An article in WaPo in the January 28 print edition made the case for decreasing the workload of Americans based on the fourfold increase in the productivity of US workers since the 1950s. I liked the article and didn't disagree with much of it, which is a pretty rare experience for me in the last decade when reading WaPo articles on economic policy. However, there was this little gem appearing in it, that I think is worth at least a small comment, in passing.
”Companies argue that grueling work schedules are necessary to boost productivity.”
Richard Schiffman, the author of the piece didn't defend this view, but used it as a jumping off place for part of his argument that productivity gains have been great enough to justify reduced work weeks for Americans. But in doing this he failed to mention how patently silly the argument is.
The primary measure of productivity used by the OECD today is: the ratio of Gross Domestic Product to hours worked which stands in for the more basic notion output per labor hour input into the production process. Anyway, if people work longer hours and have “grueling schedules,” then output per hour will not increase. As a matter of fact, the opposite will be the case, since much empirical evidence gathered over the years, which I won't bother linking to here, shows that if one controls for the productivity contribution from capital, the productivity contribution from labor decreases as the number of hours worked per week increases beyond 30 hours.
Also, if one looks at the OECD productivity measures for 2007, the most recent comparative figures, one finds that the US is behind, France, Netherlands, Belgium, Ireland, Norway, and Luxembourg. Luxembourg has the highest productivity rating, about 37% higher than the US. But when we look at average hours worked per year among those nations we find that every one of them has an average work year in 2007 substantially lower than the US's which was 1798, with the Netherlands at 1388, Norway at 1419, and Luxembourg at 1515.
So, to summarize, longer hours of work and productivity don't go together, whether one looks at this theoretically, or from the viewpoint of data. I think Schiffman should have pointed that out in his very good article.