Misconceptions about Heterodox Economics in general and Sraffian in particular

In fact, it is my view that, in general, not only heterodox groups, which by definition tend to be small and often concerned with specific topics, but the mainstream is quite fragmented. On macro issues, even with the New Neoclassical Synthesis, there is a lot of disagreement between New Keynesians and RBC types, particularly on policy issues. So it should not be surprising that heterodox groups are quite fragmented too. Sometimes the fragmentation within the mainstream gives the false impression than some groups are breaking away, or on the edge (see my view on that here). And it is not just fragmentation, but also confusion. Even in the mainstream there is lack of understanding about the meaning of the mainstream (see here for example).

So it should not be surprising that, given the fragmentation of the heterodoxy, several misconceptions arise. Sraffians are particularly vulnerable to this since Sraffa wrote so little, and even though his Production of Commodities is not a difficult book to read, it is one that has been often misunderstood. Marc Lavoie has responded, from a Post Keynesian and friendly standpoint, to some of these misconceptions (the fact that some people consider Austrians heterodox, but are not certain about Sraffians speaks volumes about confusion among heterodox groups too).

Yet, I was a bit surprised by some of the misconceptions in John King's book. He says that: "There is no role in Sraffian models for fundamental uncertainty, money or the principle of effective demand." In the quote he suggests that this is the position of Hart and Kriesler (2014), but if he disagrees he does not say anything. So it is safe to assume that he concurs. He complements this by arguing that: "since the relationship between the wage rate and the rate of profit in Sraffian models is monotonically declining, it is difficult (if not impossible) for the their models to incorporate a 'wage-led' growth regime." Hm, were to start.

So Garegnani pointed out long ago (in the 1960s, but the English publication was in the late 1970s; links to the English versions of both the 1960s and 1970s papers here) that not only Sraffians believe in effective demand, but that a coherent presentation of the Principle of Effective Demand requires the abandonment of marginalism (particularly the marginal efficiency of capital argument). Not only that, Garegnani is very clear in his 1960s papers that (from link above):

"As regards consumption, increases in real wages lead to a rise in consumption and hence, provided the economy has accumulation capacity that is not fully utilized, to an expansion of the productive system and to an increase in employment. Given the level of productivity in the economy, the increase in real wages will in fact cause a redistribution of income in favour of a class that consumes a major portion of its income, and with that an increase in the first component of final demand... 

a steady and continuous rise in real wages along with the consequent steady and continuous increase in consumption can serve to instil in entrepreneurs a confidence in the continuous expansion of the market for their products, inducing them to undertake investments and increases in employment and output that will in turn help to raise final demand."

So yeah the economy is wage-led. Actually, in Garegnani's debates with Marxists, with Joan Robinson and other authors on the causes of long run growth, he could be seen exactly as taking the position that profit-led regimes are not possible, in the sense that he understood that firms would not invest because of a higher rate of profit. Firms are interested in adjusting capacity to demand, and to maintain a normal level of capacity utilization. Also, one should note that Garegnani wrote his paper in the 1960s under the direct influence of Sraffa, who might also have understood the idea of the accelerator. That is why Franklin Serrano is correct in suggesting the supermultiplier is Sraffian. Not only Sraffians have effective demand, they do in the long run (not in the short run as a result of imperfections).

Two things are important in this context. First, in Sraffa's price equations, which uses the method of given quantities developed by classical political economy authors, imply that there is class conflict, and an inverse relation between the real wage and the rate of profit. However, in a theory of the determination of output in the long run, output by definition is not given and the effects of income distribution on output might be ambiguous, even if the demand regime is wage-led (for example, higher wages might lead to loss of external competitiveness, and lower exports than more than compensate the increase in consumption associated to higher wages). Second, as I noted on my previous post on Garegnani's 1960s paper, the Sraffian project was the revival of the classical theory of distribution, concomitantly with the extension of the Keynesian Principle of Effective Demand to the long run.

On the absence of money in the Sraffian system also a lot of ink has been wasted. In his Production of Commodities Sraffa famously suggests that it is the rate of interest, as determined by the monetary authority, which is the exogenous variable. So prices and real wages are determined for given technical conditions of production and the long term interest rate as set by the monetary authority. Pivetti referred to this as the monetary theory of distribution. So monetary policy has important distributive implications, and this view is perfectly compatible with Keynes' views on a normal, conventional and not psychological, rate of interest. It is also compatible with endogenous money views, that hark back to Tooke and other Banking School authors. There is money, and not in the sort of Monetarist way in which the central bank controls its quantity.

Finally, on uncertainty. True for Sraffians uncertainty is not central for unemployment (one in three), yet that does not mean is completely irrelevant or that it does not play any role. Here it is important to note that any good discussion of uncertainty suggests that the one that is central is the uncertainty about future demand. See for example Davidson in this example. So the problem is lack of demand. Autonomous demand that is. And that works, as it should, with the supermultiplier.