Knowledge that Transforms

This paper integrates the labor and assets markets equilibria to determine and evaluate the wage differentials generated for cyclical and noncyclical risks of unemployment. The relative wage differential is a linear function of unemployment risk measured by the covariance of an index of employment with the rate of change in aggregate output. Seniority and the hoarding of skilled labor are characteristics of minimum cost contracts because employees with more human capital prefer safer jobs. Empirical results suggest that a 14%-41% wage differential can be explained by interindustry differences in unemployment risks.

Among fraternal pairs from the Wisconsin Longitudinal Study, we model the effects of measured and unmeasured family background factors, mental ability, and schooling on occupational status and earnings. The models are estimated from incomplete data with corrections for measurement error, and they permit direct comparisons of within- and between-family regressions. We find no evidence that the effects of family background lead to a bias in the effect of mental ability on schooling or in the effects of schooling on occupational status or earnings. Family background does have large independent effects on ability, schooling, and, to a lesser degree, socioeconomic attainment.

There are jobs for which firms employ older workers but tend not to hire new older workers. This may be attributable in part to implicit contracts that discourage worker shirking and malfeasance by shifting compensation to the end of the contract. Such "delayed payment" contracts can introduce a form of fixed costs into the employment relationship. Much as with hiring and training costs, these fixed costs lead firms to hire primarily young (long-term) workers. While firms employ older workers-workers who are serving out the last years of their contract-they tend not to hire them. This paper presents empirical evidence that is consistent with this argument.

This paper analyzes trends in the skills of immigrants to the US in the post-World War II period. Changes in the supply, demand, and institutional factors determining immigration are analyzed for their implications for immigrant skills. During the past 4 decades immigration has shifted from being predominantly European and Canadian in origin to being predominantly Asian and Latin American, and there have been changes in the criteria for rationing immigration visas. Immigrant skills can be analyzed within the context of a model of the supply of immigrants and the US demand for immigrants. Of the Asian immigrants subject to numerical limitation, the proportion who were occupational preference principals declined from 18.2% in 1970, to 11.9% in 1975, to 8.1% in 1981. A growing stock of the foreign-born population who are illegal aliens may lower immigrant quality; for low-skilled workers in neighboring low-income countries the economic incentives for illegal migration are very large. Immigrants from the UK have the highest annual earnings, with Canadian, other European, South Asian, East Asian, and other American immigrants having successively lower earnings. The Mexicans and the Vietnamese have the lowest earnings. Over the period 1950 to 1980, US immigration changed from primarily drawing immigrants from countries whose nationals have high relative earnings in the US primarily drawing immigrants from countries whose nationals do less well. Recent immigrants are less favorably selected on the basis of their level of schooling. The analysis of the relative earnings of immigrants during the 1970s using 3 data files shows there has been little change for white immigrants, an ambiguous pattern for Mexican immgrants, perhaps a small decline for Cuban immigrants, and a small rise for Asian immigrants. Overall, without returning to rationing by country of origin, public policy could raise immigrant skill levels by changing the balance between kinship and the individual's skills in the rationing of visas.

"This paper uses combinations of full brothers, half brothers, and fathers and sons to measure the effect of common family background on a household's income and wealth. While the data are drawn from a nineteenth-century [U.S.] population, the intraclass correlation for income ranges from .13 to .18, which is similar to that found in modern samples. Intraclass correlations for wealth are significantly higher (.18-.35) than are those for income. Intraclass correlations of half brothers compared to those for full brothers suggest that fathers play a dominant role in the transmission of the common family effect. When unobserved background is decomposed into individual and family effects, the individual effect dominates the family effect for income, while the family effect dominates the individual effect for wealth." A comment by Sherwin Rosen is included (pp. 80-2).

Whereas previous analyses of criminal deterrence have focused on the effect of criminal enforcement on crime rates, this study analyzes the existence of compensating differentials for criminal pursuits. By analyzing the risk-rewards trade-off, this approach represents a more comprehensive test of the criminal deterrence hypothesis. The sample consisted of black inner-city youths who reported their crime participation, crime income, and self-assessed risks from crime. The risk premiums for the three principal adverse outcomes (arrest, conviction, and prison) constituted between one-half and two-thirds of all crime income on the average, providing strong support for the criminal deterrence hypothesis.

This article uses a unique data set to measure the economic and demographic consequences of mental illness. We find that mental illness significantly reduces an individual's earnings and that in some cases this effect lasts as long as 15 years. Mental illness is also shown to affect an individual's ability to marry or stay married, lowers the number of children that he has, and encourages his wife to work.

Female manufacturing workers around 1900 were far more likely to be paid by the piece and were rarely employed at the same occupation in the same firm as males. These and related aspects of work organization can be understood through a model in which workers shirk, monitoring is costly, and males and females have different turnover rates. Employers adopt either piece rates or deferred payment. Occupational segregation by sex and differences in earnings result even if workers are equally productive. Establishment-level data on supervising male and female workers in time- and piece-rate positions are examined.

"In this paper we question the pioneering work of Todaro, which states that rural-to-urban labor migration in less developed countries (LDCs) is an individual response to a higher urban expected income. We demonstrate that rural-to-urban labor migration is perfectly rational even if urban expected income is lower than rural income. We achieve this under a set of fairly stringent conditions: an individual decision-making entity, a one-period planning horizon, and global risk aversion. We obtain the result that a small chance of reaping a high reward is sufficient to trigger rural-to-urban labor migration."

"Birth-order effects are posited by many to affect earnings and schooling. We show how such effects can be interpreted to shift either the earnings possibility frontier for siblings or parental preferences. We find empirical evidence for birth-order effects on (age-adjusted) schooling and on earnings for young U.S. adults, though the latter is not robust for all specifications. The examination of intrahousehold allocations suggests that these birth-order differences occur despite parental preferences or prices by birth order favoring later borns, apparently because of stronger endowment effects that favor first borns."