The other day I commented on a piece that was run in the NYT on Wynne Godley and other Levy Institute scholars. Since then Paul Krugman has weighed in on the debate and Matias Vernengo has responded. Even though I’ve been known to be somewhat harsh on Krugman I think that the piece he wrote actually contains the seeds of a constructive conversation — unlike his typical approach to heterodox economists who are still alive, which is to dismiss them out of hand and ignore them. Krugman, it seems to me, is only comfortable debating the dead; not a particularly difficult task, mind you.
First of all, however, it should be noted that many of the errors that Vernengo points out Krugman as having made are indeed rather egregious. Krugman’s characterisation of what he refers to as “hydraulic Keynesians” as relying on a stable consumption function — that is, that consumption will rise and fall in line with income in a stable fashion — is entirely false. I have seen this mistake made many times before. In the General Theory Keynes lays out this argument, but it is clear from the context that it is a ceteris paribus condition that should be subject to empirical scrutiny (although Keynes mistakenly does say that this a priori condition can be relied on with “great confidence”). Here is the passage in the original:
Granted, then, that the propensity to consume is a fairly stable function so that, as a rule, the amount of aggregate consumption mainly depends on the amount of aggregate income (both measured in terms of wage-units), changes in the propensity itself being treated as a secondary influence, what is the normal shape of this function? (GT, Chapter 8, III)
As we can see, this really is just a ceteris paribus argument laid out to make a more general point. And as Vernengo correctly points out the Keynesian economist James Duesenberry updated this argument with his relative income hypothesis which is by far superior to Friedman’s permanent income hypothesis as championed by Krugman. (I should note in passing that I am currently waiting on data from the Post-Keynesian economist Steven Fazzari on consumption by income group that I have promised FT Alphaville I will write a post for them on. The data, from what I have seen, provides interesting insights into Duesenberry’s hypothesis. Watch this space).
Krugman’s other mistake is to discuss Godley’s work as if he adhered to the old Phillips Curve and was thus proved wrong by the inflation of the 1970s. As Vernengo points out Godley came from the Cambridge tradition which, in contrast to the neoclassical-Keynesians in the US, held inflation to be primarily wage-led. From this statement it is crystal clear that Krugman has not read any of Godley’s work (which leads one to wonder what gives him authority to pass comment on it). For example, in the book Monetary Economics, co-authored with Marc Lavoie, Godley devotes a whole chapter to inflation which they introduce as such:
Three propositions are central to the argument of this chapter. First, as we are now describing an industrial economy which produces goods as well as services, we must recognize that production takes time. As workers have to be paid as soon as production starts up, while firms cannot simultaneously recover their costs through sales, there arises a systemic need for finance from outside the production sector. Second, when banks make loans to pay for the inventories which must be built up before sales can take place, they must simultaneously be creating the credit money used to pay workers which they, and the firms from which they buy goods and services, find acceptable as a means of payment. Third, we are about to break decisively with the standard assumption that aggregate demand is always equal to aggregate supply. Aggregate demand will now be equal to aggregate supply plus or minus any change in inventories. (p284)
Inflation under these assumptions does not necessarily accelerate if employment stays in excess of its ‘full employment’ level. Everything depends on the parameters and whether they change. Inflation will accelerate if the value of [the reaction parameter related to real wage targeting] rises through time or if the interval between settlements shortens. If [the reaction parameter related to real wage targeting] turns out to be constant then a higher pressure of demand will raise the inflation rate without making it accelerate. An implication of the story proposed here is that there is no vertical long-run Phillips curve. There is no NAIRU. When employment is above its full-employment level, unless the [the reaction parameter related to real wage targeting] moves up there is no acceleration of inflation, only a higher rate of inflation. (p304)
So why did hydraulic macro get driven out? Partly because economists like to think of agents as maximizers — it’s at the core of what we’re supposed to know — so that other things equal, an analysis in terms of rational behavior always trumps rules of thumb.