A Labor Problem ? Wachter

This paper addresses two related questions. First, what effect have demographic shifts and cyclical fluctuations had on the productivity slowdown?~ Second, what labor market policies ameliorate the productivity problem? The answers to both questions are shown to depend critically on how one measures productivity. We conclude that most labor market policies are unlikely to increase the best productivity growth rate measures without creating other major problems.2 In the first section of the paper, several approaches to measuring productivity are described. The standard approach is to use the average product of labor as measured by the Bureau of Labor Statistics (BLS). An obvious adjustment is to weight the manhours of different demographic groups differently so as to compensate for the changing composition of the labor force. A second approach is to use an estimated production function to calculate marginal products instead of relying on average products. It is also possible to break labor into submeasures corresponding to skilled and unskilled labor so as to obtain two or more average (or marginal) product of labor measures. Other measures discussed include cyclically adjusted labor productivity series, total productivity seri~s, and the trends in Hicks neutral technological progress. In Section II, measures of the productivity of the labor aggregate are examined. Although all the labor productivity measures show a slowdown has occurred over the past 15 years, the magnitude and the timing of the slowdown vary across the measures. In Section III, separate measures for youths and adults are examined. The pattern of decline differs dramatically between older and younger