Regular readers of mine will probably guess – correctly – that I disagree with his mathematical framework. But I will discuss my complaints with the mathematics elsewhere. For now, I will take his results as given, but with a disclaimer that they appear to be applicable to a narrower set of model economies than John Cochrane suggests – probably much narrower.
Valuation Formula For Government Debt
The core of his paper revolves around what he refers to as the valuation formula for government debt. Written out, it is:
This one formula wraps together most of the interesting concepts for bond market economics. It provides a means to value government bonds, it tells us about inflation and fiscal policy. It is a strong result, which is surprising because it is not heavily referenced. If I had confidence in the mathematics behind it I would refer to it quite often.
Using this valuation formula he was able to examine the results of various DSGE models, and show that fiscal policy is actually driving the economy, and not interest rates as is normally assumed. This presents a challenge to the monetary policy centered interpretation of DSGE models that is made at central banks.
I had considered creating a simple idealized model of how money was introduced in European colonies, based on the verbal description in Chapter 3 of the text "Understanding Modern Money" by L. Randall Wray. My model would be cruder version of the models used by John Cochrane to derive his valuation formula. The idea is that money is forced onto an existing non-monetary economy by the imposition of a tax liability. The requirement to pay taxes with money gives money value, even the model entity has no other use for it (that is, money does not appear in the utility function).
The Post - Keynesian response
Warren Mosler has an extended quote from Mario Seccareccia, a noted Post-Keynesian economist, discussing the Cochrane paper on his blog here. His more positive comments are:
But it is very interesting that he is saying many of the same things that we have been saying for quite some time. On the question of monetary policy, as you know, here we have been paying interest on “reserves” (or more correctly “settlement balances” in Canada, which is actually a more correct vocabulary than “reserves” … since the latter is perhaps suggestive of these “reserves” backing “inside” money, which is nonsense!) for over two decades, and this institutional change in the early 1990s was hardly because of technological change! It was to conduct more efficiently monetary policy when they realized that the control of monetary aggregates was completely futile. In much the same way, the whole issue of controlling the price level via fiscal policy is pure Abba Lerner!!
However he also notes that John Cochrane completely avoids citing any relevant post-Keynesian academics in his working paper, despite the obvious overlap of content. Since I am a blogger, I do not have a dog in the academic citation fight. But as an ex-academic, I find the behavior of mainstream academic economists troubling. Ideally, academic work should take the long view and honestly take into account the full intellectual heritage that exists.
But there is a certain amount of symmetry here. The Fiscal Theory of the Price Level is itself largely ignored by a significant portion of the mainstream, based on my limited knowledge of the recent literature. This is despite the fact that the results are powerful. Since the theory appears no more incorrect than any other DSGE model results, the only reason I see for the lack of acknowledgment is ideological differences.
But returning to Mario's comments, I will note that there is a large gap between Functional Finance (associated with Abba Lerner) and the Fiscal Theory of the Price Level. The Fiscal Theory of the Price Level argues that the current price level is driven by the asymptotic trend in expected primary fiscal balances. Since those expected balances are completely unobservable, there is no empirical way of testing the theory. Functional Finance is closer to the mainstream concept of an output gap driving current inflation, which is a testable concept. Current fiscal policy matters as it will affect the amount of slack in the economy; the long-term path of fiscal balances does not matter.
(I also note some other recent articles on this topic that I came across below.)
I have other article which will look at the derivation of the bond valuation formula, and the mathematical framework used. They are:
- An article describing how the price level determination works.
- A discussion of monetary frictions within the model framework.
But if you are more accepting of the DSGE framework, this paper provides a challenge to the tight focus on monetary policy within those models.
- Nick Edmonds discusses the article on his blog, and draws parallels with Stock-Flow Consistent models. He discusses the interesting behaviour of inflation that I skipped over in my discussion here. Raising interest rates increases inflation within the framework, the so-called "Neo-Fisherite" view.
- Nick Rowe derives the same results using the Quantity Theory of Money. I just found the article as I was about to submit this one, so I did not have time to go over it in detail. It looks interesting based on a quick read.
- My "Theme Article" which provides an overview of some of my complaints with the mathematics and economics of DSGE models. The specific problems I see with the Bond Valuation Formula have not been covered directly, but there is some overlap. The derivation method he uses to derive the Bond Valuation Formula appears to side-step the issue of term premia, but it raises other problems.
(c) Brian Romanchuk 2014
This comment has been removed by a blog administrator.ReplyDelete