Wednesday, March 2, 2016

Unravelling Some Occult Mysteries of the Heterodox



Noah Smith has a new post, "Occult Mysteries of the Heterodox," criticizing Heterodox economists who seem unable to provide a simple explanation of their new, and supposedly superior, methodologies. “They show every indication of having no new methodology whatsoever.”  And later, in a "tweetstorm," Noah accuses Post Keynesians, in particular, of only being interested in methodologies that yield their favored conclusions.

In contrast to the the PKs, who offer nothing but "criticism," Noah points to “the Solow Growth model” as a useful and accessible scheme of analysis.  I’ll get to alternatives in a moment, but I think Noah is much too blas√© about the importance of “criticism,” and his reference to “the Solow Growth model” provides a good illustration of the problem.

Leaving aside the Cambridge (UK) critique of aggregate production functions (the mere mention of which brands one a hopeless ideologue), here’s a brief paper by Jesus Felipe and Franklin M. Fisher, who carefully explain the conceptual difficulties with aggregate production functions and then patiently respond to all the justifications offered by those who still use these models despite their drawbacks.  

Some defenders of mainstream economics brush off critics, claiming they just don't know the math.  Well, Kenneth Arrow and Robert Solow, who do know the math, haven't been impressed by the modeling inspired by Lucas and Sargent.  Perhaps Stephen Williamson is right to dismiss these critics because they “fail to understand the power of the work they did,” but this claim requires an argument that addresses the actual criticisms offered by Arrow and Solow, not to mention those of Frank Hahn, Franklin M. Fisher, and several other first-class economists.

Noah is mistaken when he says PKs simply favor the methods that will produce their favorite conclusions.  I think, in opposition to Noah's view, that PK approaches and methods arise from their criticisms of orthodoxy.  

For brevity’s sake, I’m going to mention one assumption widely held among PKs, which is that neither market participants nor economists “know the data-generating process.”  Awhile ago, Brad DeLong, commenting on a Lars P. Syll post, granted that there may be something worthwhile in this point of view, but asked, “what kind of economic arguments do we make” once this assumption is accepted?

Here’s a short list of the sort of practical ramifications of assuming that neither market participants nor economists know "the data-generating process."

1. First, let me note that our ignorance is manifest in the disagreement among economists about the causes of the Financial Crisis and the Great Recession, the effects of QE on interest rates and inflation, etc.  Such disagreement gives rise to many, often inconsistent, beliefs about the economy, and this diversity of beliefs must be taken into account in thinking about the economy.  Isn't it worth modeling economies in which market participants hold conflicting views of the future?  Wouldn't such an economy behave differently from one in which everyone shares the same rational expectations?  Mordecai Kurz has done interesting work in the area. And agent-based modeling (Santa Fe style) may prove useful in thinking about interactions among agents with different “views of the world.”

2. If you don’t know how “the data-generating process” works, or if there is no such thing, then history, path dependence, and hysteresis should play a larger role in thinking about the economy.

3. Since agents don't possess a complete list of states and their probabilities of occurrence, many people rely on verbal models, narratives, and stories to guide their economic decision making, and the role these informal “models” deserve examination.  See some of Robert Shiller’s recent work as well as this paper, which finds the EMH wanting once "the news" is taken into account.

4. Disequilibrium micro foundations (and mutually inconsistent expectations) attracted a lot of first-class economists from the mid-1950s until the early 2000s, including Hicks, Debreu, Hahn, Samuelson, Clower, Leijonhufvud, Negishi, and several others.  Arrow, himself, thought of the macroeconomy as "a disequilibrium phenomenon," and this vantage point is worth another look.  R. Backhouse and M. Boianovsky wrote a good book on the subject.

It may be objected that the economists mentioned above are not Post Keynesians, or at least do not identify themselves as such.  I concede this point because I’m concerned with methods and approaches that avoid some of shortcomings of mainstream thinking, rather than with the purity of Post-Keynesian economies.

No comments:

Post a Comment