Workdays, Workhours, and Work Schedules.

• Author: Hunt, Jennifer

By Daniel S. Hamermesh. Kalamazoo, MI: W. E. Upjohn Institute for Employment Research, 1996. Pp. x, 155. $24.00.

Almost all research for which hours of work are relevant use information on weekly hours or perhaps annual hours. In this monograph, Daniel Hamermesh reveals how much can be learned by looking at finer decompositions of hours, in particular hours of work per day and days of work per week, as well as the exact times of day that work is done. His results are fascinating simply from the point of view of positive economics, but they also have some interesting normative implications for policymakers. A great strength of this book is that results from both the United States and western Germany are presented, enabling readers to judge which results arc idiosyncratic to a particular country and which results might indicate something more general about working in rich economies. Most of the analysis for the United States uses the 1991 Multiple Job Holding Supplement to the May CPS, although some panel analysis is performed on the linked 1977-1978 surveys. The German analysis is based on the 1990 German SocioEconomic Panel (GSOEP), which asked detailed questions about work time, and the panel analysis of 1990 and 1992 is used.

There are four central chapters to the book. The first of these focuses on characterizing the split of weekly hours into hours per day and days per week, using cross-tabulations as well as regressions to see the correlates of the two components, and how the two components fit together. The second chapter analyzes when in the day people work, and how this is affected by the presence of children and the behavior of the spouse. The analysis again supplements tabulations with regression. Longitudinal data and the link between job changes and schedule changes are the emphasis of the third chapter. The fourth component of the analysis highlights labor demand issues by combining the data with information on output and estimating production functions for a panel of industries with hours per day and days per week as inputs.

The book is worth reading just for the tabulations of the data. For example, how many readers would already know that 5% of Americans are at work at 3 A.M. on a typical work day? Another unexpected result is that among workers with atypical schedules, having long hours per day does not imply having many days per week, and having short hours does not imply few days.

The more analytical sections also reveal...