Summer Institute

I just got back from the NBER Summer Institute. The Economic Fluctuations and Growth meeting organized by Larry Christiano and Chad Jones sparks some thoughts on where macro is and where we're going. (I also attended the monetary economics and asset pricing meetings, which were excellent and thought provoking too, but one can only blog so much.)

Review:


There were two papers on macro theory. Fist, the conference started with Gauti Eggertsson and Neil Mehrotra's "A Model of Secular Stagnation," which I discussed, slides here.

I think it's an important paper. The standard simple New-Keynesian model has a lot of trouble to produce a steady slump with positive inflation.  So if you want "secular stagnation," you need a new model. I also have a lot of trouble with the "negative natural rate." It tends to be a deus-ex-machina, output is lower than I'd like so the natural rate must be negative.  It would be much more convincing if we could separately measure the natural rate, but that too needs a model. This paper provides a model whose steady states resemble old fashioned static Keynesian relations, not the dynamic new-Keynesian ones, and a model where one could think about separately measuring the negative natural rate.

"Important" doesn't mean "right" or "conclusive." This model rules out storage, has no money, and hobbles the rate of return on capital, all of which tend to put bounds of zero or above on long-term real interest rates. More thoughts on the slides, which I may write up at more length some day. (Olivier Blanchard discussed the same paper on Friday, bringing in data from around the world. If he posts his slides I'll update.)

Second, Paul Beaudry presented his paper with Dana Galizia, Franck Portier, titled "Reconciling Hayek's and Keynes' Views of Recessions," which Ivan Werning discussed. It was a rather complex model trying to capture overaccumulation and liquidation.

There were two empirical papers. Simon Gilchrist, presented his paper with Raphael Schoenle, Jae Sim, Egon Zakrajsek, "Inflation Dynamics During the Financial Crisis," discussed by  Mark Bils. Companies short of cash in the financial crisis raised prices; companies with a lot of cash lowered them. Clean dynamic model, clean data, a nice bit of the micro data analysis going on in macro these days.

Sarah Zubairy presented her paper with Valerie Ramey, "Government Spending Multipliers in Good Times and in Bad: Evidence from U.S. Historical Data," discussed by Yuriy Gorodnichenko. As Valerie has done before, they regress output on military spending shocks to estimate multipliers. Here the question is whether the effects are larger when there is higher unemployment or a low interest rate, with a bunch of small but important methodological improvements. The conclusion is no, and multipliers a bit below one throughout, but much methodological discussion on how one interprets the facts.

There were two "Growth" papers. First, Roland Benabou presented "Forbidden Fruits: The Political Economy of Science, Religion and Growth" with Davide Ticchi and Andrea Vindigni. The basic idea is that religion blocks or adapts to new ideas, going back centuries. History, going back a thousand years, regressions of patents on religiosity, all building to a big model, with section titles like "Inequality, Religion and the Politics of Science."

Paul Romer "discussed" the paper, i.e. gave a long and thoughtful speech, covering religion, social norms, neuroeconoimcs (Southerners faced with a slight insult have big spikes in cortisol levels compared to Northerners), the shocking rate of incarceration in the US, words vs. equations in economics, and lots more.

Last but certainly not least, Ufuk Akcigit presented "Young, Restless and Creative: Openness to Disruption and Creative Innovations" with  Daron Acemoglu and Murat Alp Celik, discussed by Sam Kortum, The basic idea is that companies with young CEOs are more likely to make radical innovations rather than incremental ones. A complex model precedes regressions of patent citations on CEO age.

Thoughts:

Just how we do economics was a big theme running through all the discussion. Words vs. equations; models and empirical work; and what kinds of things we look at and what kind of work people are doing.

Most of the theory papers had some "motivating" facts. Most of the facts papers and more or less motivating theory. Not one paper wrote down a model, estimated or calibrated its parameters, and compared that model to data.  (Gilchrist came pretty close, but more the exception that proves the rule.) This isn't a complaint, really, it's just where we are. The kinds of things people want to investigate are just too hard to write down models rich enough to take to the data.

This point came up again and again. Sam gently chided Ufuk at al for presenting 24 pages of complex model all to "motivate" some regressions. He suggested that the model should be used to guide and constrain regressions, and to give a more structural interpretation to the parameters. Pat Kehoe, asking a question, complained that it's awfully hard to measure a fiscal multiplier with no guidance of which model for its possible operation. He pointed out that any model restricts how many variables together should respond to a fiscal expansion. For example the static Keynesian model says consumption should rise. The real business cycle model gives a multiplier through impoverishing people, which has joint predictions across consumption, labor, etc. Likewise I complained that in wars, the assumption that everything else is on average equal -- made when regressing output on fiscal shocks -- seems a bit stretched.

