Too many things wrong (Sargent and Field edition)



And not enough time to blog about all of them. Two that seem to be really important and worth noticing in recent debates around the blogosphere among the chattering classes are the idea  (subscription required) that State Defaults after the Jacksonian economic crisis were good to establish US credibility, and the notion that Total Factor Productivity (TFP) was essential for the US recovering from the Great Depression.

Very briefly I’ll discuss why these two propositions are just wrong. Sargent argues that by guaranteeing State debts the Hamiltonian system created moral hazard, and that the States defaults of the 1840s, which resulted from this arrangement, were instrumental in creating a credible fiscal commitment to sound finance. In his words: “in refusing to bail out the states in the early 1840s … the federal government reset its reputation vis-à-vis the states, telling them in effect not to expect it to underwrite their profligacy.” The lesson for Europe is let the periphery default, and, by the way, that would lead them to fiscal consolidation by even more austerity (yep he never heard of multipliers). At any rate, this account of the United States experience is pure fiction.

First of all the collapse had nothing to do with profligacy, and all to do with prices of cotton falling, and States defaulting on foreign debt, not domestic debt. In Europe the countries do print the money in which their debt is denominated, the problem is that the ECB is not willing to do it. The crisis is self-made, and if the ECB monetized a bit of debt there would be no danger of inflation, since the economies are really (really) far from full employment.

Further, the US national government at that point had no public debt (Jackson paid it down and caused a financial crash; he also required payments of public lands in specie, that is the crisis was worsened by austerity and sound money), and no national bank or monetary authority. Hence, it could not bail the States out. Only after the Civil War, with greenbacks, a more centralized management of debt and money were created in the US. So there is no possibility that the reputation of something that did not exist until the 1860s was built in the 1840s.

Sargent's anal fixation with austerity in order to pay debts, even those in domestic currency, and his lack of understanding of basic events in the history of the United States are appalling. And this guy got a Sveriges Riksbank Prize in Economic Sciences in Memory of Alfred Nobel (yep, it’s not a real Nobel!).

Field is an interesting case. The mistake in his book is not of his making, in all fairness, but the result of the profession's lack of understanding of basic economic principles. His point, well explained by Mark Thoma, is that part of the recovery in the 1930s was caused by rapid growth in productivity (TFP). Nothing against the argument, which might be (and probably is) true, to some extent. Note also that productivity is not only pro-cyclical but also structural and demand-led, which means that some of the increase in productivity was actually caused by the recovery. But the problem is that TFP is not a measure of productivity.

Note that TFP is based on the notion that there is a production function in which output (Y) is a function of labor (N) and capital (K), and forget for a second the problems of using the notion of a quantity of capital. In addition, we know that income (Y) is equal to the payments to labor (N) and capital (K). So we have a theoretical construct and an identity:

Y=f(N, K) and Y=wN+rK

Obviously if you derive Y with respect to time, you must obtain from either equation that the growth of Y over time is a function of growth in labor and capital, and either some additional part, which depends on the technology f(…) in the theoretical construct, and the weighted average of the growth of wages (w) and profits (r) in the identity. And yes the second is an identity and by definition (constructed in the national accounts) true. So TFP is a residual that says something about income distribution. Let’s please use labor productivity, when discussing productivity. For more on that see, for example, Felipe and McCombie (2001; subscription required).