Non-Walrasian Equilibrium Theory

Print/Save PDF

Blog readers know that the earliest version of the GEM Project’s generalized-exchange modeling played a seminal, albeit underappreciated, role in the brief period of active research on Efficiency-Wage Theory (EWT). My 1981 article and 1984 book introduced that literature to the nonconvex Workplace-Exchange Relation (WER) that describes rational employer-employee interaction in firms restricted by asymmetric on-the-job information and routinized jobs. After a decades-long hiatus, during which the Shapiro-Stiglitz (1984) market-centric model captured the efficiency-wage name while burying its stabilization relevance, the GEM Project finally figured out how to microfound  nonconvex WERs, belatedly producing one of the most consequential economic tools of the Keynesian era.

This post looks past EWT in order to examine another important macro school for which GEM modeling makes a seminal contribution. In his A History of Macroeconomics (2016),Michel De Vroey named this literature, dating from the final third of the 20th century, Non-Walrasian Equilibrium (NWE). A branch of the better known Disequilibrium Macro school, NWE theorists were mostly European, including such leading lights as Jacques Drèze, Edmond Malinvaud, Jean-Pascal Bénassy, and Jean-Michel Grandmont. An instructive overview of NWE thinking was edited by Jacques Drèze: Advances in Macroeconomic Theory (2001). 

From De Vroey: “[The NWE] approach displayed three main features. The first was wanting to construct a microfounded (i.e., choice-theoretically based) analysis. Next, the task non-Walrasian equilibrium economists set out for themselves was to construct general equilibrium models based on the assumption of [wage] rigidity and displaying involuntary unemployment as viewed by Patinkin, that is, with trade occurring off the labor supply curve. The justification for this endeavor was the belief that rigidity – or, in a lesser form, sluggishness – and involuntary unemployment were facts of life that could not be set aside [and can properly be assumed]…. The third feature is that, while non-Walrasian equilibrium economists found their inspiration in reading Patinkin, Clower, and Leijonhufvud, the line they took departed from what these economists had in mind, namely to provide a disequilibrium explanation of involuntary unemployment. By contrast, non-Walrasian economists were striving to produce an equilibrium result, a radically different project. Hence their dissatisfaction with the early ‘disequilibrium’ label and shift to the non-Walrasian equilibrium label.”

Despite the widespread respect given to the  NWE theorists, interest in their school relatively quickly died out. The reason is familiar. Mainstream New Keynesian thinking, fundamentally guided by the New Neoclassical Synthesis, insists that wage rigidity be microfounded. WR, no matter how factual, cannot be properly assumed. Consistency with the available evidence is an inadequate justification in acceptable modeling.  

Did leading NK theories believe there would be no damaging consequences from being all in on rational-behavior modeling at the expense of attention paid to the evidence (and policymaking relevance)? Among the most regrettable outcomes, a great deal of insightful research (well beyond NWE) has been marginalized for assuming factual wage rigidity, e.g., Barro and Grossman’s once famous Keynesian consumption model (1971). That intolerable situation will eventually change now that the generalized-exchange theory has rooted powerful WR in the neoclassical tenets of optimization and equilibrium. Non-Walrasian-Equilibrium macro modeling can, and should, be revived.

But there is an more important message than simply rescuing NWE thinking. The formidable new GEM Project economic tool, i.e., rational nonconvex workplace exchange, provides essential guidance to all theorists seeking to rigorously explain macro behavior in highly specialized economies. Illustrative outcomes of that more powerful modeling include:

  • Nominal downward wage rigidity and its associated mass involuntary job loss occurring in response to adverse demand disturbances both exist and are exclusively located in firms restricted by information-challenged workplace exchange and routinized jobs – surely valuable information for business-cycle modelers and policymakers.
  •  DWR is conditional on stationary cyclicality. In circumstances of nonstationary nominal demand contractions (a known condition of depressions), DWR rationally (and gradually) gives way to nominal wage cuts. The speed of that response partly depends on the severity of the spending collapse. Once spending stabilizes, DWR reasserts itself.
  • Prevention of nonstationary demand contractions is identified as the proper primary objective of stabilization policymaking.
  • Rational wage rigidity crucially has a second cojoined component, pure wage rent. PWR, also located in information-challenged workplaces, describes rational labor payment chronically in excess of market-clearing pricing.

That is just a taste; there is much more. But pause here to ponder what chronic wage rent implies for textbook economic theory.  Can there be any serious doubt that microfounded PWR immediately becomes one of the most consequential analytical tools constructed in the aftermath of the Second Industrial Revolution? The absence of the second (workplace) venue of rational exchange, an intuitively sound idea, has prevented the more timely emergence of PWR. The GEM Project’s generalization of rational exchange from the marketplace to information-challenged workplaces has enabled a powerfully improved rational-behavior macroeconomics that is uniquely consistent with a full range of important evidence. NWE theorists should be pleased.

Blog Type: Non-Market Macro Modeling

Write a Comment

Your email address will not be published.