Similarly, both of the macro theory papers stopped well short of serious confrontation with data. We didn't see anything like the standard fully specified models of the Larry Christiano type, compared to, say, impulse-response functions.  The models are so stylized you can't begin to quantify them.  (I got off  cheap shot pointing out that secular stagnation required deflation in the model. Since we do not have deflation, case closed. It's a cheap shot because I think the model could be easily modified to have stagnation with low positive inflation.) This too is not really a criticism. I've been working with simpler and simpler models, as I find it hard to keep the intuition and quantitative parable aspect alive as models get more complex. You have to walk before you can run. But questions like, how could Ed Prescott and Ellen McGrattan go off and measure the natural rate, are not yet answered.

A similar issue came up in the paper I discussed for Asset Pricing, Aytek Malkhozov, Philippe Mueller, Andrea Vedolin, and Gyuri Venter "Mortgage Risk and the Yield Curve," slides here. It developed a really nice arbitrage-free model with supply effects. And then used the model only to "motivate" regressions of returns on a measure of duration. Though the regression coefficient is tied to structural model parameters, the authors never made that link at all. Well, the model was perhaps too simple to do that. And, everyone else seems to be writing papers the same way. It's not a criticism, here, but an observation on our emerging culture.

Math vs. literature is a similar theme to atheoretical regressions/models as parables vs. estimates and tests. In my 30 years as an economist, our field has become much more literary and less quantitative. In part that reflects a different emphasis. It's really hard to build towards maximum likelyhood tests of effects of religion on the adoption of new ideas. Paul Romer commented on this at length, with "models vs. words" on his slides. In his view, math is a useful language because it removes much of the value-laden elements of language and forces logic to be out in the open. He linked language to us vs. them, social norms, morality, and those pesky cortisol levels. (I'm doing my best to recall a speech, so forgive me Paul if I don't get it all right.) He pointed to my use of "paleo-Keynesian" to describe the static models from the 1960s, guessing nobody would remember anything else from my discussion. When I complained that Paul Krugman invented the term, he pointed out (correctly) that such borrowing just made its use more rhetorically effective. There go the cortisol levels. I'm not sure in the end though whether Paul was approving or bemoaning the shift back towards literature in economic analysis. Certainly his vision for the future of growth theory, centered on values, social norms, biology, and so forth, does not lend itself easily to quantification.

The use of ancient quotations came up several times. I  complained a bit about Eggertsson and Mehrotra's long efforts to tie their work to quotes from verbal speculations of Keynes, Alvin Hansen, Paul Krugman and Larry Summers. Their rhetorical device is, "aha, these equations finally explain what some sage of 80 years ago or Important Person today really meant."  Ivan Werning really complained about this in Paul Beaudry's presentation. What does this complex piece of well worked out "21st century economics" have to do with long ago muddy debates between Keynes and Hayek? It stands on its own, or it doesn't. (In his view, it did, so why belittle it?)

Yes. Physics does not write papers about "the Newton-Aristotle debate." Our papers should stand on their own too. They are right or wrong if they are logically coherent and describe the data, not if they fulfill the vague speculations of some sage, dead or alive. It's especially unhelpful to try to make this connection, I think, because the models differ quite sharply from the speculations of the sage. Alvin Hansen certainly did not think that a Taylor interest rate rule with a phi parameter greater than one was a central culprit in "secular stagnation." I haven't checked against the speech, but I doubt he thought that inflation would completely cure the problem in the first place.

Sure, history of thought is important; tying ideas to their historical predecessors is important; recognizing the centuries of thinking on money and business cycles is important. But let's stand up for our own generation; we do not exist simply to finally put equations in the mouths of ancient economists.

But, tying it all up, perhaps I'm just being an old fogey. Adam Smith wrote mostly words. Marx like Keynes wrote big complicated books that people spent a century writing about "this is what they really meant." Maybe models are at best quantitative parables. Maybe economics is destined to return to this kind of literary philosophy, not quantified science.

Curious too what was missing. All the macro was decidedly Keynesian. General equilibrium with distortions, anything other than trend on "supply" was noticeable by its absence. So was the discussion. But maybe that's my fault for going to the NBER and not the Minnesota Macro meetings.

A last thought. Economic Fluctuations merged with Growth in the mid 1990s. At the time there was a great confluence of method as well as interest. Growth theorists were studying growth with Bellman equations, dynamic general equilibrium models of innovation and transmission of ideas, thinking about where productivity shocks came from. Macroeconomists were using Bellman equations, and studying dynamic general equilibrium models with stochastic technology, along with various frictions and other propagation mechanisms.

That confluence has now diverged. I enjoyed spending an hour or two thinking about how religion has blocked or adapted to ideas over the centuries, and Paul's view on social norms or neuroeconomics. But I don't really have any expertise to contribute to that debate. Questions like whether young CEOs head more innovative companies, or whether, like deans, what matters is the age of the faculty are a little closer to home, since I spend a lot of time consuming corporate finance. But the average sticky-price macro type does not. Likewise, when Daron Acemoglu, who seems to know everything about everything, has to preface his comments on macro papers with repeated disclaimers of lack of expertise, it's clear that the two fields really have gone their separate ways. Perhaps it's time to merge fluctuations with finance, where we seem to be talking about the same issues and using the same methods, and growth to merge with institutions and political or social economics.