The Incomplete Revolution

Abstract

This chapter deals with Keynes’s critique of marginalist theory. Keynes questioned the idea that market flexibility leads to full employment. He regarded aggregate demand as the main determinant of production and employment levels. For Keynes, as for Marx, increasing inequality in income distribution depresses aggregate demand. He believed that monetary policy could have real effects on the economy, but was hampered by distrust of the financial markets, and regarded fiscal policy as a necessary tool for full employment.

Keynes also believed that saving was not the driver of economic growth, but rather a factor that depresses aggregate demand. In his theory, investment depends on entrepreneurs’ expectations of future aggregate demand. Keynes’s main theoretical contribution lies in the idea that investment determines saving, and not vice versa as in mainstream economics. This theoretical innovation has important implications for the role of banks. According to Keynesian theory, banks do not intermediate saving, but create money out of nothing to fund investment. This view is called endogenous money theory. Finally, we examine the limits of Keynesian criticism of marginalism and how, by integrating Sraffa’s criticism with that of Keynes, a complete alternative to marginalism can be developed.

Appendices

Appendix

4.1.1 Of Bread and Fishes

Let us try to understand the logic of the Keynesian multiplier also in relation to the creation of credit by banks. We assume (Ricardian vice) an initial investment of €100 (or a million or a billion, whatever you like) made by entrepreneurs on the basis of purchasing power created by banks (see Table 4.1). The bank finances the firms by opening a deposit (first item column 5). The production of investment goods gives rise to a production and a corresponding income to households of €100 (first item in column 2). In period 2, this income is partly spent on consumption and partly saved. Assuming that households spend 80% for consumption (and save 20%), their spending gives rise to a production of consumer goods and the corresponding income of €80. Households employed in the consumer goods sector, therefore, receive €80 of income, of which they spend 64 (80%) on consumer goods and save the other 16 (20%). The production of consumer goods increases by 64. The households receive another 64 of income which they partly consume and partly save, giving rise to another round in the Keynesian merry-go-round. This shows that for subsequent rounds, from an initial expenditure for investment of 100, a product or final income of 500 is generated, which was partly spent on consumer goods (400) and partly saved (100). The saving generated is exactly equal to the initial investment. The banks initially generated credit out of nothing by creating a deposit of 100 in favour of the investor, i.e. lending without having first received saving as deposits (Keynes named this “initial finance”). At the end of the multiplication process, this deposit (which changed hands gradually as payments were made) consists of 100 in saving (named by Keynes “final finance”). In the end, we can say that the saving deposited matches the initial loan, but only ex post. As the multiplier process unfolded, the deposit always remained at 100, partly consisting of incomes and a gradually increasing part of saving. For example, in period 2, the 100 of income earned in period 1 is partly spent for consumer goods, so that 80 of the deposit changes hands, and 20 is saved; overall, however, the deposit remains at 100. The process stops exactly when the whole deposit consists of saving.

From the example we deduce that ex post, what we asserted for national accounting is confirmed, i.e. that investment is equal to the saving made during the year. While it is therefore right to say that investment is financed by the creation of bank credit out of nothing, it is also correct to say that ex post the investment and saving taking place in the economy are equal. Thus analysis of a monetary economy seems to confirm and strengthen the saving–investment relation that can be deduced from the Keynesian multiplier, namely that the latter generates the former.

In the example, we presume that production decisions follow the manifestation of demand, as when we go for a haircut or the restaurant prepares us a meal. However, in many sectors, such as manufacturing, firms undertake production on the basis of expected demand, so that we find the products already on the shelf when we want them. But if demand is less than expected, firms are forced to accumulate stock, which is a form of involuntary investment, as we saw. It is evidently a temporary situation. In the next period, if the lack of demand is expected to persist, they will reduce production, adjusting it to the demand ultimately determined by entrepreneurs’ investment decisions, via the Keynesian multiplier.

Further Reading

The most complete biography of Keynes is by Robert Skidelsky (2004). In recent years Skidelsky has been a companion in disputing European policies. Keynes and Sraffa’s essay on Hume and Smith can be found in Hume (1938). The story narrated in the booklet probably also intrigued and inspired Ludwig Wittgenstein, who at the time was engaged in an intense exchange of ideas with Sraffa in Cambridge, and was also a passionate reader of Agatha Christie, see Lucia Morra (2019).

