Arnold Kling  

Thomas Sargent, Master Macroeconomist

PRINT
David Friedman on Eggs and Qua... Oil Bubble Talk is Back...

In an interview, he says,


the more dynamic, uncertain and ambiguous is the economic environment that you seek to model, the more you are going to have to roll up your sleeves, and learn and use some math.

Pointer from Mark Thoma. Self-recommending, because Sargent is to rational expectations macro what Hicks is to Keynesian macro. Samizdat versions of Sargent's graduate macro lecture notes were for us what "Mr. Keynes and the Classics" was to an earlier generation.

You know that I am not a fan of mathematical economics, but I am a fan of Sargent. There is a great deal to chew on in the interview, including his thoughts on the relationship between modern finance and macro as well as this on the labor market:


Lars [Ljungqvist] and I believe that when people now become unemployed, they're taking a more or less permanent hit to their level of human capital, a larger one than they might have received before 1980. We have a theory that people build up human capital while they're working on a job, but lose human capital when they're displaced from a job. We think that after 1980, people in Western economies started suffering bigger drops in their human capital at the moment that they suffer a job displacement.

...Our models have cohorts of aging heterogeneous workers. Our models imply that people in Europe, especially older workers, are suffering from long-term unemployment because of the adverse incentives brought about by a generous social safety net when it interacts with these human capital dynamics.

So, we don't have a representative agent working at the GDP factory, but instead we have heterogeneous workers. And some of them take big hits to their human capital. This is very close to the ZMP and Recalculation Story. The marginal product falls, not to zero, but to the point where safety-net supported unemployment is preferable to taking a job at one's marginal product. As Sargent puts it,

Unemployment compensation systems typically award you compensation that's linked to your earnings on your last job; those past earnings reflect your past human capital, not your current opportunities or current human capital. That can make collecting unemployment compensation at rates reflecting your past (and now obsolete) human capital more desirable than accepting a job whose earnings reflect a return on your current depreciated level of human capital. This mechanism sets an incentive trap that induces the European worker to withdraw from active labor market participation.

What does that mean for the U.S. now?

the politics of the current situation can imply that so long as unemployment is high, we're going to extend the duration and generosity of benefits. And that extension, done out of the best of motives, is exactly what can lead to the trap of persistently high unemployment.

Again, I would say that we could ameliorate this risk by allowing people to continue to collect unemployment compensation after they take a new job. At the margin, this would make taking the new job more lucrative than when you lose unemployment benefits upon taking the job.

The final section of the interview discusses sovereign debt issues. As I read through it, I nodded my head vigorously in agreement. Then Sargent hit me with,


notice that throughout our discussion, Art, we've been using the vocabulary of rational expectations.

Ouch! On the one hand, I use Krugman's dismissive term Dark Age Macro to describe what Sargent unleashed on the profession. On the other hand, my guessing the trigger point paper stands on the shoulders of some of that stuff.


Comments and Sharing





COMMENTS (6 to date)
Norman Maynard writes:

Referring to rational expectations and DSGE macro as "Dark Age Macro" is rather condescending, and as you mention, the fact that you do take advantage of some of the advances provide indicates that it's not terribly accurate even for naysayers (not unlike the "Dark Ages" themselves, actually).

I suspect that rational expectations macro might better be thought of as "Macro Gaiden," to use the Japanese term. What we've been seeing has been a side-story of macroeconomics, the story retold from a different perspective. That perspective has certain important benefits, but it's hardly the main issues.

I suspect that in the next decade or so we'll see a shift away from the Gaiden and back to the main story of macro, retaining only those aspects of rational expectations that were particularly illuminating.

Troy Camplin writes:

"the more dynamic, uncertain and ambiguous is the economic environment that you seek to model, the more you are going to have to roll up your sleeves, and learn and use some math."

No. The more dynamic, uncertain and ambiguous is the economic environment that you seek to model, the less mathematics can help you. In fact, the more dynamic, uncertain and ambiguous is the economic environment that you seek to model, the more abstract the mathematics must necessarily become, until you have general models that demonstrate basic principles and little else. That is important, of course, but it can tell you nothing at all about the actual details of an actual economy. The mathematics is too complex -- we don't have the math to describe it and, if Stephen Wolfram is correct, we never can or will. More complex methodology is needed than mere mathematics.

