summer camp for econos

My kids are off with my parents for two weeks, having fun at the farm, playing with their cousins, and overloading the wifi (yes, they're mine). Their two weeks away is not randomly chosen ... I pick this time so I too can get away to "summer camp." Yes, summer camp for economists exists, or at least that's how I think of the NBER Summer Institute. I am not trying to diminish the seriousness of it, a lot of cutting-edge research and insightful comments are packed in those many sessions. Still it's easy to overdo the seriousness, so summer camp gives a nod to the awkward social of it too.

Twice I have gotten to present research in a consumption group ... the first time when my son was 5 weeks old (ahem, macromom) and the second time last year right after I started at CEA (also exhausting). This year was more typical for me: go and learn about new research, and, importantly, what other research economists think of it. The one twist this year is that I went to a macro group, so some new camp counselors and campers.

Here are some of my general reactions. This is a poor substitute for being there or even reading the papers but I think it's a good to try to put the technical in a wider conversation.

Better data and statistical tools will never substitute for economist intuition. One paper on the effects of extended unemployment benefits turned into a colorful "he said/she said" between the authors and the discussant ... or more factually a "he-he said/he said." Such debates are not uncommon and underscore why a single empirical study will never settle an argument. Nor should it. I don't know enough about this topic to weigh in (but good luck finding meaningful, exogenous variation in macro policy). What I found interesting was how others tried to assess the results of the paper in light of the vigorous rebuttal. An appeal to the workhorse theoretical model and intuition about the opportunity costs of unemployed workers was brought to the defense of the authors. That's all fine and good but you didn't need clever empirical work to get you to the conclusion then. And there were other examples of economist intuition doing some heavy lifting. An interesting paper on the complex relationship between uncertainty and the macroeconomy rolled out some fancy, new econometrics that were disciplined by a handful of intuitive "reality checks." Letting the data speak is not the same thing as letting the data do all the talking.

But better data and statistical tools will eventually overturn most economist intuition. Now is an exciting time to be working on consumer behavior: there is an explosion of new micro data on spending and the recession/slow recovery created a lot of big questions. It feels like the "representative agent" and the workhorse theory (even with all its extensions), is finally starting to collapse under the weight of empirical patterns. Adding to this groundswell was a paper on the response of consumer spending to income. Once again, spending behavior among higher-income households in the data ran against economist intuition (and a canonical pro-theory paper was taken down on measurement error in income). I think the empirics are running ahead of the theory at this point, which should give us pause since a) intuition is crucial, see above and b) there is a lot about these data we don't understand yet. Attempts to 'square the circle' by relaxing rationality are interesting, which also came up in another, more theoretical paper. But I am skeptical. Plus a dirty little secret is that it is harder to calculate a near-rational choice than a rational one, since you also have to weigh the cost of being sub-optimal. Plus I don't see how we can come up with new models by simply looking at behavior or intuiting as economists. Thankfully, data comes in many forms, including listening to people. For example, I've been pondering this quote from a recent Atlantic Monthly article (the highlighting is my emphasis) and keep looking for a model that fits:

I have not found one yet (though identity economics holds promise and research on wealthy-hand-to-mouth has pushed this inquiry along). Still it feels like real progress that the profession does not insist on shoehorning everyone in standard theory. Or explaining away such behavior as irrelevant. Economist intuitions do change.

Simple questions do not have simple answers. It is 2016, more than eight years after its start and the answer to "What caused the Great Recession?" is as contentious as ever. Economists are often ridiculed for an inability to forecast such major events, but it seems more telling (and deeply related) that we cannot even agree on the causes well after the fact. For all the data, all the econometrics, all the attention of smart economists ... this "simple" question is probably not going to be settled, ever. I began forecasting consumer spending in early 2008 so it is an understatement to say that I have been following the topic with interest. Along the way I have learned a lot and not just from the academic research. Still it was a little odd to hear things described as "the new narrative of the crisis" yesterday that felt pretty familiar to me. One thing that struck me was how much this research debate is starting to feel like peeling back the skin on an onion. And despite the accompanying tears, it makes me very happy to see us digging down deep in the guts of our models. Economic models can be quite powerful (I do think we pass a market test to some extent), however, that power rests on key parameters and relationships being relatively stable. Recessions are almost by definition a time of instability, and it is hard to trace down the roots of instability in models that largely assume it away. I am a big fan of belief shocks, I don't think we can fully understand recession/recovery without appealing to shifts in expectations. And yet, I have no idea how you cleanly, credibly separate beliefs from credit supply. I am happy to plug away at those micro foundations and follow efforts of others ... but the limits of economic inquiry are worth reflecting on too. Policy is done in real time ... no one could wait for eight years of research to make their decisions, but even today's research would not have told policymakers what to do. Clearly years of research have shed light on the Great Recession. Yet it is hard to escape the feeling that we need policies and institutions that are resilient to a good bit of uncertainty, as opposed to thinking with more work or better models will be able to foresee or even counteract the shocks next time.

In summary, my one day at summer camp this year was great and gave me plenty to think about, and a lot of it more technical than these broad-brush impressions. The Summer Institute is a real credit to the organizers, authors discussants, and participants ... and underscores how collaborative research really is. And on that note, I had lunch with a co-author who reminded me that I have some research to be working on ...

**Opinions here are mine and should not to be attributed to anyone with whom I work.**


economist - my views here are my own

Also on this blog

SHARE:  Email · Facebook · Google · Twitter · Tumblr · Kindle
SUBSCRIBE:  Receive an email on new posts from claudiasahm


I read that Atlantic article, slowly and thoughtfully, beginning to end, as my main career is in personal finance. Honestly, the highlighted quote looks pretty straightforward to explain to me. There are often big non-pecuniary rewards to a career (but these can be hard to estimate, and easy to overestimate, relative to the benefits of higher income level and higher income security.)
The first thing I think of is just simply adding to a model a variable for non-pecuniary utility. Of course, doing this well may be quite difficult. And, of course, I can tell you after teaching personal finance to over 6,000 students at the University of Arizona, and a great deal of counseling on top, most people are ridiculously far from perfect public information and expertise with regard to these decisions.
2016-07-18, Richard H. Serlin

  • Notify me upon new comments

☺ Got it