Hi friends (but mostly Nikita and Alli because this might be something we want to consider),
First, please excuse the stream of consciousness (I promise I tried to keep this organized).
Second, when I refer to modern macro as being “broken” I’m referring to it being broken with regards to its ability to predict/explain behavior and to inform policy decisions, not its ability to teach principles/intermediate students.
Anyway, I did some thinking (gasp) after last class and I think I have an important question with which we need to grapple when we try to define “broken.” That is, I think we need to draw a distinction between two different types of “broken”-ness.
Popperian Broken means that mainstream macro models were unable to explain or predict important phenomena (i.e. the Great Recession). A hard science example of this might be like how existing theories of biology/aerodynamics can’t explain why a bumble bee is able to fly or how physics can’t understand what glass is. They’re important questions, but it’s probably safe to say that we don’t need to radically overhaul these disciplines to understand them, we probably just need to make some revisions.
I think that Popperian Brokenness means macro as a discipline isn’t truly *broken*, but rather that it just needs some polishing. So far, we’ve really only pointed out ways that macro suffers from Popperian Brokenness – DSGE models are oversimplified, don’t have a financial sector, etc. All of these could probably be fixed by adding more and more equations into the models as computing power gets stronger and none of them necessarily imply that we need to overhaul our understanding of macro.
Kuhnian Broken means that mainstream macro models are paradigmatically wrong and macroeconomists are headed in completely the wrong direction. The hard science equivalent of this would be the geocentric models of the universe – our explanations are totally wrong.
I’m still trying to establish a threshold for considering macro to be Kuhnian Broken. The best I can come up with would be a fundamental upending of the micro-founded assumptions of modern macro models – i.e. that consumers don’t seek to maximize utility, firms don’t try to maximize profits, etc. I’m also not sure whether information failures (people act rationally based on incomplete information) count as Kuhnian or Popperian Brokenness, but I think that it depends in part on whether those information failures can be solved by adding equations.
Of course, I also realize that these two probably aren’t mutually exclusive – Popperian Brokenness can be a symptom of Kuhnian Brokenness. Retrograde planets could be considered Popperian Broken but they ended up proving that the geocentric model of the universe was Kuhnian Broken. That said, this still raises the question of how we would know whether macro suffers from Kuhnian Brokenness or just a few instances of Popperian Brokenness, which seems tricky. Loathe as I am to concede Alli’s point that economics is a soft-science, she has a point that it’s hard to make a definitive argument in macroeconomics. Astronomers can point a telescope into space and say “this planet is here” to disprove the geocentric model of the universe (OK maybe not that simple, but I only took 100-level astronomy). Economists rarely have access to good experimental data, especially in the study of macroeconomics. Instead we have to interpret regression results using often imperfect variables and proxies.
I thought Alli made a pretty insightful counter-point to this whole discussion yesterday. She said that in order for macro to be “broken” it had to have worked at some point – Classical Economics worked until the Depression, Keynesian models worked until Stagflation in the 1970s, etc. This point would imply that Popperian Brokenness is the only type of broken that we should consider because if we found that macro was Kuhnian Broken (and that macro never really worked), then that would just mean that macroeconomics is a pseudo-science and was never worth pursuing in the first place.
Finally, I realize that this whole meta-rant might have been better suited for the very beginning of the class, but this just now came to me in a semi-coherent form. Looking forward to hearing others’ thoughts on this