Cyclicality of Productivity, Labor Hoarding, and the Marginal Efficiency of Capital

There has been a lot of talk in the blogosphere about the cyclicality of productivity in modern recessions. It used to be that labor productivity is procyclical. Now it appears to be counter-cyclical. In no particular order, here are thoughts from Tyler Cowen, Paul Krugman, Alex Tabarrok, Stephen Williamson, and Brad DeLong. The Economist offers a link farm for other discussion of the issue.

With all the work on doctoral applications and a few reports to finish for work, I haven't gotten the chance to look at this or think about it in any great detail, but here are a few thoughts:

1. First, I think we have a (correct) tendency to look at the labor productivity numbers as effect and not cause. American workers probably aren't different now - firms just react differently to them. This is why productivity gets connected with talk of labor hoarding. Several decades ago, an unproductive worker would be kept on as a human capital investment, lowering average productivity. Today unproductive workers are dropped more quickly, raising average productivity. This is probably the right way to look at it, but we also shouldn't discount the prospect of some technological development that is causing some technological unemployment. These "modern recessions" did emerge with the spread of computers, after all. This is the Arnold Kling take on things. I doubt that explains all of it, but it's always a possibility.

2. We also need to remember that productivity figures are measures of current productivity. If you think of human labor as a raw material, it makes sense to talk about that. But if you think of human labor as an investment, it doesn't. What firms care about when they think of human labor as an investment is the discounted value of the stream of income that can be derived from the employment relationship. This raises another explanation - maybe labor hoarding behavior has changed not because the current productivity of an individual worker has behaved all that differently. Maybe it has changed because the expected discounted value of the stream of income from the employment relationship has been reduced.

3. Tyler Cowen talks a lot about the zero marginal product worker... maybe... but I wouldn't fixate on the "zero". What we may be dealing with here is a change in the marginal efficiency of human capital, which is lower than the real interest rate. Firms drop workers not because unions are weak or because their current productivity is strictly zero, but because it pays more to hold idle assets than it does to make a human capital investment. Wages are rents paid to a special kind of capital. My workload ebbs and flows, as does the workload for a lot of people in the modern economy. The paycheck I get when work is lighter is like rent paid on capital. I'm essentially on retainer. And when the workload gets heavy, I still get paid the same. I am an investment. I am not a raw material. And my employer treats me like an investment, rather than a raw material.

4. There was one guy that talked a lot about marginal efficiency of capital... I forget what his name was...



Maybe more on this later... I had to cut short - got an early conference call.