between whether capitalist economies can stay locked in below-trend growth after a slump or not.

And it’s all about what is called unit roots. Unit roots are a statistical phenomenon where, say, when there is a collapse in output or unemployment, this may be only temporary and output or unemployment will start to return towards its previous trend, but not all the way. This contrasts with ‘stationary’ phenomena where the shock is eventually corrected and the previous trend is re-established and a ‘random walk’ where the trend remains at a new (lower) level permanently. The graph is from John Cochrane’s blog post (

unit roots

Roger Farmer has been arguing that economies can suffer a slump in demand and thus in output, investment and employment that can move an economy from one equilibrium trend to another lower one which is where it settles – and this change is due to a chronic weakness in demand not to changes in long-term supply-side factors like productivity or population growth, as neoclassical growth theory reckons (

Farmer reckons that there are both transitory and permanent elements in the business cycle, so output or unemployment can exhibit movements close to unit roots. Actually, a unit root description of a business cycle that does not return to the previous trend is very close to my own schematic characterisation of a depression as taking the form of a square root – see my post,

Farmer reckons that business cycles are caused by changes in ‘animal spirits’: “My answer is that aggregate demand, driven by animal spirits, is pulling the economy from one inefficient equilibrium to another.” So “If permanent movements in the unemployment rate are caused by shifts in aggregate demand, as I believe, we can and should be reacting against these shifts by steering the economy back to the socially optimal unemployment rate.”

So Farmer reckons, as the business cycle is a unit root, it does not self-correct and governments must intervene to smooth out the fluctuations and get unemployment down. John Cochrane is not convinced of the need for government intervention, of course, but he does recognise that there can be a chronic or permanent element in changes in unemployment and it may not be totally self-correcting. A unit root is a good compromise, it seems.

Good news, eh! Keynesian and neoclassical economics have moved to a compromise in theory that capitalist economies do fluctuate (due to ‘shocks’ in demand or supply) and may not always return to previous trends. And something exogenous (government) may have to act to correct it.

The fact that neither neoclassical non-intervention policies nor Keynesian policies of macro-management had any effect in stopping the reappearance of the ‘business cycle’ from the 1970s onwards or controlling them appears to have escaped both Farmer and Cochrane. Instead, they continue to debate the nature of the ‘shocks’ to the system.

There is little doubt that capitalist economies are not ‘self-correcting’ and a level of unemployment or real GDP growth that existed before a major slump may well not return after the recession ends – indeed the current slow crawl of ‘recovery’ since the trough of the Great Recession in 2009 proves that with a vengeance.

And there is new evidence of that theoretically. A new working paper by Daron Acemoglu, Ufuk Akcigit, and William Kerr looks at the pattern of how economic disturbances propagate throughout the industrial and regional network ( They examine several types of disturbances such as changes in Chinese imports, government spending and productivity. Some of these ‘shocks’ propagate upstream through the value chain, from retailers to suppliers. They call these demand shocks. Others move in the opposite direction, and they call these supply shocks.

Keynesian blogger, Noah Smith is very excited at this research. As he puts it: “the implication is that the rosy picture of the economy as a smoothly functioning machine isn’t necessarily an accurate one. The tinker-toy web of suppliers and customers and regional economies in Acemoglu et al.’s paper is a fragile thing, easily disturbed by the winds of randomness”.

The authors of the paper conclude: “Quantitatively, the network-based propagation is larger than the direct e§ects of the shocks, sometimes by several-fold. We also show quantitatively large effects from the geographic network, capturing the fact that the local propagation of a shock to an industry will fall more heavily on other industries that tend to collocate with it across local markets. Our results suggest that the transmission of various different types of shocks through economic networks and industry interlinkages could have Örst-order implications for the macroeconomy.”

Also, Smith concludes that “one of the biggest and longest-lasting economic debates is whether government spending can affect the real economy. Lucas and others (the neoclassicals) have claimed that it can’t. But in Acemoglu et al.’s model, it absolutely can, since the government is part — a very big, very important part — of the network of buyers and sellers.”

But all the paper confirms, by using bottom-up input-output connections, is that the collapse or bankruptcy of a large firm or bank or sharp change in trade can trigger a crisis by cascading through an economy. This shows how slumps can start but suggests that they are due to random events. That does not explain the recurrent nature of slumps and so explains nothing.

