Friday, September 08, 2017

Realism in macroeconomic modeling

Via Tyler Cowen, I see that Ljungqvist and Sargent have a new paper synthesizing much of the work that's been done in labor search-and-matching theory over the past decade or so.

This is pretty cool (and not just because these guys are still doing important research at an advanced age). Basically, Ljungqvist and Sargent are trying to solve the Shimer Puzzle - the fact that in classic labor search models of the business cycle, productivity shocks aren't big enough to generate the kind of employment fluctuations we see in actual business cycles. A number of theorists have proposed resolutions to this puzzle - i.e., ways to get realistic-sized productivity shocks to generate realistic-sized unemployment cycles. Ljungqvist and Sargent look at these and realize that they're basically all doing the same thing - reducing the value of a job match to the employer, so that small productivity shocks are more easily able to stop the matches from happening:
The next time you see unemployment respond sensitively to small changes in productivity in a model that contains a matching function, we hope that you will look for forces that suppress the fundamental surplus, i.e., deductions from productivity before the ‘invisible hand’ can allocate resources to vacancy creation. 
The fundamental surplus fraction is the single intermediate channel through which economic forces generating a high elasticity of market tightness with respect to productivity must operate...The role of the fundamental surplus in generating that response sensitivity transcends diverse matching models... 
For any model with a matching function, to arrive at the fundamental surplus take the output of a job, then deduct the sum of the value of leisure, the annuitized values of layoff costs and training costs and a worker’s ability to exploit a firm’s cost of delay under alternating-offer wage bargaining, and any other items that must be set aside. The fundamental surplus is an upper bound on what the “invisible hand” could allocate to vacancy creation. If that fundamental surplus constitutes a small fraction of a job’s output, it means that a given change in productivity translates into a much larger percentage change in the fundamental surplus. Because such large movements in the amount of resources that could potentially be used for vacancy creation cannot be offset by the invisible hand, significant variations in market tightness ensue, causing large movements in unemployment.
That's a useful thing to know.

Of course, I suspect that recessions are mostly not caused by productivity shocks, and that these business cycle models will ultimately be improved by instead considering shocks to the various things that get subtracted from productivity in the "fundamental surplus". That should affect unemployment in much the same way as productivity shocks, but will probably have advantages in explaining other business cycle facts like prices. Insisting that the shock that drives unemployment be a productivity shock seems like a tic - a holdover from a previous age. But that's just my intution - hopefully some macroeconomist will do that exercise.

But anyway, I think the whole field of labor search-and-matching models is interesting, because it shows how macroeconomists are gradually edging away from the Pool Player Analogy. Milton Friedman's Pool Player Analogy, if you'll recall, is the idea that a model doesn't have to have realistic elements in order to be a good model. Or more precisely, a good macro model doesn't have to fit micro data, only macro data. I personally think this is silly, because it ends up throwing away most of the available data that could be used to choose between models. Also, it seems unlikely that non-realistic models could generate realistic results.

Labor search-and-matching models still have plenty of unrealistic elements, but they're fundamentally a step in the direction of realism. For one thing, they were made by economists imagining the actual process of workers looking for jobs and companies looking for employees. That's a kind of realism. Even more importantly, they were based on real micro data about the job search process - help-wanted ads in newspapers or on websites, for example. In Milton Friedman's analogy, that's like looking at how the pool player actually moves his arm, instead of imagining how he should move his arm in order to sink the ball.

It's good to see macroeconomists moving away from this counterproductive philosophy of science. Figuring out how things actually work is a much more promising route than making up an imaginary way for them to work and hoping the macro data is too fuzzy to reject your overall results. Of course, people and companies might not search and bargain in the ways that macroeconomists have so far assumed they do. But because labor search modelers tend to take micro data seriously, bad assumptions will probably eventually be identified, questioned, and corrected.

This is good. Chalk labor search theory up as a win for realism. Now let's see macroeconomists make some realistic models of business investment!


For some reason, a few people read this post as claiming that labor search theory is something new. It's not! I was learning this stuff in macro class back in 2008, and people have been thinking about the idea since the 70s. In fact, if anything, there seems to be a mild dampening of enthusiasm for labor search models recently, though this is hard to gauge. One exception is that labor search models have been incorporated into New Keynesian theory, which seems like a good development.

Sadly, though, I haven't seen any similar theory trend dealing with business investment. This post was supposed to be a plug for that.


