This past weekend I attended the rethinking economics conference in New York City. On Friday Sept. 12th Paul Krugman was part of a panel. I have rarely, if ever, agreed with Paul Krugman about anything but he did occasionally defend neo-classical economics against some of the more radical views and I found myself nodding in agreement. But alas the point of this post is to highlight another disagreement.
During one of his comments Mr. Krugman mentioned that French workers were just as productive as U.S. workers and used this statement as a reason for why people in the U.S. could work less without affecting our quality of life. I am not sure what he meant by this since as far as I know there is no law anywhere that mandates how many hours people have to work and thus if people want to work less they are free to do so. But aside from Krugman being Krugman and using a non sequitur to make a point, I was curious as to what he meant when he said that French workers are just as productive as U.S. workers. So I did a little research and came across some articles talking about GDP per hour worked rather than GDP per capita.
GDP per capita in the U.S. was $53,143 in 2013; in France it was $34,140 in 2013. So clearly American workers are more productive when using this measure. But workers also work a lot less in France than in the U.S. According to this article in Fortune, GDP per hour worked was $60 in the U.S. in 2011 and $57 in France. There are other articles that make a similar point. So once hours are taken into account there is not much of a difference between American and French worker productivity. At least that is what it looks like.
What the per hour productivity measures do not take into account as far as I can tell is unemployment and this is a problem. Assuming that unemployed workers are the least productive workers (which is why they are unemployed) a higher unemployment rate will increase the average productivity number because the least productive workers will not be bringing the average down.
For example, suppose both the U.S. and France have 5 workers producing $10, $20, $30, $40, and $50 worth of output per hour respectively. The average for each country is $150/5 = $30 of GDP/worker/hour. Now suppose the least productive workers become unemployed. Suppose 20% of the U.S. workers lose their job (which is 1 worker in this small example) and 40% of the French workers lose their job (which is 2 workers). The U.S. average is now $140/4 = $35 since the $10 worker lost their job and the French average is $120/3 = $40 since both the $10 and the $20 worker lost their job. Thus a higher unemployment rate biases the GDP/worker/hour number up by removing the least productive workers from the labor market and thus the official statistics.
So what were the unemployment rates in France and the U.S.? The overall unemployment rate in Dec. 2011 was 8.5% in the U.S. and 9.4% in France. The youth unemployment rate for France was nearly 23% in Dec. 2011 and was roughly 16.5% in the U.S. The youth unemployment gap is the largest and perhaps the most important since youth are usually relatively unskilled and in low output jobs. Eliminating a relatively large amount of youth from the official statistics is going to make the average larger.
It is always a good idea to take any statistic with a grain of salt before knowing how it was calculated. It might be the case that an official unemployment adjusted statistic of worker output per hour might not affect the average number too much, but the gap would certainly be larger. This makes the number cited in that Fortune article and likely the one Paul Krugman was alluding to essentially meaningless.