It's particularly good reading for PhD students. These are better models to emulate than older economists, as they tell you a bit more what directions are going well now -- at least what gets you baptized a star! Here are a few things I noticed.
Most of these economists are pursuing interesting questions about the real world. They're not following literatures, expanding on their thesis adviser's methodology, pursuing well established lines of research, or trying to impress people with fancy models.
Important questions are a dime a dozen, however. They also have good hunches for innovative answers to important questions.
Nick Boom:
Why are developing countries poor?... I am personally working on management practices: people in developing countries are poor because wages are low, and wages are low because firms are very unproductive, and firms seem to be unproductive in large part because of bad management.Bad management? Cool, new idea. Raj Chetty:
I plan to pursue research on two sets of projects over the next few years. The first will try to identify the determinants of intergenerational mobility, with an eye towards finding policies that increase equality of opportunity. Should we be focusing on increasing access to higher education? Changing the structure of elementary schooling? Revamping the tax code?Gauti Eggerttson:
...if one looks back 20 years from now, one will notice that a shift occurred towards studying the basic, big-picture, policy-relevant questions of macroeconomics—e.g., optimal currency areas, bank runs, fads and herding in financial markets, and automatic stabilizers—that have the power to change the course of history.Glen Weyl is after Hayek and how information disperses through an economy, a "big question" if there ever was one:
information technology is leading individuals to delegate their most “private” decisions to automated processing systems. Choices of movies, one of the last realms of taste one would have guessed could be delegated to centralized expertise, are increasingly shaped by services like Netflix’s recommender system. While these information systems are mostly nongovernmental, they are sufficiently centralized that it is increasingly hard to see how dispersed information poses the challenge it once did to centralized planning.(It's interesting that Hayek had one of the most penetrating analyses of a market system in "the use of knowledge in society," and other than Tom Sowell's great book "Knowledge and Decisions" it hasn't led to further research. Maybe we finally have the tools to do it!)
Practically every single one of them said, our main challenge is to increase long-run growth. The new growth theorists really won, at least the agenda.
The tools have changed dramatically. I went to grad school in the era of math. We were advised to take more of it. If you had extra time you took real analysis or measure theory.
But science always progresses by new telescopes. Galileo didn't upend Copernican cosmology with measure theory, he had a new telescope. Science also progresses from intellectual arbitrage. In my era, people who knew math, time series, control theory, dynamic programming had just revolutionized the field. But today, the new telescope and intellectual arbitrage is obviously computers and the data collection power of the internet. All of the young stars echo this change:
Bloom again:
The progress is likely to be heavily empirical—simply because more and more data is becoming available, and it is easy to analyze with fast computers (so empirics is now advancing faster than theory)—and spread across many hundreds of topics. So economics has gone from Victorian science, where one genius in his shed could invent the steam engine over the weekend, to industrial science, where innovation comes in thousands of tiny steps made by dozens of research teams.Raj Chetty again:
Looking ahead, I am most excited about the prospect of having clear, evidence-based answers on which policies have the most beneficial economic impacts. I am especially optimistic that the expansion of access to large administrative datasets, such as earnings data from social-security records or student-achievement data from school districts, will yield sharp, quasi-experimental evidence
One can argue with this. After all "inconsistent" means you get the wrong answer no matter how much data you have. Cause and effect is never easy. But that's an old grumpy guy talking, you see the trend
Gita Gopinath:
Addressing these areas will require breakthroughs in theory and empirical work, with more micro-level datasets on prices, trade, and capital flows being brought to bear.Justin Wolfers:
Today, every interaction we have in our lives leaves behind a trail of data. Whatever question you are interested in answering, the data to analyze it exists on someone’s hard drive, somewhere. This background informs how I think about the future of economics....Specifically, the tools of economics will continue to evolve and become more empirical. Economic theory will become a tool we use to structure our investigation of the data.The implication is clear. If you're going in to economics these days, learn python, R, stata, html, java; know how to scrape data from the web, run a large programming task in a disciplined style, manipulate and clean large data sets. That's the key intellectual arbitrage behind the young stars' work today, and way more important than measure theory and real analysis!
There was also a certain amount of inward-looking assumption-driven thought, how to bring less rational agents into economic theory. I find behavioral economics has been most interesting in bringing uncomfortable facts to economics rather than rethinking assumptions. And I'm curious why people going this way, like Chetty, want to assume people are behavioral so they can offer better planning advice to the super-rational "policymaker". Surely, it's time for a behavioral Stigler or Buchanan to step up to notice that policy making isn't all that "rational" either. But I'm supposed to be reporting, not editorializing, so make up your own mind whether this assumption-driven line of research will be productive.
