EMPIRICAL STUDIES OF LABOR TURNOVER play an important role in improving our understanding of the labor market. For example, theories of frictional unemployment increase in significance if total turnover is found to be large. If a dominant form of turnover is temporary separations without a permanent job change, theories of temporary layoff unemployment (based on a view of the labor market in which firms and workers form long-term attachments) gain importance. Similarly, the problems associated with structural unemployment are most likely to be of concern if permanent separations due to plant closings or cutbacks make up a large part of turnover. Additionally, separations are likely to result in larger earnings losses if a high-quality job match is destroyed,o r if the worker had accumulated firm-specific human capital. Because such losses are likely to be high, and because firms also incur losses in the form of hiring and training costs when turnover occurs, both parties have an incentive to reduce turnover in these cases. Thus turnover patterns can be informative on the nature of the matching of workers to jobs and on the accumulation of firm-specific human capital. Despite the importance of turnover, though, our knowledge of it is surprisingly slight, and much of what is known comes only from the manufacturing sector.