Understanding the 2007-09 Great Recession centrally requires modeling that explains GR evidence. Involuntarily lost jobs in the millions heads the list. The GEM Project has constructed a exceptionally powerful model of labor-market failure, featuring downward wage rigidity (DWR) consistent with optimization and equilibrium. The model is rooted the generalization of rational price-mediated exchange from the marketplace to workplaces restricted by asymmetric employer-employee information. The Project’s approach builds on the economic maxim that price rigidity forces market adjustments to focus on real variables. By contrast, mainstream New Keynesians (NK) construct their GR analysis on foundations featuring how unemployed workers search for jobs and are matched with existing vacancies. Such unemployment is not dependent on what appears, from their market-centric perspective, to be rationally inexplicable DWR. It’s OK to dump inconvenient maxims in defense of textbook modeling.
To be fair, there is really no other choice for theorists relying on mainstream friction-augmented general market equilibrium (FGME) to guide their model-building. Being even more kind, I recognize that most macroeconomists know (at least down deep) that the market-centric approach is doomed to failure. As Robert Lucas bravely argued, involuntary unemployment is obviously incompatible with general market equilibrium. Stubborn FGME theorists are stuck with figuring out how to make the voluntary joblessness produced in search/match theory fluctuate at least a bit like business cycles.
This post focuses on the first part of search/match theory. It draws on Roger Farmer’s How the Economy Works. Farmer’s insightful analysis of how stock-market confidence plays a significant role in extreme instability is familiar to readers of this Blog. Unfortunately, his book is decidedly less successful in its attempt to explain the mass labor-market failure that characterize garden-variety recessions let alone the Great Recession. Like other mainstream theorists he relies on search theory to motivate the existence of unemployment in the millions; but, unlike other modelers, he wants to be “plain-speaking”.
Farmer begins that doomed effort by citing the fundamental stabilization dilemma that plagues NK explanations of macro instability: “When there are many unemployed people, the wage does not fall to restore equality between the quantity of labor demanded and the quantity supplied. Surely these unemployed people could offer to work for a lower wage than existing workers. It would be in the interest of a firm to hire an unemployed worker because it would be able to make a profit from the transaction. But this argument doesn’t take account of the costs of matching a worker with a vacant job.” That’s how Farmer introduces search theory, the core NK labor model that critically supports mainstream analysis of macro instability that dismisses meaningful wage rigidity.
Farmer’s How the Economy Works
From Farmer: “How do workers find jobs? The answer is not as obvious as it seems. To answer it, we must think hard about the time and effort that must be expended by workers and firms to find each other…. A job is produced from the time spent searching for each other by a firm and a worker…. The inputs are the search time of an unemployed worker and the search time of a recruiter in the personnel department of a firm.”
Reality check. Large-firm practitioners know that applicants for routinized-job vacancies typically provide basic personal information and recount their job experience. Wages are posted, not bargained, Large firms’ processing applicants is a continuous exercise; vacancies are typically occurring somewhere in their operations. Hiring (with probationary periods) decisions are quickly made. The process is widely understood as neither time- nor effort-intensive. The system easily handles fluctuations in hiring; the notion that resources are transferred back and forth from production to the personnel department is laughable.
Workers searching for work are by definition unemployed. Mainstream search theory contends that the increase in unemployment in the Great Recession resulted from more workers looking for work combined with an increase in the time required to process applications. Note that Keynes’ focus on involuntary unemployment has been pushed aside. In search theory, unemployment is voluntary. Involuntary lost jobs are nowhere to be found in mainstream FGME thinking; workers who insist that they have been forced out of their jobs are assumed to misunderstand the actual process at hand. More bluntly, they do not know what they are talking about. There is also no place for the surely important fact that existing workers, not job applicants, account for almost all rational-behavior market wage rigidity and recession-induced unemployment.
Fact check. Once tasked to explain mass joblessness in recession, search story is deeply inadequate. The macro academy surely knows that their go-to search theory contradicts a pile of critical facts. They also know to avoid its mention to practitioners. In particular, they know never to claim in the presence of practitioners that the millions of workers who separated from their jobs in 2008-09 somehow did so voluntarily. The search instability theory is unrecognizable foolishness in firms it supposedly describes.
Robert Shimer’s Labor Markets and Business Cycles
Perhaps Farmer’s plain-speaking treatment of search theory fails to capture important elements of the search/match literature, which after all has been awarded multiple Nobels. Attempting to play fair, this post turns to what I believe remains the literature’s best summary of modern S/M thinking, Robert Shimer’s 2010 book. To conserve space, my critique of Shimer and the entire NK search/match literature focuses on job separation. But let’s be frank. Even in its most sophisticated treatments, the search/match explanation for mass unemployment in the Great Recession, or any other downturn, is indeed foolish. It is an analytic debacle that mainstream instability theorists must abandon.