Keynes’s first biographer was Roy Harrod (1900–1978) (Harrod 1951). At the end of the 1930s, Harrod was also the initiator of modern growth theory. His was an attempt to extend Keynes to the long run. Keynes had in fact stated that in the short period, when productive capacity is given, the degree to which it is used depends on aggregate demand. In the long run, however, productive capacity (or capital stock) increases. The question is therefore what determines this increase. In the late 1930s, Harrod came up with a model as simple as it was intriguing, which was neither Keynesian nor marginalist. So much so that both the orthodox and the unorthodox growth theories departed from it. A good introduction to the theories of growth, albeit a little dated, is Hywel Jones (1976). A harsh judgment on recent marginalist theories of growth can be found in Cesaratto (1999). Robert Solow, founder of the modern neoclassical theory of growth (and “Nobel Prize” winner), appreciated this criticism of mine.

The state of the art in the unorthodox field is in Cesaratto (2015). Also as a result of this paper, the “supermultiplier” initially proposed by the Swiss (Sraffian) economist Heinrich Bortis, a former professor at the University of Fribourg (Switzerland), and Franklin Serrano of the Federal University of Rio, is slowly but surely gaining ground among heterodox growth theories. An introductory text on the “supermultiplier”, which also deals with “technological unemployment”, is in Cesaratto et al. (2003). The supermultiplier takes up the idea, implicit in mercantilism, of foreign markets as an outlet for the surplus examined in the first chapter. This idea, developed by Rosa Luxemburg (1871–1919), who called them “external markets”, was taken up by Michal Kalecki, who included state spending as an absorber of surplus. The supermultiplier includes autonomous consumption in the category “external markets”. In a beautiful essay, Kalecki (1971) compares the solutions of Rosa Luxemburg and the Russian economist Mihail Tugan-Baranowsky (1865–1919) to the problem of surplus realization, i.e. its actual sale on the market. We have just mentioned Rosa Luxemburg’s work. Tugan-Baranowsky argued that in theory, capitalism would not suffer from lack of demand as long as capitalists systematically invested all their savings (Kalecki assumes that workers’ saving is negligible).

Invested in what, Prof? In industrial machinery, says Tugan. But machines to do what? To produce other machines. Kalecki comments that this implies economic planning, which of course capitalists don’t want. In a famous aphorism, the Polish economist writes that “capitalists do many things as a class, but they don’t invest as a class.” Kalecki concludes, however, that although the hypothesis of production of machines by machines is only valid in theory, it indicates a bitter truth about capitalism: that its ultimate goal is not the production of consumer goods and the material well-being of the community. In principle, producing machines to produce other machines is just as good.

When a student applied to Garegnani to supervise his undergraduate thesis on a Keynesian topic, Garegnani suggested he study Alvin Hansen (1953). An old but unparalleled Keynesian macroeconomics manual, with a comparison between marginalist macroeconomic theory and Keynes’s theory, is Ackley (1961). Ackley also published Keynesian studies on the Italian economy. One report has striking similarities with a contemporary study by Garegnani, both conducted for the same institute in Rome. I write about it in Cesaratto (2020), in a Festschrift dedicated to Marc Lavoie and Mario Seccareccia, two distinguished heterodox economists.

A useful critical review of recent mainstream macroeconomics can be found in Antonella Stirati (2016). The complementarity of Keynes’s and Sraffa’s critique of marginalism was initially identified by Garegnani (1978–1979). In a post on his blog in the New York Times, Krugman (2014) reveals the “dirty little secret of monetary policy”, namely that “it normally acts through the housing sector, with little direct impact on productive investment”. The important thing is not to say it in textbooks, right Paul?

But why don’t they want to say it, Prof? Because they are very keen to say that saving and therefore thrift determine investment. Besides acclaiming sacrifice (parsimony is a bourgeois virtue), this provides a justification for the unequal distribution of income.

The fact that public debt is not a problem when the interest rate is less than the growth rate was recently “rediscovered” by Oliver Blanchard (2019). Take a look at the first pages, the rest is “vulgar economics” based on the concept of marginal capital product and natural interest rate. Luigi Pasinetti (1998) had already written about this at the end of the nineties.

The example in the Appendix to this chapter is inspired by Dalziel (1996). Two important economists, Claudio Borio and Piti Dysiatat (2011)—chief economists at the Bank for International Settlements, Basel and the Central Bank of Thailand, respectively—clearly support the narrative of the Appendix when they write (evoking similar passages from Keynes): “The true constraint on expenditures is not saving, but financing. … And it is only once expenditures take place that income, investment and hence saving, are generated” (ibid., p. 7). Borio and Dysiatat are exemplary marginalists, so for them the essential thing is that financing takes place at the natural interest rate. We of course disagree with them on this aspect.

But then, Prof, endogenous money and the Keynesian multiplier can be incorporated into marginalist macroeconomics. Yes, of course. In fact, dissent with the marginalists has shifted to the existence of a natural interest rate or a full-employment equilibrium wage and so on, as suggested by Sraffa.

Finally, Augusto Graziani (1990) suggested that the creation of bank credit is to finance production decisions, not investment. Cesaratto (2017) tries to reconcile the two points of view.