  1. this point about surplus has been well understood since hagedorn manovskii 2008. as a solution it is completely unsatisfactory as it implies counterfactually large effects of unemployment insurance policies. this feature just highlights the fact that if you put a lot of sand in the wheels of the labor market, you get dampened responses of market tightness to any kind of shocks ( demand or supply/productivity doesn't really matter). This just highlights that some crucial propagation mechanism is missing. Eg you see layoffs spiking in downturns, but the standard dmp model has none of that. there is also job to job flows etc

    1. What do you think are the most promising mechanisms to fix those problems? I haven't really read much of this literature in almost a decade...

  2. Typical Mortensen-Pissarides is hardly realistic. Where do I see the matching function in reality? Isn't there heterogeneity across workers, and across firms in reality? Where are the banks? Where are the cows? It's good that you like search models, but that can't be because they're "realistic."

  3. But if the business cycle is not due to productivity shocks then why do we even need this? Why not stick to sticky wages? After all, we already know why and how that happens thanks to Truman Bewley.

  4. "The next time you see unemployment respond sensitively to small changes in productivity in a model that contains a matching function, we hope that you will look for forces that suppress the fundamental surplus, i.e., deductions from productivity before the ‘invisible hand’ can allocate resources to vacancy creation."

    I have trouble understanding this, but it sounds as if the "invisible hand" were supposed to convert the surplus generated by increased productivity into job vacancy. How about treating it as going the other way? The surplus generated by increased productivity results is eliminated by laying off people -- if you can't sell it, why produce it?

    Better yet, realize in some cases that increased productivity is a consequence of laying people off and eliminating their positions.

    Let's see: a supermarket increases productivity at the check-out stand by creating customer-operated check-out stands, and eliminating several employees there, but replacing them with one assistant who goes around as needed. How the heck is the "invisible hand" supposed to do anything other than suppress the labor market? (And consequently, reduce their purchasing of goods?)

  5. It's interesting to see this branch of the search literature lauded for realism in a discussion of the Shimer puzzle. While description of certain features of the labor markets was motivation for the development of these models, they are focused on aggregates in a way quite similar to many of the variety of DSGE-type models with microeconomic features more or less loosely tied to data. As I understand it, while some job flow data has been available for a longer period, many of the key tests of these models came from data sets collected after (and partially in response to) their development, like the JOLTS survey. And in terms of macro aggregates, the point of the Shimer puzzle is that empirical success has also been limited on that front.

    From my point of view, the part of the search literature that *has* been laser-focused on matching micro-level process and data on worker wages and transitions is the wage-posting literature following Burdett and Mortensen (1998), which generates full distributions of wages and flows across workers and firms, and has been fit nonparametrically to firm-worker micro-data at least since the late 1990s (see e.g. for one of the early examples, or much of the recent work of Jean-Marc Robin for extensions and refinements).
    The big problem with this literature, of course, is that the frameworks are static, so they don't have anything at all to say about business cycles. This has been changing recently too, though the business-cycle versions of these models still have many issues to work out, in part because there are so many moving parts that authors have to impose stringent economic restrictions (like the existence of a monotonic "job ladder" in which jobs vary only in one dimension) just to get numerical results. As a result, it's not clear whether mediocre descriptive performance reflects fundamental issues with the approach or if it's a byproduct of auxiliary restrictions that could be relaxed. Notably, these papers do not rely on a reduced form matching function in aggregate vacancies, which is often regarded as a crude approximation to some decentralized process, but might be capturing some things that are missing from the more explicit models (like unmodeled forms of congestion externalities which could create dependence between firms' behavior that's hard to see in micro-data but adds up in the aggregates).
    So overall, yes this area is promising and yes researchers in it do pay more attention to data on micro behavior than those in many other fields of macro, but the business-cycle component of this area seems to be facing challenges and data limitations in the same way that other aggregate-focused subfields are.

  6. "a good macro model doesn't have to fit micro data, only macro data"

    Take the “Marshmallow Test”: (1) banks create new money (macro-economics), and incongruously (2) banks loan out the savings that are placed with them (micro-economics).

    I rest my case

  7. I've been working on the backend and matching algo of a large job site for about a year now. I may be missing something (many things), but I don't see any of the actual problems with job matching (at least as exists for a few million job postings, candidates on our site/apps) really addressed head on in any of the models in that help wanted paper.

    The reality calculating resume relevancy is extremely hard and not done by anyone very well at this point in time. There are a ton of different vague titles out there. How should distance of candidate to job be factored into relevancy? How do you account for career progression or switching in the relevancy score? Because none of these things are handled very well, most of time seakers/recruiters won't be able to each other unless they are in the same area and the seaker has already had a job that the clustering algo says is similar. And that assumes the persons resume is machine learning friendly.