Traditional human capital models suggest that age, race, and education affect individual labor market opportunities and levels of participation in crime. A unique data set containing 17 years of criminal activity and 9 years of earnings records for 1,176 men released from Georgia prisons was used to examine which demographic and criminal history characteristics affect labor market patterns among prison releasees. The examination yielded 3 main results. First, labor market participation rates and earnings levels were extremely low. Second, traditional findings on the effects of race and education on earnings levels were supported for this criminally active subset of the population. However, race and education did not affect employment rates. Employment rates of older individuals were especially low, even after controlling for characteristics expected to affect labor market attachment. Third, the amount of time incarcerated did not significantly affect earnings in the community, conditional on any employment.

Dramatically increasing incarceration rates in the United States have led to large concentrations of formerly imprisoned people in poverty-stricken urban areas. Therefore, identifying ways to help inmates who exhibit multiple, serious problems and who are at great risk of experiencing poor postrelease outcomes is especially important to urban communities, as well as to service providers and policymakers concerned about these communities. Our research provides evidence about the effectiveness of one strategy, called Health Link, which recruited adult women and adolescent men while they were incarcerated in a New York City jail and offered case management services during the especially challenging first year after release. About 1,400 participants who enrolled during a 3-year period were randomly assigned either to a group that was eligible for intensive discharge planning services and community-based case management services or to a group eligible for less-intensive discharge planning and no community-based services. We investigated whether the availability of these services reduced rates of drug use, HIV risk, and rearrest. Using data from interviews and hair analysis to measure impacts during a 1-year follow-up period after clients' release from jail, we detected increased participation in drug treatment programs and weak evidence for reduced drug use. However, we did not observe reductions in rearrest rates or in activities with high risk of HIV infection. We conclude that a well-executed case management program can make modest differences in a few short-term outcomes of former inmates. However, the intervention did not lead to the hoped-for changes across a range of outcomes that would clearly indicate greater success in community reintegration or improved health.

The pay of college graduates, of professionals and managers, and of other white-collar workers increased relative to the pay of less-educated and blue-collar workers; joblessness increased among the less-educated but not among college graduates. Dispersion of earnings within educational groups increased. The rise in earnings and employment differentials was greatest among younger men: from the early 1970s through the 1980s the real earnings of 25-34-year-old men with high school or less education fell by some 20 percent. Their employment-population rate dropped by over 10 percentage points, while college graduates suffered no such losses. What happened to earnings and employment differentials between moreand less-educated workers in Canada in this era of rising economic inequality in the United States? Did supply and demand for labor shift in the same way in Canada as in the United States? Did Canadian wage-setting institutions respond "more gently" to the market twist against the less skilled than those of the United States? What does the Canadian experience tell us about the causes of the 1980s rise in skill differentials in the United States? To answer these questions we analyze data from the Canadian Survey of

Ever since the U.S. federal-state system of unemployment insurance was founded in the 1930s, it has provided partial, temporary replacement of wages to eligible workers who lose jobs -- through no fault of their own -- (as determined by state-level regulations). Unemployment insurance is one of the largest social insurance programs in the United States, with benefits paid totaling about $34 billion in 2004. Economic theory can help us understand the challenges this complex program is likely to face over the next few years. We begin by summarizing the salient characteristics of the unemployment insurance program and then examine the theoretical and econometric research. Much of this research revolves around the main goals of the program, which include: 1) sustaining consumption for workers and their families; 2) helping recipients to make efficient job choices during a period of financial stress; and 3) minimizing the adverse incentives that may accompany partial wage replacement. Of course, these goals can come into conflict -- for example, if replacing wages for an unemployed worker also discourages that worker from aggressively searching for or accepting a new job -- and our discussion will focus on these conflicts. In conclusion, we address the key policy issues that the unemployment insurance system is likely to face in upcoming years and ways policymakers may be able to use economic analysis to adjust the program so that it remains effective in addressing the needs of unemployed workers.

The extreme labor market weakness during and after the Great Recession led to the passage of extensive federal legislation related to unemployment compensation. In this paper, we summarize that legislation and some of the research related to it. Although our particular focus is on the very long potential durations of benefits (up to 99 weeks) initially implemented in 2008-2009, we also look at a variety of other initiatives. Most of the research we review comes from the United States. But we also provide a brief look at the vast European literature that addresses many of the same policy issues.