Smith claims that the paper “would solve the problem of what causes recessions. Currently, we have very little idea of what tips economies from boom over to bust — there is usually no big obvious change in productivity, technology or government policy at the beginning of a recession. If the economy is a fragile complex system, it might only take a small shock to send the whole thing into convulsions.”

So the cause of recessions is random shocks that multiply. That is no more an explanation than that proffered by Nassim Taleb in his book, Black Swan, or by the heads of the American banks during the financial crisis, that it was a chance in a billion (

In none of these current debates is there any mention of the role of profit in the ‘business cycle’ in what is essentially a profit-making system of production. Business cycles are back according to macroeconomics – rather belatedly. But it’s random or it’s not random; it’s demand or it’s not demand; it’s monetary or it’s not monetary. Mainstream theory remains in a fog of confusion.


11 thoughts on “Business cycles, unit roots and animal spirits

  1. A central aspect of Marxist thought is that things don’t remain the same. Before, Britain became the pre-eminent industrial power, the baton rested with the Netherlands. Before that the centre of capitalism resided with the mediterranean city states, which were developing a form of mercantile capitalism, which ultimately destroyed itself, because, as Marx outlines, surplus value comes from production not exchange.

    In Theories of Surplus Value, Marx explains that it was the Physiocrats who first correctly analysed production as the source of surplus value, whilst English Political Economy was stuck in the rut of Mercantilism and the Money School, because the English economy at the time was dominated by Britain’s role as a trading and colonial power, whilst the Physiocrats were able to see quite clearly with France’s agricultural economy that its surplus product, and surplus value arose in production, because the new value produced was greater than the value of the labour power consumed.

    In the same way, during the 19th century, Britain’s pre-eminent position gave way to challenges from Germany, and the US, and then from Japan.

    Consequently, I find the idea of “the major capitalist economies” being defined in rather fixed and static terms as those economies that held that position in the past, rather odd. In just the same way that the US and other economies surpassed the UK, in the past, so today China is surpassing the US, and has already way surpassed Britain et al.

    If we compare China’s growth over the last 15 years, it rather knocks the socks of the 4% growth of the major capitalist powers in the previous “golden age”. In fact, with above 10% growth during most of that period, its growth has been so rapid, that even today with growth at “only” 7%, in absolute terms it is higher than it was a decade ago at 12%, because the economy itself is more than twice as big!

    But, its not just China, because the process of combined and uneven development means that other economies in Asia and elsewhere have been growing rapidly too. Growth in much of developing sub-saharan Africa has been running at around 10%.

    I suppose that if you keep watching Linford Christie run, you will come to the conclusion that times from sprinters are slowing down. If you watch Usain Bolt, you will come to a different conclusion.

    1. “as Marx outlines, surplus value comes from production not exchange”. True enough, but what then is the status of mercantile (money) “profits”? Did not medieval merchants trade commodities, and exchange them against money? And was money a commodity before the capitalist era?

      If the answers are no, simply because mercantile trade was not associated with the production of surplus value, they why do we persist in reference to merchant “capital”?

      These are not merely historical questions. “Merchant” – purely commercial – profits continue to exist today, either in (decreasingly) independent form, or much more likely, in combination with industrial or banking capital. In Vol II, Part 4, Marx only analyzed the independent form, but in either case Marx completely overlooked the analysis of commercial surplus profits. As with the surplus profits transformed into the rents of landed property, their bases lies in “extra-economic” force. Historically associated with the “so-called primitive” accumulation, so-called by Marx because he held that these surplus profits were a permanent feature of contemporary capitalism, these remain permanent because these forms of surplus profits arise out of the necessary conditions for the existence of productive capital – private property (rent), the state (commercial arbitrage). Both rent and arbitrage intertwine in close connection with that sine qua non of capitalist production, labor power and its reproduction. Needless to say, both have a special financial connection to money capital.

      Finally, these forms of surplus profits not only continue to exist under modern capitalism, in the era of capitalism’s decay – another concept completely forgotten by Marxists, including those claiming adherence to “Leninism” – but grow as ever more prominent features of capitalist accumulation, precisely as the whole productive circuit of capital lands in ever deeper historical crises. Forgotten just in time for what looks to be the early stages of capitalism’s greatest historical crisis ever!

