Turbulence and unemployment in matching models

Steven Weinberg (2018) says: (1) new theories that target new observations should be constrained to agree with observations successfully represented by existing theories; and(2) preserving successes of earlier theories helps to discover unanticipated understandings of yet other phenomena. Weinberg’s advice helps us to answer the question: how do higher risks of skill losses coinciding both with involuntary layoffs (“layoff turbulence†) and with voluntary quits (“quit turbulence†) affect equilibrium unemployment rates? An earlier analysis that had included only layoff turbulence had established a positive relationship between turbulence and the unemployment rate within generous welfare states, but the absence of that relationship in countries with stingier welfare states. A subsequent influential analysis found that even very small amounts of quit turbulence would lead to a negative relationship between turbulence and unemployment rates. But that finding was based on a peculiar calibration of a productivity distribution that generates returns to labor mobility that make the model miss the positive turbulence-unemployment rate relationship that has been a theoretical basis for explaining the the persistent trans-Atlantic unemployment divide that emerged in post 1970s data and also miss observations about labor market churning. Repairing the faulty calibration of that productivity distribution not only brings models with quit turbulence into line with those observations but also puts the spotlight on macro-labor calibration strategies and implied returns to labor mobility.