A Mankiw Monday: On Healthcare and Keynes

Greg Mankiw is an economist with whom I often disagree when it comes to policy but nevertheless is someone who I think is an extremely smart and thoughtful purveyor of his discipline. It does not bother me that he often comes to different conclusions in policy matters, for this is generally a matter of opinion about the magnitude of the market failure and the ability of government to correct it efficiently and effectively. I tend to have more faith in government than he and see market failures as a bigger deal, but I always try to continually challenge my own assumptions and reading his take is always useful (and occasionally convincing).

Mankiw is also an exceptional writer and he had two articles in major newspapers over the weekend that are well worth reading. The first, in Sunday's New York Times, is a worthwhile piece on the realities of modern medicine and societies role is its distribution. An excerpt:

An optimist might hope that my doctor, or someone higher up in the health care hierarchy, made a rational cost-benefit calculation on society’s behalf. To figure out whether my treatment makes sense, one would have to weigh the cost of the drug against the benefit of an extended life. And to do that, one would have to put a dollar value on my life — the kind of calculation that makes everyone but economists squirm.

The second is a review of a new book on Keynes by the author of the authoritative three part biography on Keynes, Robert Skidelsky. Skidelsky's new book talks about the sudden and swift revival of Keynes's theories in the recent crisis. Mankiw, writing in the Wall Street Journal, is impressed with Skidelsky as an historian but less so with Skidelsky as an economist. He takes umbrage at Skidelsky's distain for math in economics:

To economists his discussion of macroeconomic theory will seem pedestrian and imprecise. To laymen it will seem abstract and hard to follow.

As an ardent fan, Mr. Skidelsky fails to give Keynes's intellectual opponents their due. In academic circles, the most influential macroeconomist of the last quarter of the 20th century was Robert Lucas, of the University of Chicago, who won the Nobel Prize in 1995. His great contribution to the discipline was to analyze how government policies influence the economy in part through their effect on people's expectations—a lesson that Keynes would likely have appreciated but that early followers of Keynes often ignored.

Yet Mr. Skidelsky chooses to make Mr. Lucas sound like some kind of idiot savant, more interested in playing with mathematical models than in trying to understand how the world actually works. Mr. Lucas, we are told, is following in the tradition of the "French mathematician Leon Walras [who] pictured the economy as a system of simultaneous equations." The very idea is made to sound slightly crazed.

This brings us to the biggest problem with "Keynes." Mr. Skidelsky admits to being poorly trained in the tools that economists use: "I find mathematics and statistics 'challenging,' as they say, and it is too late to improve. This has, I believe, saved me from important errors of thinking."

Has it, really? Mr. Skidelsky would like to think that his math-aversion allows him to focus on the big ideas rather than being distracted by mere analytic details. But mathematics is, fundamentally, the language of logic. Modern research into Keynes's theories—I have conducted such research myself—tries to put his ideas into mathematical form precisely to figure out whether they logically cohere. It turns out that the task is not easy.

Keynesian theory is based in part on the premise that wages and prices do not adjust to levels that ensure full employment. But if recessions and depressions are as costly as they seem to be, why don't firms have sufficient incentive to adjust wages and prices quickly, to restore equilibrium? This is a classic question of macroeconomics that, despite much hard work, is yet to be fully resolved.

Which brings us to a third group of macroeconomists: those who fall into neither the pro- nor the anti-Keynes camp. I count myself among the ambivalent. We credit both sides with making legitimate points, yet we watch with incredulity as the combatants take their enthusiasm or detestation too far. Keynes was a creative thinker and keen observer of economic events, but he left us with more hard questions than compelling answers.

I am like many PhDs of my generation in that I was raised on Keynes as an undergrad (and as a masters policy student) and then disabused of all things Keynes in grad school in economics where the 'real business cycle' school was dominant. To use the vogue term of art, I was taught by a freshwater economist (in this case a Minnesota-trained one). But I also had the very useful experience of having previously taken a graduate level macro class in Keynesian economics which was engaged in just the endeavor to which Mankiw refers: trying to apply modern mathematical logic to Keynes and see if it can survive. [The answer, by the way, was yes, but with some pretty strong assumptions that may or may not be realistic - however the notion is sticky prices and wages is pretty well established now and is perhaps the most important assumption to get Keynes in the modern economic world]

I agree completely with Mankiw in his take on math, it is indeed the language of logic and for that we can all be thankful that it has a prominent place in economics. And just as it can be taken too far, where mathematical elegance is prized above economic insight, dismissing economics too quickly as just mathematical abstraction is equally wrong.

