Program meeting on labor studies.

Members of the NBER's Program on Labor Studies met in Cambridge on April 18 to discuss recent research. Organizers Richard B. Freeman and Lawrence F. Katz, both of the NBER and Harvard University, selected the following papers for discussion:

Steven D. Levitt, NBER and Harvard University, "Juvenile Crime and Punishment"

George Baker, NBER and Harvard Business School; Robert Gibbons, NBER and Cornell University; and Kevin J. Murphy, University of Southern California, "Implicit Contracts and the Theory of the Firm"

William T. Dickens, The Brookings Institution, "Neighborhoods and Networks"

Michael Kremer, NBER and MIT, "Should Taxes be Independent of Age?"

Alan B. Krueger, NBER and Princeton University, "Experimental Estimates of Education Production Functions"

Over the last two decades the punitiveness of the juvenile justice system has declined substantially relative to that of the adult courts. During that same time period, juvenile violent crime rates have grown almost twice as quickly as adult crime rates. Levitt finds that juvenile offenders are at least as responsive to criminal sanctions as adults are. Moreover, sharp changes in criminal involvement with the transition from the juvenile to the adult court suggest that deterrence plays an important role. Changes in relative punishment account for 60 percent of the differential growth rates in juvenile and adult violent crime between 1978 and 1993. There does not appear to be a strong relationship between the punitiveness of the juvenile justice system and the extent of criminal involvement later in life.

Baker, Gibbons, and Murphy analyze the role of "implicit contracts" (that is, informal agreements supported by reputation rather than law) both within firms, for example in employment relationships, and between them, for example as handin-glove supplier relationships. They find that the optimal organizational form is determined largely by what implicit contracts it facilitates. Among other things, they also show that vertical integration is an efficient response to widely varying supply prices. Finally, their model suggests why "management" (that is, the development and implementation of unwritten rules and codes of conduct) is essential in organizations.

Dickens presents a model of a labor market in which workers come from either a high rent or a low rent neighborhood. Employed people prefer to live in the high rent neighborhood, and can help unemployed people in their "network" (neighborhood) to find jobs. He shows that complete segregation of workers by employment status lowers the level of employment: beyond some point, making it easier for employed people to leave the low rent neighborhood begins to reduce employment. The level of mobility at which this change takes place is higher for populations with lower unemployment rates. Thus decreased racial housing segregation that has allowed affluent blacks to leave depressed inner city neighborhoods may have hurt black employment levels overall. And, because increasing mobility is detrimental at lower levels for populations that have more of a tendency to be unemployed, the same level of mobility that is helpful to whites could be harmful to blacks who face discrimination in finding a job and who on average are not as well prepared for work as whites.

Raising marginal tax rates at a given level of income reduces work incentives for people with that amount of income and increases tax revenues collected from people with higher incomes. Kremer finds that the number of people for whom work incentives are distorted per dollar of revenue raised in this way is typically more than five times greater for 17-to-21 year-olds than for 31-to-64 year-olds, and typically more than twice as large for 22-to-26 year-olds as for 31-to-64 year-olds. Reductions in marginal tax rates for youth, coupled with revenue-neutral increases in marginal tax rates for prime-age workers, will equalize the distribution of lifetime income, Kremer concludes, since income when one is young is almost uncorrelated with lifetime income.

Krueger analyzes data from Project STAR, an experiment in which 11,600 Tennessee kindergarten students and teachers were assigned randomly to one of three types of classes beginning in the 1985-6 school year: small (13-17 students); regular-size (22-25 students); and regular-size with a teacher's aide. Students in regular-size classes were re-assigned randomly at the end of kindergarten, and about 10 percent of students moved between class types in second and third grade. Attrition was also common. Krueger finds that: 1) on average, performance on standardized tests increases by about 4 percentile points the first year that students are assigned to a small class, irrespective of the grade in which the student first attends a small class; 2) after initial assignment to a small class, student performance increases by about one percentile per year relative to those in regular-size classes; 3) teacher aides have little effect on student achievement; 4) class size has a larger effect on test scores for minority students and those who receive free lunches.