Knowledge that Transforms

This article examines the determinants of changes in the U.S. wage structure from 1976 to 2000. Our main empirical observation is that changes in both the level of wages and the returns to skill over this period were primarily driven by changes in the ratio of human capital to physical capital. We show that this pattern conforms extremely well to a simple model of technological adoption following a major change in technological opportunities. In contrast, we do not find much empirical support for the view that ongoing (factor‐augmenting) skill‐biased technological progress has been an important driving force over this period.

It is common practice to use sensible rules of thumb for cleaning data. Measurement error is often the justification for removing (trimming) or recoding (winsorizing) observations where the dependent variable has values that lie outside a specified range. We consider a general measurement error process that nests many plausible models. Analytic results demonstrate that winsorizing and trimming are solutions for a narrow class of error processes. Indeed such procedures can induce or exacerbate bias. Monte Carlo simulations and empirical results demonstrate the fragility of cleaning. Even on root mean square error criteria, we cannot find generalizable justifications for these procedures.

Relicensing requirements for professionals who move across borders are widespread. In this article, we measure the effects of occupational licensing by exploiting an immigrant physician retraining assignment rule. Instrumental variables and quantile treatment effects estimates indicate large returns to acquiring an occupational license and negative selection into licensing status. We also develop a model of optimal license acquisition that, together with the empirical results, suggests that stricter relicensing requirements may lead not only to practitioner rents but also to lower average quality of service in the market for physicians.

I use panel data to examine whether long‐term changes in industry wages are positively related to long‐term changes in industry employment. Previous research using repeated cross‐sectional data found no systematic relationship between these variables. Using standard fixed effects models to deal with individual heterogeneity, I find a robust positive relationship between changes in composition‐constant industry wages and industry employment. This suggests that growing industries attract less skilled individuals in a manner that biases down the estimated relationship between industry employment and wages in repeated cross‐sectional data. The results imply that supply curves facing industries are elastic but upward sloping.

In certain industries, sales agent contracts include provisions for sales commissions and clawbacks of commissions if clients are not retained. We show that contracts with these features arise in environments having up‐front selling costs recouped from ongoing sales; heterogeneous customers; limited agent access to capital markets; and imperfect commitment to long‐term contracts. We test the model using information on insurance sales agent contracts from New Zealand prior to and after bank entry into insurance sales. The evidence indicates that banks cream‐skimmed customers. We predict that this should reduce the values of sales commissions and clawbacks. The data support this prediction.

Dans cet article, nous cherchons à développer un modèle par lequel le salaire d'un travailleur est fonction de ses qualifications. Le marché ainsi que le travailleur sont au préalable dans l'incertitude quant à certaines de ces qualifications. L'endogénéité à la fois des changements de salaire et des décisions de changements du secteur d'affiliation résulte du processus d'apprentissage relié aux qualifications du travailleur. Nous montrons ensuite comment le modèle peut être estimé par les méthodes des variables instrumentales non-linéaires. Nous appliquons notre méthodologie à l'étude des salaires et de l'allocation des travailleurs aux différentes occupations et industries. Nous trouvons que les secteurs à salaires élevés emploient des travailleurs ayant davantage de qualifications et que ces secteurs rémunèrent ces qualifications à un taux supérieur relativement aux secteurs à faibles salaires. Les estimés des rendements associés aux qualifications qui ne tiennent pas compte du fait que les rendements diffèrent d'un secteur à un autre sont par conséquent erronés. Nous proposons enfin d'autres applications possibles de notre méthodologie.

Turnover falls with tenure, but wages do not always rise (and sometimes fall) with tenure. We reconcile these findings by revisiting an old issue: how gains from firm‐specific training are split between workers and firms. The division is determined by a stationary distribution of outside offers. The lower the wage a firm pays to a specifically trained worker, the more profit it makes but the more likely the employee is to leave. The optimal time paths of wages and turnover show that, if marginal product is increasing, wages need not be increasing but it always implies a falling turnover rate.

In many countries, an expired labor contract is automatically extended during the often‐protracted delay before the new contract is signed. Our theoretical model focuses on macroeconomic factors in explaining the delay. It emphasizes the importance of the realized nominal and real shocks, and of the levels of nominal and real uncertainty. The model is tested using Israeli collective wage agreements where long delays are frequent. The empirical findings strongly support the theoretical model. Thus, nominal uncertainty is found to increase the delay, and real uncertainty to decrease the delay, but less in the public than in the private sector.

There is renewed interest in ways to enhance secondary education, especially among disadvantaged students. This study evaluates the short-term effects of a remedial-education program that provided additional instruction to under-performing high-school students in Israel. The program targeted 10th ntwelfth graders who needed additional help to pass the matriculation exams. Using a comparison group of schools that enrolled in the program later and implementing a differences-in-differences estimation strategy, we found that the program raised the school mean matriculation rate by 3.3 percentage points. This gain reflects mainly an effect on targeted participants and the absence of externalities on their untreated peers. The program was found to be less cost-effective than two alternative interventions based on incentives for teachers and students.

Many people are fired from their jobs for poor performance. However, it is difficult to distinguish whether they are fired because they are not well suited for their job (sorting explanation) or because the firms are trying to provide incentives for effort (incentive explanation). This article develops a dynamic incentive model of dismissal and proposes a methodology to distinguish between these two explanations. The methodology is based on learning‐by‐doing and changes in the slope of dismissal probability with respect to tenure. Using personnel data from a large U.S. company, this study finds significant evidence for the incentive explanation.