Vox has the article The radical plan to change how Harvard teaches economics.
While reading this long article, I think I was more troubled than encouraged.
There is some discussion of the way economics is taught at Harvard traditionally.
Mankiw’s textbook covers the abstract theory that underpins economics as it has been understood for decades. It is about supply and demand, about how prices can be used to match production of a good to its consumption, and about the power of markets as a tool for allocating scarce resources. Students in Ec 10 are asked to plot supply and demand curves, to solve simple word problems about what happens when the mayor of Smalltown, USA, imposes a tax on hotel rooms.
The idea is to impart a basic theory, to lay a foundation for understanding how society works. And that theory strongly implies that markets tend to work without much intervention, and that things like minimum wages might hurt more than help.
.
.
.
Feldstein and Mankiw were perfect leaders for the course. They both write frequently for popular audiences and are somewhat heterodox for Republicans. Feldstein became an irritant in the Reagan White House by bemoaning soaring budget deficits and demanding that tax increases kick in if Reagan’s cuts continued to add to the debt.
I have known about Feldstein and Mankiw for quite a while. I knew how pernicious was their influence on economic “thinking” is the USA. This article emphasizes just how big an influence they have been.
The part of the above excerpt talking about Mankiw is a good demonstration of the fallacies that crop up when you build macro economic theories on the framework of microeconomics. When you get to size of the economy that macroeconomics covers, there are feedback effects that become important that were not even measurable on the microeconomics scale.
The mention of Feldstein and federal government deficits showed how little Feldstein understood about macroeconomics that had all been thoroughly explained by John Maynard Keynes in the 1930s.
Data analysis was so subjective, so easily pliable to one’s own pre-chosen conclusions, as to feel almost useless. Then a new generation of economists — like Card, the late Alan Krueger, MIT’s Joshua Angrist, and many others — took it upon themselves to change that status quo, by carefully adopting research designs better able to determine causation (not just correlation), and focusing heavily on actual experiments and quasi-experiments where it’s clearer what factor is causing what phenomenon.
There is little in the article that talks about exactly how the new generation of economists are better able to determine causation not just correlation. I have read a number of books by Nassim Nicholas Taleb, Skin In the Game, Antifragile, The Black Swan, Fooled by Randomness, and The Bed of Procrustes. In one of them, he had a simple thought experiment about the difficulty of discerning causation by reconstructing it from the evidence you can gather. If you see a puddle on the floor in the kitchen, you would be hard pressed to give a detailed description about the ice cube that melted to cause the puddle. You would need a lot more information than you can get just by observing the puddle. He emphasizes that you are on much firmer ground describing what you see than in explaining how it happened.
George Soros introduced the idea of reflexivity in showing the difference between Social Science and Physical Scinece. (This may explain reflexivity – Reflexivity and Economics: George Soros’s theory of reflexivity and the methodology of economic science).
When you are developing the theory of planetary motion, the planets are not going to read your theory and change their behavior just to trip you up. On the other hand, when you write about how economic markets work and you pass laws to regulate those markets, the key players in these markets are avidly reading what you wrote and the laws you passed to figure how the loopholes that they can exploit. What you wrote about the market does change the way the market operates.
I’d be very interested to learn how these new economists are using big data statistics to avoid just these two pitfalls.
The article extols the use of differences of differences. I have heard the term before, but I don’t have much knowledge of what the experts are meaning with this phrase. All I have at this point is this excerpt from the article.
But most of Chetty’s discussion of the paper was about his methodology, what’s known in economics as a “differences in differences” approach. The key was to compare how sales of unaffected products in the stores changed from the start (26.48 sold per week) to the end (27.32 sold per week) of the experiment to how sales of affected products with the new label changed: from 25.17 per week to 23.87 per week.
The discussion sections for the class, run by Chetty’s graduate student teaching fellows, hammered home the point further. Michael Droste and John Macke, the grad students whose sections I attended, emphasized that differences-in-differences is a general technique that can be used in cases even when an actual experiment hasn’t been conducted.
I have used differences of differences in some personal financial software I have written. I have found the data calculated this way to be close to useless. Since I have other ways at getting at the information that are more likely to tell me something useful, I don’t know why I haven’t ripped out the differences of differences part of the software. Sometimes I find it hard to throw away software that I have worked so hard to write.