I find the ubiquity and ambition of Search/Match/Bargaining theory alarming. Nearly all macro analysis that touches on the labor market relies on modern search theory that has been extended from its original focus on frictional joblessness to the stationary behavior of unemployment around its natural rate. Especially troubling is S/M/B theorists’ increasingly assertive claim of relevance to stabilization policymaking, therein ignoring a debilitating problem. Decades of S/M/B research has made clear what the GEM Project easily demonstrates: Absent free parameters, the go-to model class cannot accommodate involuntary job loss (IJL), recognizable involuntary unemployment (IU), and related business-cycle characteristics. S/M/B reach is effectively limited to voluntary unemployment and quits. As a result, its stabilization-policy implications are consistently misleading.
My examination of the S/M/B model class is divided into two parts. The first looks at its most debilitating problems. It is an unhappy exercise. The second, postponed to next week, answers questions that follow from the first. Why did mainstream theorists decide to pour so much time and effort into a model class that, even if developed to its highest extent, has no hope of being both coherent and stabilization-relevant? Why did prominent New Keynesians (NK) abandon the early Keynesian research program focused on nominal wage rigidity that, if pursued to a successful conclusion, would have coherently modeled recognizable cycles and yielded plausible stabilization policies?
Space limitations restrict this assessment of S/M/B troubles to be brief and incomplete. In-depth analysis can be found in the GEM Project’s eBook. (Chapters 4 and 6) But most readers will not need elaboration. The critical failures are self-evident, raising an uncomfortable question. Are mainstream NK theorists engaged in a deliberate attempt to mislead? They must by now know that the coherent S/M/B model class cannot explain forced job loss and involuntary unemployment, cannot explain how adverse demand disturbances induce IJL, and cannot justify central-bank discretionary management of total spending. They must know that S/M/B modeling cannot adequately support effective stabilization policymaking and that, as a result, the Federal Reserve simply ignored mainstream S/M/B thinking in its design and execution of policies to deal with the 2008-09 extreme instability. I had a front-row seat for the Fed’s magnificent performance during the Great Recession and can assure that the huge body of work on matching efficiency was wholly irrelevant. Anyone suggesting that more recruiters were needed to deal with the very real problem of six-million job losers would have been jeered out of the room.
We know, as a matter of neoclassical logic, that IJL existence requires suppression of rational wage recontracting. Absent suppression, optimizing employees must in lieu of losing their jobs accept nominal wage cuts that do not violate their opportunity costs. And, in consensus market-centric analysis, profit-seeking employers must offer such cuts. No market imperfection prevents the worker’s rational on-the-job choice to accept the wage reduction or the firm’s rational decision to offer it. In mainstream market-centric DSGE modeling, IJL does not exist. That inherent nonexistence is a huge stabilization problem for S/M/B theorists, especially in light of the overwhelming evidence that identifies IJL as the engine of rising unemployment in recession.
The GEM Project introduces continuous-equilibrium IJL into coherent macroeconomics via the generalization of optimizing price-mediated exchange from the marketplace to the large-establishment (LE) workplace. The intuitively broader reach of rational exchange microfounds the meaningful wage rigidity, defined by its capacity to rationally suppress wage recontracting. (Chapter 2) MWR combines with nominal demand disturbances to generate IJL and recognizable cyclical unemployment. (Chapter 6) Coherent macro theory becomes stabilization-relevant.
On the few occasions that mainstream theorists consider the actual nature of cyclical joblessness, they either posit forced layoffs or create some unfailingly disingenuous diversion around endogenous job separation. The first strategy, illustratively used by Shimer (2010), unhappily forfeits model micro-macro coherence. The second is interestingly illustrated by the original leaders of the S/M/B movement. Mortensen and Pissarides (1994) squeezed some cyclicality into joblessness by subjecting employees to productivity shocks that sufficiently alter the threshold wage governing employment exit to induce them to voluntarily quit their jobs. The diversion strategy always twists its way back to voluntary quits.
Moreover, some NK theorists have been attempting to link MWR, albeit prior to understanding how to define it, to the labor-market frictions featured in S/M/B modeling. The most prominent effort has been to posit that match-efficiency imperfections rooted in idiosyncratic human capital create large hiring-process wage indeterminacies. That research program has become a minor growth industry, despite debilitating flaws:
- NK theorists vastly overestimate, for model-building convenience, the importance of match capital. For applicants seeking nonsupervisory production jobs, the lion’s share of LE employment, idiosyncratic human capital and related wage indeterminacies are trivial. (Chapter 4) Who doesn’t know that large firms post hiring wages for routinized jobs? Who doesn’t know that fact, by itself, destroys the NK bargaining narrative?
- Posit some alternative universe which produces significant routinized-job labor-pricing indeterminacy to be bargained. There is still no endogenous NK mechanism that enables rational negotiations to produce a determinant wage. The surplus division is instead governed by arbitrary assumptions that are no more defensible than early Keynesians’ assumption of wage stickiness. If the latter justified banishment, why is the former immune?
- In the alternative universe, any “rational” labor price emerging from idiosyncratic bargaining is inherently downward flexible. It permits wage recontracting and prevents involuntary job loss. NK indeterminate wages are not stabilization relevant.
- Labor-price bargaining with individual routinized-job new hires must create a chaotic wage structure that cannot be consistent with profit-seeking. The mainstream model does not pass practitioners’ laugh test.
In closing, there is another, even more fundamental problem with the NK strategy of linking MWR to labor-market frictions featured in the S/M/B model. The GEM Project microfounds a relevant, readily apparent fact. In highly specialized economies, LE wage determination rationally locates inside large establishments, severing its connection to labor-market imperfections. MWR linkage to product markets, however, is critically robust. (Chapters 2, 3, 6)
Blog Type: Wonkish Chicago, Illinois