      1. Matthew,

        Yes, medieval merchants did trade commodities, and money was a commodity, though as Marx describes a special type of commodity, one that gives up its own specific use value, in acting as money, in order to take on the use value of money as universal commodity.

        The reason Marx talks about “money-capital” and “merchant-capital” in relation to these forms during this period, is precisely for the reason I have set out in describing these elements of Marx’s analysis on my blog. It is also as Marx, describes, and as Engels draws attention to in his own Preface to Volume III. That is that Marx’s method cannot be understood if you work with fixed frozen definitions.

        In describing why those who seek some kind of “objective” fixed definition in Marx simply have not understood him, Engels writes,

        “They rest upon the false assumption that Marx wishes to define where he only investigates, and that in general one might expect fixed, cut-to-measure, once and for all applicable definitions in Marx’s works. It is self-evident that where things and their interrelations are conceived, not as fixed, but as changing, their mental images, the ideas, are likewise subject to change and transformation; and they are not encapsulated in rigid definitions, but are developed in their historical or logical process of formation.”

        This is what marx says specifically when he talks about money-capital and merchant capital. he defines them as “ante-diluvian” forms of capital, in other words they are capital in the process of becoming. But, the point that the profits they obtain are a consequence of surplus value created in production still holds, and as Marx goes on to describe the means by which those forms of capital obtain these profits is the same in medieval times as it is under capitalism, what is different is the conditions under which it does so, and the laws that govern it.

        The merchant and money-capital in pre-capitalist modes of production were able to obtain a share of the surplus value produced by the actual peasant producers – indeed just as were the feudal lords and the Church, for the reason that Marx says the Physiocrats were right to identify. It is that the surplus value exists initially in the shape of a surplus product.

        The peasant producer requires a certain quantity of use values – seeds etc. – as means of production, and these must by physically reproduced on a like for like basis out of current production. This is why as marx always does, following on from the Physiocrats, he values this means of production on the basis of its current reproduction cost, rather than its historical cost. It is the proportion of current social labour-time, of the current gross output that is relevant for the determination of social reproduction and assessing how much is left over as surplus, not how much labour-time was required at some point in the past.

        Of the peasant producers gross output, therefore, Marx says, this physical product equal to the means of production is simply reproduced both in terms of use value and value, because its value is determined by its current reproduction cost. It is then only the new production of labour that represents revenue, i.e. is available for distribution as wages, profits, rent and interest.

        But, as marx demonstrates in Theories of Surplus Value, and in the concluding Chapters of Volume III examining social reproduction, it is not just the means of production that must be physically reproduced on a like for like basis out of current production. If social reproduction is to occur, then the means of consumption must also be likewise reproduced on a like for like basis. Otherwise, the producers themselves cannot reproduce their labour-power, they cannot then produce.

        The basis of the surplus product, and surplus value of any society then comes down to the difference between the value of labour-power, i.e. how much of social labour-time is required to produce the commodities required to reproduce the producers, and the value produced by that labour-power.

        The Physiocrats could get away with a definition of value and surplus value based upon use value rather than value, Marx says in TOSV, because in an agricultural economy, there was effectively no difference between the two. In other words, what the agricultural producers consumed each year did not change much, and consisted of what they produced. If the use values they consumed, were less than the use values they produced – discounting the use values required to simply reproduce the means of production – they made a surplus, and that surplus could be appropriated by the landlord.

        But, it could also be appropriated by merchants and money lenders. The fact remained that the surplus product and surplus value was created in production. The merchant and money lender merely realised a portion of it in exchange, against money.

        Merchant capital and money lending capital obtain profits from the same source today, out of the surplus value created in production, but the conditions under which they do so, and the laws governing that are different.

        Merchant capital obtains a share of surplus value on the same basis as any other capital, i.e. as capital it obtains the average rate of profit. As Marx sets out, Money-dealing capital is only a form of merchant capital, and obtains the average rate of profit.

        Money lending capital, by contrast, is fictitious capital. It does not obtain the average rate of profit, but obtains the market rate of interest, determined by the demand and supply of money-capital.

