This post resumes the GEM Blog’s two months of Ptolemy. The review has tapped into a seemingly endless supply of implausible analytic reaching needed to accommodate relevant evidence within micro-coherent market-centric mainstream modeling. The embarrassing hoops ancient astronomers jumped through in order to sustain their geocentric theory of the solar system, the original Ptolemaic research agenda, have been insufficiently cautionary to modern macro theorists. The message, however, should not be ignored. There must be a limit to the defense of consensus thinking that the evidence reveals, again and again, to incorrectly describe actual stabilization phenomena. The reason Ptolemy and his school could not keep the earth at the center of the solar system is that it is not at the center of the solar system.
Today’s macro consensus, populated by New Classical, RBC, and New Keynesian theorists, is tempting the ignominious fate of Ptolemaic astronomers. In order to be stabilization relevant, the modern mainstream most needs some micro-coherent means to reestablish Keynesian nominal-to-real causation. The substantial NK effort to identify one or more market frictions that can microfound meaningful wage rigidity (MWR) has failed; it never coming close to rationally suppressing labor-price recontracting in market-centric thinking. Mainstream research agendas have been organized around that quest. From Chari, Kehoe, and McGrattan (2009): “Modern macroeconomists … agree on which reduced-form shocks are needed for models to fit the data.” The reduced-form shocks most frequently cited are time-varying disutility of work and time-varying bargaining power. Other ‘market imperfections’, such as Calvo’s random product-price adjustment and Taylor’s staggered wage contracts, are no more than incoherent free parameters and are, as a result, rejected by consensus theorists.
The time-varying disutility of work argues that workers generally become more lazy in recession – an idea that has been inexplicably accepted by many theorists as a serious explanation for high-frequency employment behavior. Chari et al. cite Rotemberg and Woodford (1997); Erceg, Henderson, and Levin (2000); Smets and Wouters (2003); and Galí and Rabanal (2004). Galí and Rabanal’s variant of the standard market-centric DSGE model includes a crudely specified work-preference shock that “explained” 57 percent and 70 percent of real output and employment variance respectively.
In time-varying bargaining power, workers are hypothesized to become more greedy (or firms less greedy), altering their use of labor-pricing power, in recession. This convenient, albeit baseless, assumption has also been offered as a serious explanation for the high-frequency behavior of employment by serious mainstream macroeconomists. They notably include Smets and Wouters (2003 and 2007); Hall (2005b); Galí, Gertler, and Lopez-Salido (2007); and Smets and Wouters (2007). In the last example, wage-markup movement (thought to represent bargaining-power shocks) accounted for a fifth of output variation and more than half of inflation variation.
Who is this literature trying to fool? Modern labor-wedge interpretations of high-frequency evidence never rise above the Ptolemaic. They provide a sharp contrast to the meaningful wage rigidities derived from axiomatic model primitives in the GEM Project that directly, powerfully, and intuitively explain the employment-variation facts in question. Countercyclical preferences or market power makes no sense other than to defend the authors’ micro-coherent market-centric DSGE priors that job separations are always voluntary. Miscasting voluntary separations to play a starring role in actual business cycles must be particularly disheartening to macroeconomists who aspire to policy relevancy. While the miscasting rescues mainstream theorists from what they have learned to be a difficult, unrewarding task (i.e., rationally suppressing wage recontracting), the price of that convenience is unacceptably high. Theorists are required to pretend ignorance of the overwhelming evidence that quits are never a cause of rising job separation and unemployment in recession. They must also ignore that stable preferences are a central building block of micro-coherent rigorous economic analysis, eligible for compromise only in the face of incontrovertible evidence. The facts do not, and never will, support preference shift against consumption in recession. We all know, at least deep down, it to be a foolish assumption. It is embarrassing to pretend otherwise. This literature is unmistakably a Ptolemaic convenience.
The variable bargaining-power argument, additionally hypothesized to produce large countercyclical movement in quits, is also indisputably Ptolemaic. That recession pushes workers to become irrationally obstinate, refusing to recontract in order to save their rent-paying jobs, is no more convincing than the countercyclical shifts in reservation wages resulting from time-varying disutility of work. Who does not understood that the relevant power here is the capacity to impose job loss on workers who rationally prefer wage cuts? Mainstream labor-wedge interpretations are taken seriously by nobody other than mainstream macroeconomists and by them only because the convoluted logic dresses up the evidence to be a tiny bit more aligned with micro-coherent market-centric DSGE theory. Neither this nor the earlier model-building convenience passes the laugh test.
The message is harsh but unavoidable. The casual use of labor-wedge reduced-form shocks to satisfy the stabilization-policy ambitions of mainstream theorists can never be more than a Ptolemaic exercise, constructing and disguising free parameters that defend consensus theory. Mainstream gatekeeper acceptance of incoherent, factually incorrect, make-believe rigidities organized around counter-cyclical variation in work disutility or bargaining power, while continuing to reject the Early Keynesian strategy of simply positing MWR, struggles to rise above intellectual fraud.
Labor-wedge capacity to closely track employment and production, despite being a finding of questionable consequence (given its empirical doppelganger’s incorporation of average hours at work), has encouraged troubling over-reaching by mainstream macro theorists. By pushing their consensus model into stabilization policymaking, they badly exceed its very modest limits, provide damaging advice, and reduce their research to being the worst sort of Ptolemaic exercise. The GEM Project, generalizing exchange and thereby enabling the derivation of meaningful, continuous-equilibrium nominal wage rigidity, provides a coherent, stabilization-relevant, and ultimately face-saving way out of the destructive mainstream macro quagmire.
Blog Type: Wonkish Saint Joseph, Michigan