How economics went from theory to data

(Bloomberg View) — One of the most striking things about attending the annual meeting of the American Economic Association after a long absence is that economics is now really all about the data.

Older theorists such as Eric Maskin (who won the 2007 economics Nobel), Jean Tirole (the 2014 Nobelist) and Bengt Holmstrom were still accorded prominent roles as luncheon speakers at this week’s gathering in San Francisco, but in the sessions where actual research was being presented most of the activity and excitement surrounded empirical work. (It’s really the AEA and a bunch of other economics groups, and the official title of the gathering is the Allied Social Science Associations annual conference. But I figured it would be confusing to call it that in my first sentence, and the AEA is the dominant organization at the conference.)

Daniel S. Hamermesh of the University of Texas documented this shift in a 2013 article in the Journal of Economic Literature. In 1963, 1973 and 1983, the majority of the articles published in the American Economic Review, Journal of Political Economy and Quarterly Review of Economics, three of the field’s most influential journals, were works of theory – with theory’s dominance peaking in 1980. By 2011, theory’s share was down to 27.9 percent.

One cause seems pretty clear. The biggest shift toward empirical work occurred between 1983 and 1993, and it was between 1983 and 1993 that personal computers became commonplace. That made crunching data much easier for economics professors; the subsequent rise of the Internet and digitization of much that was once analog in the economy opened up a huge new array of data for them to crunch.

In Hamermesh’s taxonomy, borrowed data means “ready-made … government-provided … macroeconomic time series or … large household surveys,” while own data means that the authors of the article created the data set — even if the source was government records, as it was with Thomas Piketty and Emanuel Saez’s famous work on top incomes. The continued rise in empirical research since 1993 has been entirely in this latter category. Economics has also seen the advent of experimental work, most of it taking place in campus “labs” where students and other subjects participate in market-related games and exercises.

Disillusionment with theory has also been an issue. From the late 1930s through the 1970s,economics was full of excitement about grand mathematical models that seemed to explain everything about the world. Then some things happened that the grand models — particularly the macroeconomic ones — didn’t explain very well, while a new generation of theorists took things in increasingly narrow and convoluted directions. The goal was often to make the theories more realistic, but the result, as Hamermesh puts it, was that:

Economic theory may have become so abstruse that editors of the leading general journals, recognizing that very few of their readers could comprehend the theory, have cut back on publishing work of this type.

Piketty, who was a promising young theorist at the Massachusetts Institute of Technology in the early 1990s, wrote in the introduction to “Capital in the 21st Century” that he decided to move back to France in part because economists are less respected there and thus must “set aside their contempt for other disciplines and their absurd claim to greater scientific legitimacy, despite the fact that they know almost nothing about anything.” Then he went looking for some data to crunch.

Now that’s what all the cool economics kids are doing. Here’s a sample session title from this week: “Data Gold! Exploiting the Rich Research Potential of Lifetime Administrative Earnings Data Linked to the Census Bureau’s Household SIPP Survey.”

I actually didn’t attend that one (seemed a little too technical), but the most entertaining presentation I did see involved Philippe Aghion of the College de France bounding about the front of a crowded hotel ballroom explaining how he and two co-authors had linked data from the European Patent Office, the Finnish statistical agency and the Finnish military to study whether people with high IQs invented more things and made more money than others. They did, even controlling for parental education and income. (This research is still preliminary, and doesn’t appear to be available online yet.) The meritocracy reigns! In Finland, at least.