Macroeconomic theory is chock full of mathiness. It’s not just Lucas and Prescott, it’s the whole scientific culture of the field.
I think you find this going all the way back to John Hicks’ famous 1937 paper, “Mr. Keynes and the Classics.”
The “i” in this model could be a short-term interest rate, or it could be a long-term interest rate. It could be a risk-free rate, or it could be a risky rate. It could be a nominal rate, or it could be a real rate.
And, as Smith points out once again, none of the equations in the IS-LM model, or any other mathematical macro model, has any demonstrated empirical validity. The equations are, at best, a way of organizing and expressing the economist’s opinions about macro.
My own opinion, as you know, is that thinking about the economy as if it were a single business (or as a single consumer who also runs a single business) is wrong-footed from the very start. Instead, I believe that it is in the shifting kaleidoscope of patterns of specialization and trade among multitudes of businesses that employment fluctuations take place.
It is fascinating to me that there are critics who will not buy the PSST story until they see it expressed using math. To me, that is as beside the point as arguing that it has no validity unless it can be told in Latin or Swahili or Yiddish.
Is mathiness stronger in macro than in other sub-fields? I was at a small conference in honor of one of the top 10-15 labor economists in the world (a very mathy one), and one of the presenters, also a very distinguished and senior figure, said something about how “Without math, we would just be throwing words around.” I think this is pervasive in economics. It is believed that math is a necessary condition for any explanation of economic concepts. Without it, we are just tossing words around (like mere sociologists or historians, it seems to be implied).
It is fascinating to me that there are critics who will not buy the PSST story until they see it expressed using math. To me, that is as beside the point as arguing that it has no validity unless it can be told in Latin or Swahili or Yiddish.
Oy.
Many people have a really hard time with “emergent” phenomena.
Hence, many, have a hard time swallowing the idea of evolution, many aspects of politics, and often, even straightforward to observe even if somewhat surprising, results in markets.
So saying you won’t believe PSST (or any like theory) until you see it in equations is like saying you won’t believe in evolution until you see it in equations, etc.
Now of course, one might raise (and I will raise right here), that it’s not obvious to me what test would prove or disprove or inform study of PSST. But that same complaint applies to a lot of economics….
I want to see it expressed in math because I think that it what is required to have legs in academia.
Robin Hanson has a recent post in which he says:
But you know, then he gives the math, and it only takes six brief paragraphs, and someone can follow along, make sure there are no cheating steps, and even learn a little bit about the exponential relationship that one would not have expected from the verbal explanation alone.
And aren’t purely verbal explanations even more fraught with ambiguity and prone to abuse than manipulations of equations?
So why complain so much about what’s not that big a deal, and which some people find better than words alone? Hanson just makes it seem like a deniable brag. “Sigh… I suppose lesser minds can’t see this trivial result from the words alone, or are so brainwashed by the mainstream academic rituals that they refuse to think without seeing some variables. Oh well, fine, I guess if I must …”
Her is why: when PSST gets taken over by the mathturbators it will be taken down the math rabbit holes just as everything else does.
So both things are true. It probably should and could be mathematize to death AND that will probably be a waste of time.
Oh, I almost forgot this. Since math is mistaken for utility and status how long will it take a new research program to achieve the sophistication on that dimension to be taken seriously in that faulty contest? Who will do it if it is harder? Will it ever achieve that level if no one has the incentives in academia to do it?
What sorts of observations in the economy would be inconsistent with PSST? Can it be falsified? Is there anything in the economy that can be measured which would indicate something about PSST? Can you connect such measures to simple numerical examples (similar to what Ricardo and Smith do)?
Economic models can serve a useful purpose if they concisely present an idealization of numerical examples. They are even better if they can illuminate connections between different simple numerical examples. Game theory is great for this (as is a great deal of microeconomics and price theory). I can understand not wanting to present a general equilibrium account (equilibrium in economics always struck me as a metaphysical claim rather than connected to anything real). But, I think there is a role for math here, so that we can give impersonal indicators that test ideas (rather than waffling like Krugman’s 2013 “market test”), or using consensus of economists as a measure (the “academic market test” Blanchard brought up with respect to DSGE models).
If you want something more substantial for PSST than just so stories and a pastiche of growth economics, you really need to focus on the other half of macro, recession and unemployment, which have little to with innovation and growth directly, and is not a story of entrepreneurship but of retreating and failing enterprises, large and widespread, and why this can go on for so long and become so large. In this it seems PSST is even more primitive than the gdp factory, saying something was or became unsustainable and went out of business when the questions are why, how, when, and wherefore, to what magnitude. These may all be idiosyncratic, but even then it should be possible to say something. Not everything fails, not everything ceases, not everything is crop yields and sunspots, not everything recovers at the same rate, and not everything recovers entirely.
My exposure to PSST is limited to the descriptions you’ve provided on this blog. Based on that level of understanding, it seems to me that simulating the economy as some sort of genetic optimization problem with aperiodic, pseudo-random, changes in both the boundary conditions and the fitness function would demonstrate the PSST concept in a way that would satisfy those looking for a mathematical basis.