John Maynard Keynes: first invest

Keynes or Marx?

What is capital’s driving force? Michael Roberts explores the profit-investment nexus

The main themes of this year’s Historical Materialism conference in New York1last month were the Russian Revolution and the prospects for revolutionary change one hundred years later.

But my main interest, as always, was on the relevance of Marxist economic theory in explaining the current state of global capitalism - if you like, understanding the objective conditions for the struggle to replace capitalism with a socialist society.

On that theme, in a plenary session, professor Anwar Shaikh at the New School for Social Research (one of the most eminent heterodox economists around) and I looked at the state of the current economic situation for modern capitalism. Anwar concentrated on the main points from his massive book published last year - the culmination of 15 years of research.2 This is a major work of political economy, in which Anwar uses what he calls the classical approach of Adam Smith, David Ricardo and Karl Marx (and sometimes John Maynard Keynes) under one umbrella (not specifically Marxist apparently). His book is essential reading, as is the series of lectures3 that he has done to accompany it.

His main points at the plenary were to emphasise that capitalism is not a system that started off as competitive and then developed into monopoly capitalism, but it is one of turbulent “real competition”. There has never been perfect competition (as mainstream economics implies), from which we can look at ‘imperfections’ like monopoly.

Anwar went on to say that crises under capitalism are the result of falling profitability over time in a long down-wave. The neoliberal period from the early 1980s was a result of the rejection of Keynesian economics and the return of neoclassical theory and the replacement of fiscal management of the economy, which was not working, with monetarism from the likes of Milton Friedman. But even neoliberal policies could not avoid the great recession, he said. And since then there has not been a full recovery for capitalism. Massive monetary injections have avoided the destruction of capital values, but at the expense of stagnation.

In my contribution, I emphasised the points of my book, The long depression,4 which also saw the current crisis as a result of Marx’s law of profitability in operation. I argued that mainstream economics failed to see the slump coming, could not explain it, and do not have policies to get out of the long depression that has ensued since 2009, because they have no real theory of crises. Some deny crises at all; some claim they are due to reckless, greedy bankers; or to ‘changing the rules of game’5 by the deregulation of the finance sector, causing instability; or due to rising inequality squeezing demand.6

In my view, none of these explanations are compelling. But neither are the alternatives that are offered within the labour movement. Anwar was right that neoclassical economics dominates again in mainstream economics, but I was keen to point out that Keynesian economics is dominant as the alternative theory, analysis and policy prescription in the labour movement. And, in my view, Keynesianism was just as useless in predicting or explaining crises and thus so are its policy prescriptions.

Indeed, that was the main point made in the paper I presented at another session at HM, at which Anwar Shaikh was the discussant. In my paper, entitled ‘The profit-investment nexus: Marx or Keynes?’,7 I argued that it is business investment, not household consumption, that drives the booms and slumps in output under capitalism. For crude Keynesians it is what happens with consumer demand that matters, but empirical analysis shows that before any major slump it is investment that falls, not consumption, and indeed often there is no fall in consumption at all.

Moreover, what drives business investment is profit and profitability, not ‘effective demand’. That is because profits are not some ‘marginal product’ of the ‘factor of capital’, as mainstream marginalist economics (that Keynes also held to) reckons. Profits are the result of unpaid labour in production, part of the surplus value appropriated by capitalists. Profits come first before investment, not as a marginal outcome of capital investment. In the paper, I showed that the so-called Keynesian macro identities used in mainstream economic textbooks fail to reveal that the causal connection is not from investment to savings or profit, but from profits to investment. Investment does not cause profit, as Keynesian theory argues, but profits cause investment.

Shaikh commented in his contribution that Keynes was also well aware that profits were relevant to investment. That sounds contradictory to what I am arguing. But let Keynes himself resolve how he saw it, when he says:

Nothing obviously can restore employment which first does not restore business profits. Yet nothing in my judgement can restore business profits that does not first restore the volume of investment.

To answer Keynes, my paper demonstrated that there is plenty of empirical evidence to show that profits lead investment into any slump and out into a boom - the Marxist view. And there is little or no evidence that investment drives profits - the Keynesian view.

‘Profits of enterprise’

Shaikh argued at HM that it is the “profits of enterprise” that matter, not profits as such. By this he means that the interest or rent taken by finance capital and landlords must be deducted before we can see the direct connection between business profits and business investment. Maybe so, but the evidence is also strong that the overall surplus value in the hands of capital (including finance capital) is the driving force behind investment. Interest and rent can never be higher than profit, as they are deductions from total profits made by productive capital.

