This post extends last week’s essay, using the troubles of mainstream macroeconomics to illustrate a message of the GEM Project: Rational-behavior theory must replace the modern dependence both on assumptions of convenience and on models that are “patently false” (Blanchard’s description of two of the three equations in the basic go-to NK analytic framework). The Project’s essential point is that analysis rooted in optimization and equilibrium provides crucial guidance to macro model-building, necessary to realize the field’s explanatory potential.
There is little doubt that NK market-centric general-equilibrium theory provides inadequate model-building roadmaps, making macro theorists vulnerable to debilitating wrong turns. Failure is especially acute when market-centric models are used to explain optimizing exchange occurring inside highly specialized firms. Market-centricity has been a wellspring of harmful macro model-building guidance for more than a century. Olivier Blanchard’s paper describing the state of stabilization theory as “good”, our topic a week ago, illustrates the Project’s case. OB makes many wrong turns as he takes direction from badly constructed mainstream analysis. A few of the most egregious missteps take up the remainder of this post.
Existence of unemployment. From Blanchard: “One striking (and unpleasant) characteristic of the basic NK model is that there is no unemployment! Movements take place along a labor supply curve, either at the intensive margin (with workers varying hours) or at the extensive margin (with workers deciding whether or not to participate). One has a sense, however, that this may give a misleading description of fluctuations.”
Is he putting us on? “One has a sense” that a macro stabilization theory without unemployment “may give a misleading description of fluctuations”? I am surely not alone in understanding that such a model is misleading, to the point of being dangerously wrong, in just about every sense possible.
NK unemployment modeling. Blanchard does point to some on-going research that he believes will set things straight on the “misleading description” front. Taking guidance from NK market-centric general equilibrium model, he then takes some serious wrong turns. “The first question is then how to think about and introduce unemployment in a macro model. Here, fortunately, we can build—and are building—on a parallel effort, developed over the past twenty years by, in particular, Peter Diamond, Chris Pissarides, and Dale Mortensen…. In this approach, unemployment arises from the fact that the labor market is a decentralized market, where, at any time, some workers are looking for jobs, while some jobs are looking for workers. This has two implications. The first is that, by necessity, there is always some unemployment—and, symmetrically, some vacancies. The second is that, as it takes time for a worker to find another job, and for a firm to find another worker.”
The bad guidance from mainstream NK market-centric theory is breathtaking. OB is describing frictional unemployment. He is betting the ranch on the capacity of frictional unemployment to explain layoffs and rising joblessness in recessions. That egregious wrong turn will not do, for at least one simple reason. FRICTIONAL UNEMPLOYMENT IS WHOLLY VOLUNTARY. OB must know – everybody else does – that most of the increase in unemployment in recession results from involuntary job loss. Forced layoffs are the famous business-cycle fact that was once at the center of Keynesian theory. A stabilization model absent layoffs is akin to staging Hamlet without the Prince.
OB engages in hand-waving to divert attention from involuntary unemployment: “… it makes one think of the labor market as a market characterized by large flows, flows of job destruction and creation, flows of workers between employment, unemployment, and non-participation. It allows one to think about the effects of labor market institutions on the natural rate of unemployment. It allows one to think about whether and how fluctuations affect reallocation, and whether some of the fluctuations themselves may be due to variations in reallocation intensity.” He eventually tires, never getting to involuntary job loss that always occurs in response to significant weakening in nominal demand. OB’s shameful exercise in misdirection is too frequent in the NK literature not to reflect deliberate intent.
It is useful to recall here that meaningful wage rigidity, by definition capable of suppressing wage recontracting, is a necessary condition for the rational existence of involuntary job loss. (See the website’s e-book, chapter 1.) One of OB’s big problems is that MWR must be microfounded to be acceptable in rigorous NK modeling-building. But he would rather not dwell on not knowing how to microfound MWR. It only complicates his problem that the GEM Project has indeed microfounded MWR, so that achievement does not much interest OB. As I understand it, the costs of adding a second workplace venue of rational exchange is too great an upheaval, costly in time, effort, and damage to existing human capital and professional reputations, to even consider as a legitimate research topic. Involuntary job loss is not that important.
Preventing future financial crises. Many NK theorists appear to believe that the key to dealing with extreme instability is to prevent future financial crises. In pursuit of that goal, they are working to augment mainstream market-centric general equilibrium modeling with a more robust financial sector. From OB: “The current financial crisis makes it clear that the arbitrage approach to the determination of the term structure of interest rates and asset prices implicit in the basic NK model falls short of the mark: Financial institutions matter, and shocks to their capital or liquidity position appear to have potentially large macroeconomic effects.”
A more developed financial sector may or may not be worth it. But one thing is clear. The expanded market-centric theory will not show us how to prevent future financial crises. It is not realistic to prevent future financial crises. (See the website’s e-book, chapter 6.) It is surprising that OB and his NK colleagues have not figured that out. Instead, the effective direction of stabilization policy is to focus on the powerful propagation of crises that is deeply rooted in the interaction of MWR and nominal demand disturbances. That is the vulnerable aspect in extreme instability macrodynamics; it is here that stabilization-authority intervention has proved to be effective. The generalized-exchange macro model, endowed with microfounded MWR, provides crucial guidance to focus on breaking up aggregate-demand propagation. Today’s mainstream market-centric, general-equilibrium theory does not.
Central-bank objectives. In his paper, Blanchard is pleased that NK theory “leads to a strong policy conclusion: Strict inflation targeting is good, both for inflation, and for output (a result Jordi Gali and I have baptized the ‘divine coincidence’). This result serves as an important benchmark. In the presence of further imperfections however, it may no longer hold. To take a very topical example, suppose labor market imperfections lead to more real wage rigidity than would be implied by a competitive labor market. Then, an increase in the price of oil—which requires a decrease in the real wage—may lead to a large decrease in the second-best level of output: Very low output, and thus a large increase in unemployment may be needed to make workers accept the real wage cut. First-best output, which is defined as what output would be without real wage rigidities, may move much less. In this case, it may be better to allow for some inflation, and a deviation of output above its second best level for some time, rather than to stick to constant inflation.”
This is Blanchard, confused after many wrong turns. He appears proud that mainstream NK analysis helped guide central banks to adopt a single (anti-inflation) objective. At the same time, he appears to recognize that the advice turned out to be really bad. Perhaps most important, he cannot help demonstrate that a central problem is the NK insistence that market-centric, general-equilibrium modeling is settled macro theory. It provides no useful guidance to help him make sense of the facts indicating that the state of macro theory is bad. Instead, it forces him to think about big issues wholly in the context of market centrality, leading him into a tangle of wrong turns.
Microfounded MWR and rationally suppressed wage recontracting break the simple market-centric trade-off between unemployment and product prices, which can then be understood to respond on average slowly to gathering economic slack. Product prices did not, in late 2008 or in 2009, signal urgency to the Federal Reserve. Urgency was instead signaled by the real-side phenomena of millions of involuntary lost jobs and contracting output in combination with collapsing asset markets. (For elaboration, see the GEM Project’s extreme-instability model in the website’s e-book, chapter 6.) Any suggestion that those signals should take a backseat in deference to the NK-deduced primacy of product-price inflation is both stupid and dangerous.
Blog Type: New Keynesians San Miguel de Allende, Mexico