        Under Capitalism, merchant capital obtains the average rate of profit, because it is capital. But money lending capital can only obtain an average rate of interest. As Marx sets out in Volume III. Under pre-capitalist modes of production, profit is a deduction from rent, but under capitalism, rent is a deduction from profit, and interest falls into the same category as rent. The laws which determine how much can be paid as interest or as rent are themselves determined by the laws that determine the rate of profit, but the laws that determine the rate of profit, are the laws which determine the current value of commodities, i.e. the labour-time required to reproduce the commodities that comprise the means of production, the means of consumption, and the value of the commodities produced by that labour-power.

      2. Matthew,

        Incidentally, Marx does deal with the issue of surplus profits in relation to merchant and other forms of capital, arising from market frictions, monopoly power and so on, but only to discount it, because it is a distraction from the core of the analysis. It is a matter for analysis under “Competition”, which he intended to come to later, in examining the phenomenal forms of all these categories.

        But, he says, the fact is that the total of surplus value remains what it is, and if some capital obtains surplus profits because it has a monopoly or other advantage, the consequence is that it sucks surplus value from some other capital.

        This is different from a situation where, for example, merchant capital obtains higher or lower profits than industrial capital. Where that is the case, Marx says, capital will to it from productive-capital, or away from it towards productive capital, depending upon whether the profit is higher or lower than the average, i.e. no different than the process of creating an average rate of profit in general.

        Incidentally, as Marx describes in The Poverty of Philosophy this was his general attitude to monopoly itself. That is monopoly creates competition, which breaks up the monopoly, just as competition creates monopoly by bringing about concentration and centralisation. It just does so on a higher level each time.

  2. As Michael Roberts says, what’s problematic about the approach of people like Noah Smith is not the idea that (possibly) small shocks can propagate into systemic crises, but the failure to accept that the reason for this is structural (in other words the falling tendency of the profit rate).

    However, it would be a pity if this error, or the self-important misunderstandings of Nassim Taleb — who completely misses the point of the black-swan metaphor — led to a downplaying of the importance of statistical understanding of capitalism and thus its status as a complex system with emergent properties.

    The existence of power-law tails in the distributions of a wide variety of data is a sign that capitalism is, as the complexity literature puts it a system “on the edge of chaos” — hovering perpetually on the brink of crisis, which will ensue in the event of a significant challenge to the system.

    Marxists in particular need to engage with these ideas because of their potential to be adapted for apologetical purposes (see Beinhocker’s “The Origins of Wealth” as an example — although note also that it is an excellent introduction to the relevant literature).

    For a quick way into Marxist approaches to this I’m forced to self-advertise: J. Wells, “Of Fat Cats and Fat Tails: From the Financial Crisis to the ‘New’ Probabilistic Marxism”, Research in Political Economy, 28 (2013), pp. 197–228.

    For a longer way, the following is highly-recommended: A. F. Cottrell, P. Cockshott, G. J. Michaelson, I. P. Wright, and V. Yakovenko, “Classical Econophysics”, Routledge, Abingdon, 2009.

  3. “As Michael Roberts says, what’s problematic about the approach of people like Noah Smith is not the idea that (possibly) small shocks can propagate into systemic crises, but the failure to accept that the reason for this is structural (in other words the falling tendency of the profit rate).”

    The problem here is that the cause and effect nexus has been reversed in examining the correlation. The Law of the Tendency for the Rate of Profit to Fall depends upon rapid rises in social productivity caused by the introduction of new technologies on an intensive basis, to replace labour both absolutely (in some sectors) and relatively (in general).

    As Marx makes clear, for long periods, capital is accumulated extensively, i.e. on the basis of simply rolling out more of the same technology, which creates no reason for a falling rate of profit. In fact, to the extent that this extensive accumulation causes the rate of turnover to rise, it causes the annual rate of profit to rise.

    As Marx says,

    “Growth of capital, hence accumulation of capital, does not imply a fall in the rate of profit, unless it is accompanied by the aforementioned changes in the proportion of the organic constituents of capital. Now it so happens that in spite of the constant daily revolutions in the mode of production, now this and now that larger or smaller portion of the total capital continues to accumulate for certain periods on the basis of a given average proportion of those constituents, so that there is no organic change with its growth, and consequently no cause for a fall in the rate of profit. This constant expansion of capital, hence also an expansion of production, on the basis of the old method of production which goes quietly on while new methods are already being introduced at its side, is another reason, why the rate of profit does not decline as much as the total capital of society grows.”