Also Shaikh reckons that it is expectations of future profit on new investment that is decisive in the movement of business investment, not the mass or the rate of profit on the existing stock of capital invested. Yes, capitalists invest on the expectation of profit, but that expectation is based on what their actual profitability was before. So the profitability on existing capital is what matters. Otherwise, the expectation of profit becomes some ephemeral subjective measure, like Keynes’s ‘animal spirits’. Indeed, as I quote Paul Mattick in my paper,

What are we to make of an economic theory … which could declare, “In estimating the prospects for investment, we must have regard, therefore, to the nerves and hysteria and the digestions and reactions to the weather of those upon whose spontaneous activity it largely depends” (Keynes)?

My paper concludes that different economic policy conclusions flow from the Marxist or Keynesian view of what drives investment. The Keynesian multiplier reckons that it is demand that drives investment and if consumer and investment demand is low or falling, a suitable boost of government investment and spending can compensate and so pump prime or boost the capitalist economy back on its feet.

But when we study the evidence of the efficacy of the Keynesian multiplier, as I do in this paper, it is not compelling. On the other hand, the Marxist multiplier - namely the effect of changes in the profitability of business capital on investment and economic growth - is much more convincing. Thus Keynesian fiscal and monetary stimulus policies do not work and do not deliver economic recovery, when profitability of capital is low and/or falling. Indeed, they may make things worse.

At the plenary I pointed out that Donald Trump plans some limited form of Keynesian stimulus by government spending on infrastructure programmes worth about $250 billion - I have discussed these plans and their fake nature on a previous occasion.8 But even if they were genuine increases in state investment, it will do little. Business investment as a share of GDP in most advanced capitalist economies is around 12%-18%, while government investment is about 2%-4% - some four to six times less. That is hardly surprising, as these are capitalist economies! But that means an increase of just 0.2% of GDP in government investment, as Trump proposes, will make little difference, even if the ‘multiplier effect’ of such investment on GDP growth were more than one (and evidence suggests it will be little more).9

What matters under capitalism is profit, because the capitalist mode of production is not just a monetary economy, as Keynesian theory emphasises; it is, above all, a money-making economy. So without profitability rising, capitalist investment will not rise. This key point was the starting point of another session on Fred Moseley’s excellent book, Money and totality, which explains and defends Marx’s analysis of capital accumulation, his laws of value and profitability, from competing and distorting interpretations.10

Moseley shows there is no problem in reconciling Marx’s law of value (based on all value being created by labour-power) with relative prices of production and profitability in a capitalist economy. There is no need to ‘transform’ labour values into money prices of production, as Marx starts the circuit of production with money inputs and finishes it with (more) money outputs. The law of value and surplus value provided the explanation of how more money results - no mathematical transformation is necessary.

But this also means that for Marx’s law of value to hold and for total value to explain total prices (and for total surplus value to explain total profits), only labour can be the source of all value created. There cannot be profit without surplus value. That is why I disagree with Anwar Shaikh’s view that Marx also recognised profit from ‘alienation’, or transfer.11

The danger of accepting that profit can come from somewhere other than the exploitation of labour-power is that it opens the door to the fallacies of mainstream economics - particularly Keynesian economics - that creating money or credit can deliver more income (demand) and is not fictitious, but real, value. If that were true, then monetarism and Keynesian policies become theoretically valid options for ending the current long depression and future slumps without replacing the capitalist mode of production. Luckily, the view that profits can be created out of money and by not exploiting labour is demonstrably false.

Michael Roberts blogs at


1. See http://hmny.org/hmny-2017.

2. A Shaikh Capitalism: competition, conflict, crises Oxford 2016. See my review at http://isj.org.uk/real-capitalism-turbulent; and commentary at https://thenextrecession.wordpress.com/2016/04/04/capitalism-and-anwar-shaikh.

3. www.hgsss.org/anwar-m-shaikh-capitalism-competition-conflict-and-crises.

4. M Roberts The long depression Chicago 2016.

5. https://thenextrecession.wordpress.com/2016/03/02/changing-the-rules-or-changing-the-game.

6. https://thenextrecession.wordpress.com/2014/03/11/is-inequality-the-cause-of-capitalist-crises.

7. https://thenextrecession.files.wordpress.com/2017/04/the-profit-investment-nexus-michael-roberts-hmny-april-2017.pdf.

8. https://thenextrecession.wordpress.com/2016/11/22/a-trump-boom.

9. See https://thenextrecession.files.wordpress.com/2017/04/leeper_ltw_fmm_final.pdf.

10. F Moseley Money and totality Leiden 2015. See my review at https://thenextrecession.wordpress.com/2016/04/29/fred-moseley-and-marxs-macro-monetary-theory.

11. See https://thenextrecession.wordpress.com/2016/10/13/shaikh-at-greenwich.