Economist's Notebook: We Are All Keynesians Again

When I was in graduate school in the late nineties, the field of macroeconomics was having a bit of an identity crisis. Early twentieth century Keynesianism was out because it did not rely on the careful building up of a theory of aggregate economics from the simple rational economic agent that was the basis of modern neo-classical microeconomics. How could you, the argument went, have a macro theory that is disconnected to micro theory and still call it well-founded?

Thus, Robert Lucas, at the University of Chicago, and others set out to re-invent macroeconomics based on 'microfoundations.' Starting with the individually rational decision maker and adding up, in essence, led to a new macroeconomics that has been called many things but is most commonly described by the rubric 'real business cycle' theory (RBC). It had its origins in the monetarism of Milton Friedman, but was (is) a much more fully realized theory (and in that I mean, among other things, highly mathematical - which I consider a good thing in that mathematics helps prevent logical inconsistencies). A number of graduate programs became associated closely with it, Chicago, Rochester, and, most prominently (or perhaps better said, the department whose identity is most deeply enmeshed in RBC theory), the University of Minnesota.

It was a fresh Minnesota PhD who taught me most of my graduate macro (after a breathtaking few weeks with Bruce D. Smith who was an amazing teacher and researcher). The lessons from this training was clear: markets are efficient and people are rational. This means that the government has little role in the economy and will generally muck it up if it tries. This is a strong departure from the Keynesian macro of the past which implicitly assumed that markets had lots of frictions and would often find themselves in less than efficient outcomes, so the federal government has a very necessary role to play.

By the mid 1990s RBC macro was ascendant and Keynesians were struggling trying to proffer arguments why frictions existed and incorporate them into models that looked nothing like Keynsian ones. The most compelling of these frictions was the idea of 'sticky prices' - that prices and wages were slow to adjust to changing economic circumstances like the RBC models assumed. Union contracts, cognitive dissonance and the simple cost of reprinting menus we all given as evidence of this phenomenon (as was a wealth of empirical evidence).

It wasn't until a new class of macroeconomists came along, most prominently Michael Woodford and Julio Rotemberg, who were able to construct a microfounded model of Keynesianism (essentially a melding of the RBC models and some Keynesian principals) that this gulf was bridged -partly. This is still process and will take a generation - Minnesota is still pretty orthodox RBC, for example - and, unfortunately it came too late for me (Woodford's now canonical book Interest and Prices: Foundations of a Theory of Monetary Policy was published in 2003). These new 'Dynamic Stochastic General Equilibrium' models (DSGE) incorporate less than fluid prices but are essentially still based on monetary policy, not fiscal policy. [If you want a more thorough treatment of DSGE models, there is no better place than this]

These DSGE models are still mostly based on rational agents but with economic frictions. One way to view the current crisis is that it is being driven by a panic. This implies a degree of irrationality. The only place to turn for theories where agents are behaving less than perfectly rationally is to the macroeconomists working in the area of 'bounded rationality' who reject the perfectly rational agent in favor of agents who have a hard time figuring it all out and so use rules or learning mechanisms. George Evans of the U of O and his student, and my colleague, Bruce McGough of OSU work in these areas, but many mainstream macroeconomists are uncomfortable with such departures from the basic rational agent framework we are used to in economics.

So where does the modern macroeconomist turn for guidance on the effects of fiscal policy? Well, it appears that most of them are turning all the way back to 1936 and Keynes's General Theory. International economists (like Krugman) will claim that New Open Economy Macro (NOEM) does talk about fiscal policy, but I would argue that it is still mostly monetary based, and doesn't tell us a whole lot about our current situation. Keynes, on the other hand, speaks to just these types of systemic economic crises and offers solutions.

So we are all scrabbling around and opening up our intermediate macroeconomics textbooks (that still rely heavily on Keynes's model as a pedagogical tool). What I wonder is if the 'great moderation' in business cycles showed us that modern RBC theory was essentially right or was it the other was around, the serendipitous moderation of business cycles gave us false confidence in the RBC framework. Either way, we are in new territory now and virtually all modern, cutting-edge macro models don't help much in guiding us through these troubled times.

We are all Keynesians again.