Monday, May 17, 2010

Simple pictures against bad ideas

The Green Party is doing exceptionally well in Sweden right now. Educated voters, especially women, like their mix of environmentalism, social liberalism and perceived economic centrism.

Unfortunately and despite their rhetoric, the Green Party has a lot of bad economic ideas. One in particular is work sharing, a government regulation that forces everyone to work as standard no more than 35-hours per week. Their idea is that if you force people to work fewer hours, there will be more job for others.

The consensus belief among academic economists is that work sharing does not work.
Unemployment does not arise because there are too many people. It is because there is some imperfection in the market (either policy induces or due to market failure) that causes the market to generally not be able to match jobs to people.

We have to remember that normally in functioning economies, there are very strong forces that create jobs for everyone who wants to work. To illustrate this for non-economists, please allow me to put up a graph with a high "duh" factor (but which really is quite important).

This is the relationship between number of working age adults in 2007 and number of jobs in 2007, for the OECD countries. Source is as usual OECD.

The correlation between potential workers and jobs in the OECD is 0.99!

I have also done the same plot without the U.S and Japan so you can see the individual countries better.

To an economist this is trivial, and just says that there is no connection between employment rate and country size among the OECD countries. But savor the pictures for a moment. They have a profound implication. It means that there are extremely powerful forces in market economies that create jobs for ordinary people, no matter how many people we have, and regardless of if we can perfectly understand these forces.

It is not easy to describe this magic when people demand "where will jobs come from?". You may even sound naive if you say that "the market will take care of it", and refer to history or to the graph above. But in this case what sounds naive is in fact the most profound answer. Empirically, we can observe that the market does seem to take care of creating jobs.

The problems that cause unemployment is never the number of people, it is things like the skill composition combined with wage rigidity, cyclical demand conditions, search friction, taxes and regulations, and market imperfections. None of the core economic forces that create unemployment is affected by permanent work sharing for all workers.

Let me also look at this a little more directly. Here is average hours worked for workers and the unemployment rate, again for OECD, and again for 2007.

There is no statistically significant relationship between the typical workday and unemployment rate (p value 0.52). Countries that have reduces the average hours worked have not been able to achieve lower unemployment rate. Now, correlation is not always causation. Maybe the unemployment rate in France would have been even higher if they worked more hours. But I strongly doubt it.

Work sharing is guaranteed to harm the economy, by making everyone earn less and by dramatically lowering tax revenue for health care etc. Meanwhile there is no evidence that it reduces the unemployment rate, and strong suggestive evidence that indicates that it has no effect.

If people choice to work less, great! But legislation to shorten the workweek like the Greens in Sweden propose to do is very bad economic policy.


  1. in the last years we can't calculate the curve between the numbers if jobs and the people working in this moment, is like the people who work in made it and the people who Buy Viagra in the world.

  2. What a brilliant article, extremely informative and well written! Thanks for sharing.
    Funny Clips


Google Analytics Alternative