I greatly admire Michel De Vroey. The Italian economist has done more than anyone else in the world to keep Early Keynesianism and its keystone meaningful wage rigidity alive in the age of market-centric neoclassical macroeconomics. His essential book, A History of Macroeconomics from Keynes to Lucas and Beyond (2016), stars John Maynard Keynes and Robert Lucas. In De Vroey’s telling, Keynes commands the first thirty years, roughly 1940–70, while Lucas dominates the next thirty. He interprets the two most recent decades as an uneasy balance between Keynesians and Lucasians
Critical milestones. The development of macroeconomics within De Vroey’s chronological divides is punctuated by milestones. The first, of course, is Keynes’s own General Theory and the related work of the estimable Early Keynesians. Keynes focused on the 1930s global depression and used his masterwork to reorient macro theory. By assuming away his Second Classical Postulate, i.e., the equality between wages and the value of the marginal disutility of work, he engineered a reversal of neoclassical real-to-nominal causation, establishing the macro centrality of aggregate nominal demand. EK theorists, including Modigliani, Patinkin, Klein, Samuelson, and Solow, pursued the reversed-causation analysis, positing short-term meaningful wage rigidity (MWR) as the keystone of their mainstream Neoclassical Synthesis. That time-dependent separation of neoclassical and Keynesian causations generated stabilization theory uniquely capable of suppressing wage recontracting and incorporating involuntary job loss (resulting from weakening aggregate demand), persistent unemployment, recognizable business cycles, and discretionary management of aggregate demand into mainstream textbooks.
De Vroey’s second milestone is rooted in the late-1960s work by Phelps and Friedman on the natural rate of unemployment. His third is the “long-lasting” attempt by Clower (1965), Leijonhufvud (1968), Barro and Grossman (1971), Benassy (1975), Dreze (1975), Malinvaud (1977) and others to microfound Early Keynesianism within the consensus market-centric general-equilibrium model class. De Vroey argues that their innovation was “disequilibrium” clearing of markets by long-side rationing when wages or prices are rigid. The disequilibrium school ran out gas as theorists gradually gave up trying microfound MWR within the market-centric analytic framework.
That failure to rationally motivate wage rigidity, combined with the growing impatience of a younger generation of theorists to work within macroeconomics that is coherent with established microeconomics, ushered in the Lucasian era. Milestone four is the construction of the Dynamic Stochastic General-Market-Equilibrium (DSGME) framework that enabled micro-coherent macro analysis. Led by Lucas (1972), the anti-Keynesians emphasized rational expectations and market clearing, spurning as arbitrary the EK keystone assumption of meaningful wage rigidity.
De Vroey identifies milestone five as a collection of related ideas in the 1970s and 1980s that explored imperfections in markets largely arising from incompleteness or private information. Those frictions were used to motivate a variety of analyses, including labor contracts, credit rationing, and efficiency wages (the subject of last week’s post). Contributors notably include Fischer (1977), Taylor (1980), and Stiglitz and Weiss (1981).
Before market frictions had time to dominate the macro literature, they were swamped by the next wave of ideas, represented by the fully microfounded and calibrated Real Business Cycle (RBC) model Kydland and Prescott (1982) that restored neoclassical real-to-nominal causation, suppressing market disequilibrium and the Keynesian role of money. Milestone six can be understood as the second coming of DSGME, now cast in the RBC framework constructed on the Cass–Koopmans model of optimal economic growth. Augmented by stochastic nonmonetary shocks to the aggregate production function, Cass–Koopmans can be made to produce fluctuations, albeit feeble and truncated, in national income and its components. Many macroeconomists were pleased, some extraordinarily so, that a bit of cyclicality had been made consistent with continuous market equilibrium.
By contrast, central bankers and other stabilization authorities were alarmed by the growing acceptance of a deeply unrealistic description of causes and consequences of macro instability. Policymakers were unsurprisingly receptive to the last (seventh) milestone on De Vroey’s list. This New Keynesian innovation occurred mainly in the late 1990s and early 2000s and marks the second, more ambitious wave of rational market frictions, this time more explicitly in the context of general market equilibrium. Moreover, and more quietly, NK theorists carved a critical hole in full micro-foundations by patching together a partial return of Keynesian thinking that arbitrarily reasserted nominal wage rigidity and the adaptive-expectations Phillips curve. Among the NK theorists who thereby reclaimed the macro mainstream are Calvo (1983), Taylor (1993), Galí (1999), Rotemberg and Woodford (1997), and Christiano, Eichenbaum, and Evans (2005).
The GEM Project identifies the crucial New Keynesian departure from Early Keynesian thinking to be the former’s unwillingness to assign priority to research on MWR microfoundations. NK theorists prefer to pay as little attention as possible to their wage-rigidity assumption, sweeping its inherent irrationality under the rug. Relying on glossed-over, shallow guesswork about the crucial wage determination process, which NK theorists believe must occur wholly in the marketplace, has burdened macroeconomics with a wildly inaccurate version of MWR. The damage caused by irrational labor pricing to the stabilization-relevance of mainstream NK macro theory turned out to be fatal.
Critical mistakes. As companion to De Vroey’s catalogue of milestones, the remainder of this post provides my list of critical mistakes made in the development of macro theory:
- Critical mistake #1: The EK belief that rational MWR can be rooted in the neoclassical market-centric general-equilibrium model class is the most fundamental error. Adequate descriptions of wage determination in modern economies featuring costly, asymmetric workplace information require the generalization of rational exchange from the marketplace to highly-specialized workplaces. If mainstream theorists were to learn one lesson, this would be it.
- Critical mistake #2: EK reliance on “disequilibrium” analysis was wrong-headed. Optimization, equilibrium, and market-centricity have long been the three fundamental tenets of macroeconomics, providing the core foundations of analytic rigor. Market-centricity, not rationality, is the root cause of stabilization-irrelevance and must be eliminated in modern analysis.
- Critical Mistake #3: Promising that RBC modeling, with its neoclassical real-to-nominal causation, would (with on-going refinement) produce something that looks like actual postwar business cycles, let alone an episode of extreme instability such as 2008-09 or the 1930s, is a mistake of epic proportions. At the very least, it reflects extraordinary innocence. RBC advice to stabilization policymakers is inherently dangerous. The sort of monetary-policy interventions that prevented the 2008-09 Great Recession from morphing into an unimaginably devastating depression has never had, and never will, a place in the Cass–Koopmans market-centric model.
- Critical Mistake #4: The mainstream New Neoclassical Synthesis, embracing market-centricity and proper microfoundations, forces NK theorists to search for a super market friction capable of rationally suppressing wage recontracting. Given that the GEM Project has demonstrated that rational MWR is an inherently nonmarket phenomenon, the NK quest has always been hopeless. A start to better macro thinking would be to stop relying on voluntary unemployment to model involuntary job loss.
- Critical Mistake #5: The mainstream use of the market-centric wage model of Stiglitz and Weiss (1981) to represent Efficiency Wage Theory fatally misdirected the most promising attempt to microfound MWR, building on the generalization of rational exchange from the marketplace to information-challenged workplaces. Like almost everybody else today, De Vroey makes this mistake in his history of macro theory. As described in last week’s post, the Stiglitz-Weiss rerouting of EWT to market-centricity doomed the efficiency-wage effort to microfound MWR.
Blog Type: Wonkish San Miguel de Allende, Mexico