Jobs, Training, and Worker Well-being: Volume 30

Early models of the functional distribution of income assume constant labor productivity among all individuals. Not until human capital theory developed did scholars take into account how productivity varied across workers. According to early human capital models, this variation came about because each individual invested differently in education and training. Those acquiring greater amounts of schooling and on-the-job training earned more. However, these models neglected why one person would get training while another would not. One explanation is individual heterogeneity. Some individuals are smarter, some seek risk, some have time preferences for the future over the present, some simply are lucky by being in the right place at the right time, and some are motivated by the pay incentives of the jobs they are in. This volume contains 10 chapters, each dealing with an aspect of earnings. Of these, the first three deal directly with earnings distribution, the next four with job design and remuneration, the next two with discrimination, and the final chapter with wage rigidities in the labor market.

The aim of this chapter is to analyze the sources of earnings dispersion between trainees and nontrainees. We stress three mechanisms by which investment in general training may affect wage inequality: directly via participation to a general training program and indirectly via the selection process of trainees or the existence of heterogeneous returns on training. This chapter adopts an approach originally proposed by Fields (2003) but extends it to the breakdown of inequality by population subgroups – those who received training and those who did not. The empirical illustration is based on four French surveys, the 2006 Adult Educational Survey and the 2004, 2005, and 2006 Labor Force Surveys that complement it.

We use household panel data to explore the wage returns associated with training incidence and intensity (duration) for British employees. We find these returns differ depending on the nature of the training, who funds the training, the skill levels of the recipient (white- or blue-collar), the age of the employee and if the training is with the current employer or not. Using decomposition analysis, training is found to be positively associated with wage dispersion: a virtuous circle of wage gains and training exists in Britain but only for white-collar employees.

In the United States, there is little difference in annual income inequality and income mobility between the rural and urban sectors of the economy. This forms a sharp contrast with China where income inequality is greater and income mobility lower among rural households than among urban households. When incomes are averaged over three years and when adjustments are made for the size and composition of households, income inequality among all households differs little between China and the United States in the 1990s. Moreover when pooling rural households and urban households and when measuring annual income inequality and income mobility of the pooled households, the mobility of incomes of households in the United States differs little from that in China. Social welfare functions are posited that allow for a trade-off between increases in income and increases in income inequality. These suggest strong increases in well-being for urban households in China. The corresponding changes in rural China and in the United States are smaller. Four sets of data on households are drawn on to document these findings.

In this chapter we study job design. Do organizations plan precisely how the job is to be done ex ante, or ask workers to determine the process as they go? We first model this decision and predict complementarity among these following job attributes: multitasking, discretion, skills, and interdependence of tasks. We argue that characteristics of the firm and industry (e.g., product and technology, organizational change) can explain observed patterns and trends in job design. We then use novel data on these job attributes to examine these issues. As predicted, job designs tend to be “coherent” across these attributes within the same job. Job designs also tend to follow similar patterns across jobs in the same firm, and especially in the same establishment: when one job is optimized ex ante, others are more likely to be also. There is evidence that firms segregate different types of job designs across different establishments. At the industry level, both computer usage and R&D spending are related to job design decisions.

In this chapter we use data from industrial plants to find out whether seniority-based pay is used as a motivational device for production workers. Alternatively, seniority-based pay could simply be a wage-setting rule independent of incentives. Unlike previous papers, we use a direct measure of seniority-based pay as well as measures of monitoring devices and explicit incentives. We find that those firms that base their wages partly on seniority are less likely to offer explicit incentives. They are also less likely to invest in monitoring devices. We also discover that these companies are more likely to engage in other human resource management policies, which result in long employment relationships. Overall these results suggest that seniority-based pay is indeed used as a motivational device.

We formulate static and dynamic empirical models of promotion where the current promotion probability depends on the hierarchical level in the firm, individual human capital, unobserved individual specific attributes, time-varying firm-specific variables, as well as endogenous past promotion histories (in the dynamic version). Within the static versions, we investigate the relative influence of the key determinants of promotions and how these influences vary by hierarchical levels. In the dynamic version of the model, we examine the causal effect of past speed of promotion on promotion outcomes. The model is fit on an eight-year panel of 30,000 American executives employed in more than 300 different firms. The stochastic process generating promotions may be viewed as a series of promotion probabilities which become smaller as an individual moves up in the hierarchy and which are primarily explained by unobserved heterogeneity and promotion opportunities. Firm variables and observed human capital variables (age, tenure, and education) play a surprisingly small role. We also find that, conditional on unobservables, the promotion probability is only enhanced by the speed of promotion achieved in the past (a structural fast track effect) for a subset of the population and is negative for the majority. In general, the magnitude of the individual-specific effect of past speed of promotion is inversely related to schooling, tenure, and hierarchical level.

We discuss a class of copula-based ordered probit models with endogenous switching. Such models can be useful for the analysis of self-selection in subjective well-being equations in general, and job satisfaction in particular, where assignment of regressors may be endogenous rather than random, resulting from individual maximization of well-being. In an application to public and private sector job satisfaction, and using data on male workers from the German Socio-Economic Panel for 2004, and using two alternative copula functions for dependence, we find consistent evidence for endogenous sector selection.

We empirically study gender segregation in privately owned Swedish establishments, and the correlation between gender segregation, survival and growth of establishments. We find that the overall inter-establishment gender segregation in Sweden has been constant between 1987 and 1995 and at the same level as that found in US manufacturing. Our results show that establishments dominated by males or females have a higher probability of exiting the market than more integrated establishments and that establishments dominated by females grow more slowly than other establishments. An important additional finding is that establishments with a skewed workforce in terms of educational background have lower survival probabilities. Furthermore, establishments with skewed age distributions have both lower survival probabilities and grow less compared with other establishments. These findings are consistent with theories suggesting that workers with different demographic characteristics contribute to a creative working environment as a result of their different experiences, a greater variety of information sources and different ‘thinking’.

Using Becker's ‘taste for discrimination’ model, the chapter analyzes the current legislation against wage discrimination and finds it counterproductive. Using a costly apparatus of auditing, detecting and fining violators does not deliver results. If a fine is levied on discriminators and reimbursed to the disadvantaged workers in order to undo the discrimination, it affects equally the demand for and the supply of those workers, because their expected wage includes the fine, and has no real effect. If the fine is collected and kept by the government, it shifts employment away from the workers it seeks to help, to others, depressing the total employment. In contrast, levying a tax on the favored workers effectively curbs discrimination in the labor market. A quota is a possible substitute for a tax with questionable side effects. Affirmative action is in essence a sort of tax on employing favored workers, only administered in an indirect, clumsy and costly way. Yet, the chapter explains its humble impact in the right direction. An explicit and direct tax would do much more and with a negative cost. Alternatively, subsidizing the disfavored workers is a costly but as effective policy that, in addition, boosts total employment.

We study the distortions that downward nominal and real wage rigidity would induce to a flexible form of a notional, rigidity-free, distribution of wage change using the histogram-location approach. We examine alternative methods of generating the histograms that support the econometric search for rigidity distortions and implement our approach to inflation sub-periods that should be characterised by different patterns of nominal and real rigidities. We establish the general applicability of the approach to these sub-periods and find results consistent with expectations.