Kevin Donoghue writes:

"...notice that throughout our discussion, Art, we've been using the vocabulary of rational expectations."

That's clever rhetoric. It makes RE sound much more impressive than it really is. Before Muth, economists managed quite well by distinguishing between events which were foreseen and events which were not. The former were taken into account in the construction of supply and demand curves, the latter gave rise to shifts in those curves and windfall gains or losses.

The problem with Dark Age macro isn't that the economist-theologians have learned a new vocabulary. Learning new words doesn't have to mean forgetting hard-won old ideas. But that's what happened.

Paul writes:

I'm a fan of Sargent, but its quotes like this that makes it hard to take academics seriously.

"The Kareken and Wallace model’s prediction is that if a government sets up deposit insurance and doesn’t regulate bank portfolios to prevent them from taking too much risk, the government is setting the stage for a financial crisis."

really? that's a prediction of the K&W model? agency problems? but they don't predict that participants will circumvent the regulations?

I guess these are the subtle insights that mathematical models expose.

B.B. writes:

Here is the problem with the proposal to allow people to keep collecting benefits even after they get a job.

One the one hand, you are right that your proposal eliminates the incentive to not take a job because the person would lose benefits. That is a supply-sider story about marginal tax rates.

On the other hand, there is a game theory problem. Workers would have an incentive to 'lose' a job, get benefits, then get their old job back, and keep collecting benefits. The system can be gamed. Employers have every incentive to play along because it boosts the earnings of their workers.

Net/net, the current system may work best.

fundamentalist writes:

Sargent: “The criticism of real business cycle models and their close cousins, the so-called New Keynesian models, is misdirected and reflects a misunderstanding of the purpose for
which those models were devised.6 These models were designed to describe aggregate economic fluctuations during normal times when markets can bring borrowers and lenders together in orderly ways, not during financial crises and market breakdowns….The authors of papers in this literature usually have made it clear what the models are designed to do and what they are not. Again, they are not designed to be theories of financial crises.”

So these models are no good at predicting crises because they weren’t designed to predict crises. So how do you predict crises?


Sargent: “Art, it is just wrong to say that this financial crisis caught modern macroeconomists by surprise. … Enlightened by those data, researchers have constructed first-rate dynamic models of the causes of financial crises and government policies that can arrest
them or ignite them.”

So we have the models to predict crises after all? But they are different models from, the macro models mentioned above. And boiled down, those models consist of how deposit insurance can prevent bank runs but also create moral hazards.

Sargent: “What triggers a bank run is patient depositors’ private incentive to withdraw early when they think that other patient investors are also choosing to withdraw early. Technically speaking, that amounts to multiple Nash equilibria. There are situations in which I run (i.e., withdraw from the bank early) because I expect you to run, and when you also run because you expect me to run. But there are other situations in which we both trust that the other person isn’t going to run and we don’t run. Which equilibrium prevails is anyone’s guess, or something resolved only by an extraneous random device for correlating behavior, a device that economists sometimes call a “sunspot.”

Isn’t this just Keynes’ “animal spirits” applied to depositors? How is that new?

I’m sorry, but the Sargent’s responses were very unsatisfying. All I get from them is that modern macro worries about bank runs and that anything can be a bank in the sense that a lot of people will want to bail out at the same time. We have no idea why people suddenly get the desire to take their money out of investments and cause bank runs. Deposit insurance helps to some degree but also creates moral hazards, so we should require anything that even smells like a bank to hold greater reserves.

Any economist in the 1920’s could have written this. It shows no progress at all in economic thinking. Economists have succeeded in translating these ideas into math expressions and collecting a lot of data about it, but they haven’t advanced the theory of business cycles for almost a century. Read Hayek’s “Monetary Theory and Trade Cycles” and you’ll see what I mean.

If mainstream macro understands crises so well, why didn't they inform the federal reserve and the government of what was wrong? Why weren't they shouting from the rooftops that we were headed for a major crisis. Why did the Queen ask the LSE how they missed that one? Against Sargent's protests, it seems clear to everyone else that mainstream econ was caught flat footed. It's stunning that Sargent is oblivious!

Comments for this entry have been closed
Return to top