I was culling my library last week and ran across Peter Chinloy’s Labor Productivity (1981). I remembered that the book, nearly four decades ago, clearly illustrated the difference between insightful and misleading interpretation of evidence. Its lesson is even more relevant today.
Chinloy’s book is one of the literature’s many exercises in Solow accounting, which attempts to identify the nature and size of various influences on productivity growth. Since such growth is the wellspring of rising living standards, the rapid accumulation of publications was justified by the importance of the subject. Chinloy concluded that labor quality played an outsized role in the postwar behavior of productivity. He measured quality by sex, age, education, occupation, class of worker, and relative wages – the last used to construct an index of labor quality. During the entire period of investigation (1947-1974), his labor quality was estimated to increase by 0.6% per year, accounting for more than two-fifths of labor-input growth. Chinloy did the same exercise for the subperiod 1971-74, finding that the contribution of labor quality to the total gain in labor input was nil. That substantial change largely informs his policy conclusions that directly target labor quality.
In the GEM project, microfounded meaningful wage rigidity (MWR) greatly helps in interpreting the deterioration in labor input and productivity in 1971-1974. The GEM analysis usefully recognizes that the early-1970s short period is dominated by a sharp slowdown in economic activity that was resulted from weakening effective demand that was in part engineered by the Federal Reserve. (I know because I was there and can remember the frustration of then-Chairman Arthur Burns that we couldn’t get the 1974-75 recession going sooner.) The slowdown in total spending combined with MWR in large, highly specialized firms to force them to cut production and lay-off millions of employees. The GEM Project also demonstrates that the affected employees were rationally paid significant wage rents, enabling their employers to cream job applicants. It follows that laid-off workers would both weaken Chinloy’s measurement of labor input and be collinear with his measures of labor quality. Moreover, his index of labor quality, constructed on relative wages, would be directly damaged by the huge number of involuntary job losers who had been receiving wage rents. (Relative wages are useful as an indicator of labor quality in the circumstances of general market equilibrium, which Chinloy assumes. Nobody who is familiar with 1971-1974 actually believes it to be a period of market equilibrium.) Chinloy’s faulty interpretation of the evidence led to, to be nice, misleading policy recommendations.
Labor Productivity was published in 1981. He would have been trained in Early Keynesian macroeconomics that featured evidence-relevant instability. Why didn’t he see the obvious? His work on the Solow growth-accounting problem measures macro instability, not policy-relevant changes in labor quality. Chinloy was caught up in a sad story. By the end of the stagflation decade, the macro academy was deep into the pursuit of a Ptolemaic objective, insisting that published work demonstrate fealty to the neoclassical model class, featuring general market equilibrium, at the cost of stabilization relevance. Readers of the GEM Blog know that a rational, market-centric framework cannot accommodate MWR and recognizable recessions. In the emerging neoclassical consensus, the Solow growth model is constructed on market-centric general-equilibrium and simply posits continuous full employment.
In the early 1980s, just prior to my Price of Industrial Labor, I believed that the work being done of morale-centric efficiency-wage theory would lead to proper microfoundations for meaningful wage rigidity. Properly motivated EWT, given a fair hearing, would close the growing and dangerous practice of badly misinterpreting macro evidence. I was, of course, disappointed.
Today, the question is whether the disappointment will persist. Given that the GEM Project has made MWR consistent with optimizing exchange organized by general decision-rule equilibrium, will the damaging, chronic misinterpretation of the macro evidence to be consistent with the mainstream market-centric general-equilibrium model class continue? If so. why are mainstream New Keynesians content with stabilization irrelevance and bad policy advice?
The GEM Project’s most basic message is that rational-behavior models are the best, indispensable guidelines for interpreting evidence. But rational behavior was altered by the Second Industrial Revolution. Optimizing exchange can no longer be usefully restricted to the marketplace. Just as the Ptolemaic attachment to market-centric analysis misled Chinloy in his analysis of the productivity slowdown, it has generally prevented mainstream NK theorists from constructing macroeconomics that is simultaneously micro- coherent and stabilization relevant.
Blog Type: Wonkish Saint Joseph, Michigan