Rather than chase after hundreds of papers that I think are uninteresting and will be rapidly made obsolete by the ongoing wave of research, I would rather focus on a niche area that I think is interesting. Roger E. Farmer has been publishing models for some time that fit within the broad neoclassical tradition, yet have some similarities to post-Keynesian thinking. The book Prosperity for All (Amazon affiliate link) gives a readable introduction to his work.
The book by itself is interesting for heterodox readers in that it discusses the strengths and weaknesses of the DSGE methodology from an insider perspective. A lot of heterodox critiques are based on literary criticisms that were essentially developed decades ago.
I do not want to go too far into the book, since I will probably write a summary that would be turned into a section in my manuscript.
Although I have some concerns about Farmer's results, this article will just outline why I think his approach -- or a similar approach -- is somewhat interesting.
The Advantages of more Respect for Mathematics
If I were a student researcher, the main attraction of Farmer's work is that it is not obviously wrong like standard DSGE macro, yet it is close enough from a methodological basis that one can pass as a mainstream economist. Since I am not exactly concerned about applying for jobs at central banks at this point in my life, I am indifferent to this point, but it certainly factor in my thinking if I were in that position.
My view is that mathematical economic theory runs into brick walls, and we cannot expect to develop models that do what people demand that they do. If forecasting the economy is an inherently impossible task, then we should not expect people to develop better models than existing ones -- they will all fail, just in different ways. That said, we might be able to find nooks and crannies where mathematical models do a decent job, which will be enough to keep people busy.
My interpretation of post-Keynesian economics is that they have correctly described most of the fundamental issues with mathematical approaches. The problem is that once that is done, there is not a lot more to say. Academic output is literary, and one can debate how much forward progress there is -- which is a problem for all mature academic fields in the publish-or-perish era.
Instead, you need to look at the current mathematical economics research, and explain why it in particular does not work.
The Demise of NAIRU
Prosperity for All has a long list of critiques of the dominant New Keynesian paradigm, but one of the core issues is the natural rate of unemployment. (Depending on the model, this might be replaced by an output gap that is assumed to revert to zero.) In order to get the models to fit the data, it has to continuously adjust. He argues that this makes the models unfalsifiable (which is exactly my view).
He instead has a nonlinear model which allows for multiple equilibria. The result is that nominal GDP tends to stick near its previous level, and not revert to trend. This allows for the economy to be mired in unemployment, and not return to full employment through the some attractive force that is allegedly part of capitalism.
My view is that economists of all persuasions need to give up on the pseudo-science that has sprung up around the concept of equilibrium. If you write down a model of a system, it either has no solution, one solution, or multiple (possibly infinite). This is true of any class of models. Mathematics is the study of sets, and the logical properties of sets. There are no little people living inside the system moving prices back and forth and coming to an "equilibrium." In this case, if we drop the equilibrium nonsense, all that has happened is that Farmer has a different set of equations, with the solution having different properties. There is no reason to pretend that the nature of equilibria has anything interesting value to add.
The resulting models address many of the issues that post-Keynesians gripe about. Getting rid of NAIRU eliminates many of the existing arguments about how to manage the business cycle from a practical perspective.
Farmer's models offer a plausible explanation for recessions. This is unlike standard DSGE models that are close to the real business cycle/New Keynesian core: they are arbitrage-free pricing models that give forward prices that always converge to a stable steady state. Like interest rate forwards, the model trajectories are always smoother than realised data. By contrast, the "multiple equilibria" allows for jumps in GDP.
The reason is that there is a belief function at the core of the model. Farmer refers to this as "animal spirits." My argument (which I developed before hearing of Farmer's work), also is that "animal spirits" is a core hidden variable in realistic macro models. Since that is how interpreted Keynes (and Minsky), the overlap is not a surprise, since there is a common root.
I am not entirely convinced by some of Farmer's arguments about this animal spirits model, but my view is that it represents the best explanation of the properties of recessions. My argument is that we can get to the same place using stock-flow consistent models, without bringing in the ideological baggage of neoclassical methodology. However, I need to look into his methodology in more detail (I have scanned some papers) before making that assessment for my book.
It would be unsurprising to get qualitatively similar results starting from the SFC methodology or Farmer's with respect to modelling recessions.
I expect to write a cycle of articles about Farmer's methodology. I might start with Prosperity for All, since that would be the best starting point for a general reader. I will then attempt to discuss the methodology from some of his journal articles. Although I do not want to spend too much time moping about weaknesses of the neoclassical methodology in my book, I might outline some of the points made as well.
(c) Brian Romanchuk 2020