From Shimer (pp.104-105): “This section extends the search model with capital to consider the impact of fluctuations in the employment-exit probability…. Such shocks have a direct impact on the employment and unemployment rates by affecting the flow of workers from employment to unemployment. These shocks do not play any role in a model without labor market frictions since they can be costlessly reversed, but they can be an important contributor to aggregate employment fluctuations in an environment where finding a job is time-consuming….
“There are two basic approaches to introducing fluctuations in the employment-exit probability into this model. The first is to generate the fluctuations endogenously. In Mortensen and Pissarides (1994), employed workers are subject to idiosyncratic productivity shocks and quit their job when the idiosyncratic shock is too bad. Shocks to aggregate productivity change the endogenous threshold for exiting employment, a potentially important amplification mechanism. The second, simpler approach treats the employment-exit probability as an exogenous shock, possibly correlated with productivity…. I follow the simpler approach here, allowing for the correlated shocks to the deviation of productivity from trend and the deviation of the employment-exit probability from its normal value.” Each choice in Shimer’s introduction to job separation ignores the actual process that is evident both in the data and in descriptions of practitioners.
The GEM Project’s identification of downward wage rigidity as a structural outcome of LEV profit-seeking labor pricing is the missing element in the NK story. Modern S/M theorists choose to ignore mass involuntary job loss and, consequently, have little that is useful to say about its nature and implications when the crucial phenomenon occurs in the millions.
Two points about Shimer’s approach to the problem of mass labor-market failure are most relevant. First, the explanation is as good as it gets in the mainstream literature. Second, his analysis does not come close to passing practitioners’ laugh test. Since actual behavior is captured in the economic data, the mainstream story also disappoints careful applied macroeconomists. And, repeating for emphasis, it is difficult to imagine that search theorists themselves believe it. (I know that Farmer doesn’t.) Shimer et al. simply ignore what everybody knows: Voluntary quits do not cause rising unemployment in recession.
S/M theorists, despite their Nobels, cannot explain why huge numbers of rational employees quit their jobs in order to search for an alternative while total employment is contracting. It is unsurprising that countercyclical quits, the endogenous engine of S/M/B modeling, is nowhere to be found in the evidence. In the 2007-09 Great Recession, involuntary job loss (IJL) exceeded six million. Centrality of IJL is an indisputable fact. For each of the eight downturns since detailed job-separation data became available, forced job loss was by far the principal cause of rising joblessness.
Involuntary job loss is not a subtle, easily overlooked occurrence. When occurring in the millions, it is really big news that stirs national attention and debate. Ignoring IJL in the construction of macro models of employment stability requires willful theorist action. We must ask why so many economists, especially those who aspire to policy-relevance, are so determined to look the other way. Why do they stubbornly choose to be so obviously out of touch with what is actually happening? Shimer’s personal choice to eschew overtly relying on quits to generate job cyclicality by making job separation an exogenous black box does not effectively deal with the problem of either fact or conscience.
Contrast Shimer’s story with the GEM Project, which provides a powerful alternative to the dead-end use, both logically and empirically, of the S/M model class to explain job instability in modern, highly specialized economies. It identifies that today’s mainstream macro theory falls short almost wholly because of the ill-considered consensus assumption that restricts rational, price-mediated exchange to the marketplace. Once that groundless free parameter is played, for no better reason than model-building convenience, NK modeling cedes any rational claim to policy relevance.
How did smart, serious macro theorists get stuck with implausible stories about important phenomena? Why are we stuck with folly when proper modeling is so obvious? I believe the answer to be that we have pushed aside what we all used to know: Any adequate analysis of mass unemployment must feature wage rigidity capable of suppressing labor-price recontracting. The GEM Project’s MWR story has a lot to recommend it, notably including its consistency with the huge best-practices management literature. Both offer the same description of how highly-specialized economies reconcile differing employer and employee objectives in the aftermath of the Second Industrial Revolution and the consequent advent of large, specialized, bureaucratic corporations. Theorists who care about stabilization relevance must reject restricting job-exit analysis to market-centric modeling that cannot rationally accommodate meaningful wage rigidity. The GEM Project, deriving MWR as a continuous-equilibrium phenomenon from axiomatic model primitives, can help modern theorists become stabilization-relevant without giving up micro-macro coherence. Isn’t that too good to turn down?
Blog Type: New Keynesians
Write a Comment