    But, the points in time when these surges of new technology and intensive investment occur are themselves a function of the actual crises of overproduction, which arise because the available supplies of exploitable have been used up within capitalistic limits. As Marx puts it again,

    “Given the necessary means of production, i.e. , a sufficient accumulation of capital, the creation of surplus-value is only limited by the labouring population if the rate of surplus-value, i.e. , the intensity of exploitation, is given; and no other limit but the intensity of exploitation if the labouring population is given. And the capitalist process of production consists essentially of the production of surplus-value, represented in the surplus-product or that aliquot portion of the produced commodities materialising unpaid labour.”

    It is precisely because its no longer possible to extract surplus value from this given size of labour-force, with this given level of technology, which makes it necessary to develop new forms of technology so as to increase the rate of exploitation.

    Profits fell in the preceding period, not because of the Law of Falling Profits – this was in Marx’s terms a period of relatively static development of technology – but because existing labour supplies were being used up, pushing up wages, and thereby reducing the rate of surplus value.

    “As soon as capital would, therefore, have grown in such a ratio to the labouring population that neither the absolute working-time supplied by this population, nor the relative surplus working-time, could be expanded any further (this last would not be feasible at any rate in the case when the demand for labour were so strong that there were a tendency for wages to rise); at a point, therefore, when the increased capital produced just as much, or even less, surplus-value than it did before its increase, there would be absolute over-production of capital; i.e., the increased capital C + ΔC would produce no more, or even less, profit than capital C before its expansion by ΔC. In both cases there would be a steep and sudden fall in the general rate of profit, but this time due to a change in the composition of capital not caused by the development of the productive forces, but rather by a rise in the money-value of the variable capital (because of increased wages) and the corresponding reduction in the proportion of surplus-labour to necessary labour.”

    It is to resolve this situation of an adequate size of workforce given existing technology, to be able to extract additional surplus value that causes the need for new technology to be developed and introduced, which both then creates a relative surplus population, and also raises the rate of surplus value.

    It is precisely THIS process, which creates the conditions for the operation of the Law of the Tendency for The Rate of profit to Fall. The correlation between the Law and Crises is not that the Law causes crises, but that crises create the conditions for the need for a period of innovation and intensive accumulation, which rapidly raises social productivity, which raises the rate of surplus value, whilst simultaneously causing the rate of profit to fall.

    It is not the Law that is the cause of crises, but crises which lead to the need to resort to the Law as a means of resolving those crises! The reason that crises occur with a regular cycle is due to the reasons Marx describes, which is that fixed capital has a cycle of replacement, which itself tends to get synchronised. These cycles of replacement are related to the cycles for innovation, which in turn are related to the cycles during which existing technology can be rolled out, thereby using up existing labour supplies and so on.

    A look at Capital Volume I, itself indicates this mechanism. Marx explains that in the early days of capitalist production, prior to machine production proper, technology developed slowly. Opposition to machines did not come from workers, Marx says, who generally found that they created more work, and increased their living standards. Machines were seen he says, as a means of dealing with a shortage of labour relative to rapidly growing markets.

    Opposition to machines came not from workers, but from the existing guild producers, who were undercut. So, from the inception of capitalist production in the 15th century, right up to 1825, there was no crisis of overproduction. As marx describes all of the crises that occurred prior to that were financial crises, caused by the private banks. It is only when steam power is introduced on a large scale, which makes the power of machines expand phenomenally, that the first crisis of overproduction arises, in 1825.

    1. “As Marx makes clear, for long periods, capital is accumulated extensively, i.e. on the basis of simply rolling out more of the same technology, which creates no reason for a falling rate of profit.”

      How would population increases and demographic changes (increase in elderly) fit into this narrative?

      1. It depends upon whether any increase in population creates an additional labour supply that can be employed so as to create additional surplus value given the existing level of technology, and its effect on facilitating the realisation of the produced surplus value, i.e. by creating additional aggregate demand.

Leave a Reply

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

You are commenting using your 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 )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: