“Microfoundations”: I Do Not Think That Word Means What You Think It Means: Wednesday Focus

Ash nazg durbatalik!–no, that is wrong: “nazg” is the Orcish word for “ring”. What is the Orcish word for “economic model” anyway?

I find this morning that the intelligent, thoughtful, and extremely hard-working Mark Thoma is pleading for appropriate model-using: use the right tool for the job, rather than simply building the biggest hammer you can under the assumption that everything is a nail:

Mark Thoma: One Model to Rule Them All?:

Models are built to answer specific questions. We do not have one grand model that can explain everything…. The best map to use to drive from Seattle to Los Angeles is a lot different from the best map to find out if it is safe to dig a deep hole in someone’s yard…. And if we try to stuff even more information onto the map so that it can answer all of our questions, telephone poles, roads, elevation, every side road, every house and every store, all the bus routes, rainfall, the types of vegetation, etc. etc. the map becomes too complicated to be useful. Economic models are no different. The trick in modeling is to pare away all the inessential features so that there can be a sharp focus on the question of interest. The best maps are very specialized and highlight only what we need to know. The best economic models do the same…

Where I disagree with many of my colleagues is in the assertion that we should limit ourselves to a single class of models, e.g. variations on the New Keynesian model… built to explain a world of moderate fluctuations in GDP… featur[ing] temporary price rigidities… [with] aggregates… consistent with the optimizing behavior of individual consumers and producers. For certain types of questions–how should policymakers behave to stabilize an economy with mild fluctuations induced by price rigidities–it is the best model to use’85. When a different world emerged, a large financial shock and the ensuing Great Recession, the model was of little use…. The IS-LM model, on the other hand, was built in the aftermath of the Great Depression to examine precisely the kinds of questions we faced throughout the Great Recession, issues such as a liquidity trap, the paradox of thrift, and how policymakers should react in such an environment. Why is it surprising that a model built to explain a particular set of questions does better than a model built to explain other things?

And the last surviving wild pre-Keynesian/UCLA monetary economist David Glasner is not a happy camper either:

David Glasner: Microfoundations (aka Macroeconomic Reductionism) Redux:

In this [UCLA] context microfoundations meant providing a more theoretically satisfying, more micreconomically grounded explanation for a phenomenon–“sticky wages”–that seemed somehow crucial…. And it is not microfoundations in this sense that is controversial. The sense in which microfoundations is controversial is whether a macroeconomic model must show that aggregate quantities that it generates can be shown to consistent with the optimizing choices of all agents in the model…. If… not… the macromodel is now considered inadequate and unworthy of consideration. Here is how Michael Woodford, a superb economist… put in his paper “The Convergence in Macroeconomics: Elements of the New Synthesis”:

But it is now accepted that one should know how to render one’s growth model and one’s business-cycle model consistent with one another in principle…. It should be possible to reconcile one’s views about household or firm behavior, or one’s view of the functioning of individual markets, with one’s model of the aggregate economy, when one needs to do so….

If the world… is not like the world assumed by modern macroeconomics… estimates… reflecting [that] worldview… will be inferior…. But, on methodological principle, modern macroeconomics treats the estimates generated by any alternative econometric model insufficiently grounded in the microeconomic principles of intertemporal optimization as illegitimate….

Newtonian physics reduced the Keplerian laws of planetary motion to more fundamental principles… achieved an astounding increase in explanatory power and empirical scope. What has the methodological reductionism of modern macroeconomics achieved?… Carlaw and Lipsey… [see that] methodological reductionism in macroeconomics has resulted in a clear retrogression in empirical and explanatory… an antiscientific exercise in methodological authoritarianism…

I think that Mark Thoma has only half the story, and the other half of the story is implicit in David Glasner. Yes, New Keynesian models were built to study post-industrial North Atlantic economies undergoing small fluctuations in nominal demand and small supply shocks in a context in which inflation expectations were at least potentially unanchored and in which the financial system was stable. But New Keynesian models were also built to accomplish a different task. They were built to rebut a current of thought lamented by Robert Solow:

Robert Solow: Dumb and Dumber in Macroeconomics:

The original impulse to look for better or more explicit microfoundations was probably reasonable. It overlooked the fact that macroeconomics as practiced by Keynes and Pigou was full of… generalizations about aggregative consumption-saving patterns, investment patterns, money-holding patterns [that] were always rationalized by plausible statements about individual–and, to some extent, market–behavior. But some formalization… was a good idea. What emerged was not… a single representative consumer optimizing over infinite time with perfect foresight or rational expectations, in an environment that realizes the resulting plans more or less flawlessly through perfectly competitive forward-looking markets for goods and labor, and perfectly flexible prices and wages. How could anyone expect a sensible short-to-medium-run macroeconomics to come out of that set-up?… We want macroeconomics to account for the occasional aggregative pathologies that beset modern capitalist economies…. A model that rules out pathologies by definition is unlikely to help. It is always possible to claim that those “pathologies” are delusions, and the economy is merely adjusting optimally to some exogenous shock. But why should reasonable people accept this?… Why should the burden of proof fall on those who see an ordinary standard pathology here?

And they rebutted this current of thought by demonstrating that all you needed was very small market frictions–very small deviations from the Ramsey setup at the level of the individual agent–to produce large and macroeconomically-significant pathologies: to show that large business cycles could emerge from small menu costs, and thus the assumption that market failures were small at the level of individual economic actors should not be taken to imply that any macroeconomy was approximating the solution to any optimal central planning problem.

But then Mike Woodford and company lost sight of the goal. Yes, New Keynesian models with more or less arbitrary micro foundations are useful for rebutting claims that all is for the best macro economically in this best of all possible macroeconomic worlds. But models with micro foundations are not of use in understanding the real economy unless you have the micro foundations right. And if you have the micro foundations wrong, all you have done is impose restrictions on yourself that prevent you from accurately fitting reality.

Thus your standard New Keynesian model will use Calvo pricing and model the current inflation rate as tightly coupled to the present value of expected future output gaps. Is this a requirement anyone really wants to put on the model intended to help us understand the world that actually exists out there? Thus your standard New Keynesian model will calculate The expected path of consumption as the solution to some Euler equation plus an intertemporal budget constraint, with current wealth and the projected real interest rate path as the only factors that matter. This is fine if you want to demonstrate that remodel can produce macroeconomic pathologies. But is it a not-stupid thing to do if you want your model to fit reality?

I remember attending the first lecture in Tom Sargent’s evening macroeconomics class back when I was in undergraduate: very smart man from whom I have learned the enormous amount, and well deserving his Nobel Prize. But…

He said that in this class we were not going to make the mistake of simply putting money in the utility function: that would be ad-hoc. We were not allowed to wave our hands and say that people had a demand for money because holding money allowed them to take advantage of unexpected market parking is an so increase their utility. Instead, we had to explicitly model just what transactions holding money allowed you to undertake that you could not undertake otherwise. Hence, he said, we were going to build a rigorous, micro founded model of the demand for money: We would assume that everyone lived for two periods, worked in the first period when they were young and sold what they produced to the old, held money as they aged, and then when they were old use their money to buy the goods newly produced by the new generation of young. Tom called this “microfoundations” and thought it gave powerful insights into the demand for money that you could not get from money-in-the-utility-function models.

I thought that it was a just-so story, and that whatever insights it purchased for you were probably not things you really wanted to buy. I thought it was dangerous to presume that you understood something because you had “microfoundations” when those microfoundations were wrong. After all, Ptolemy had microfoundations: Mercury moved more rapidly than Saturn because the Angel of Mercury left his wings more rapidly than the Angel of Saturn and because Mercury was lighter than Saturn…

1559 words