In Defense of Economic Theory: Navigating the Empirical Revolution and Philosophical Engagement

The podcast discusses a significant shift in economics from abstract theorizing towards empirical work, a trend the speaker initially supported but now views with caution. The core concern is the rise of "atheoretical empirical work," which lacks a necessary interpretive framework, leading to reliance on potentially flawed "common sense." This evolution necessitates a deeper engagement with the philosophy of science, which itself has become more empirical and naturalistic. Philosophers of science have become increasingly interested in economics for two main reasons. Firstly, the challenges of modeling complex systems like climate change, which involve mathematical models with numerous untestable assumptions, mirror those found in economic modeling, prompting questions about epistemic warrant. Secondly, the "rationality wars" sparked by behavioral economics, particularly the work of Kahneman and Tversky, have intertwined philosophical decision theory with economic discussions on individual rationality, exploring both its positive (predictive) and normative (corrective) aspects. The field of economic methodology has specialized, leading to a division of labor between historians of economic thought (focused on archival work) and philosophers of economics. While this specialization can create challenges for interdisciplinary work, it also fosters deeper engagement. Surprisingly, despite the dismal failures of macroeconomics and finance in predicting or recommending policies during crises, there hasn't been a significant surge of methodological papers addressing these failures. Instead, the primary impetus for methodological discussion has come from behavioral economics, neuroeconomics, and the broader empirical revolution, including data-driven approaches and randomized control trials. The speaker notes that the impact of the recent financial crisis on economic methodology has been far less profound than the Great Depression's impact on Keynesian economics. The current focus on data-driven and experimental economics, including behavioral and behavioral welfare economics, represents a broad empirical turn. The long-term implications of this shift, particularly regarding the potential for "economics without theory" and the role of technological change (big data, fast computers) in shaping economic practice, remain to be fully understood and will require future historical and philosophical analysis.

if you ask me 20 years ago tell me one simple thing that's wrong with economics I would say it's too much abstract theorizing and not enough looking at the real world not enough data we need to be more empirical

you can't do a theoretical empirical work you have to have an interpretive framework Kuhn was right

if you don't have an interpretive framework that's informed by mainstream economics or some heterodox program you're going to just be putting common-sense on it

these models cannot be tested in a traditional way because you're talking about what's going to happen way out in the future with a whole lot of variables

what philosophers of science discovered is wow they look like economic models they're mathematical they have a lot of assumptions nothing works unless the assumptions are true and yet these are models that we trust and we believe in and we're willing to do policy on

behavioral economics has set off a fire in terms of the rationality of individual agents

Kahneman and Tversky talk about normative standards right like in libertarian paternalism if you're if you're making a mistake you need to be corrected well the mistake is that you're not behaving rationally according to expected utility theory

that old approach of sort of going and finding your favorite philosopher of science and then taking that position and just applying it to economics without thinking about what economists do that's different than physics or whatever that's not what's going on

given what a dismal failure of macro and Finance were in terms of either predicting or recommending policies you would think that people would run to economic methodology to figure out what happened and why did the science fail

I expected this would have a profound impact on how we think about economic science certainly the Keynesian a Great Depression had a profound impact on how we think about economic science but it really didn't

there is a technological component I mean you know big data and fast computers and all this stuff is important and economists you know have traditionally argued that technological change matters right and then technological change may matter in how we do economics as well