So there you are – the Thatcher revolution: Longer hours, lower productivity
They reach this conclusion by looking at average working hours for full time workers, then comparing that with productivity per hour of people in work.
Ah, no, you see, you’re not supposed to do it like that.
You can look at productivity at the economy level, what’s the productivity of all people who could be in work? But if you look at the productivity only of those who are actually in work then you’ve got to look at the unemployed as well. Those who would/could be in work but aren’t.
For it doesn’t take a planet sized genius to note that those most likely to be out of work are the least productive labour. That is, after all, what labour is hired for, productivity.
So, for example, if we wanted to measure UK productivity against French productivity we might want to adjust for different levels of unemployment. The country with the higher levels of unemployment we would naturally assume will have higher productivity per hour worked, given that they’ve some chunk of their least productive people doing nothing instead of something.
There’s another glaring error in their numbers as well. They only look at hours in market work. Again, sorry, but you shouldn’t do that. You should look at total hours of work, household production as well as market.
For after all, we’re actually interested in the residual, what is available leisure after all working hours and all personal time? As, in fact, Mssrs. Sen and Stiglitz pointed out in that report they did for Sarkozy.
When we do this results change, really quite dramatically: for example, the average American woman, what with her higher market working hours, actually has more leisure (by half an hour a week according to one paper drawing off the LIS figures) than the average German woman with her much lower market hours.
Still, there you go, Will Straw’s evidence that he’s actually done something in the real world, cashed union cheques to employ an intern, showing again that there are lies, damned lies and political statistics.