There are a lot of headlines screaming about how the U.S. unemployment rate is exploding, and has now reached levels not seen since 1983. The February unemployment numbers for February really are quite bad; unemployment increased 0.5%, from 7.6% to 8.1%. May of last year was the only month in any of the recessions of the past 25 years that saw that big a jump in the rate.
When we think about how bad unemployment gets in any given recession, however, we need to consider three numbers: the starting point, the average monthly change in the rate, and the duration of the recession. The past two recessions (2001 and early 90s) were very mild, so I'll compare the current recession to the 81-82 recession, which was by most measures the worst since the Great Depression.
The average monthly increase in unemployment rate was .225 points in the prior recession, and in the current recession it has (so far) been .229 points, or almost exactly identical. The prior recession lasted 17 months, and this one has lasted 15 months so far. But unemployment peaked in the prior recession at 10.8%, which we are extremely unlikely to approach within the next couple of months when our recession reaches 17 months. Why? Because of where we started. The U.S. unemployment rate was 7.2% at the start of the prior recession, which is a rate we only reached last December, after more than a year of recession.
The structural work on the economy is at least as important as how we deal with the recession. Doing something, anything, to stop the pain of the current recession, no matter what its structural effects on the economy, might seem practical, but it is not. Bailing out failing companies, preventing write-down of assets like mortgages, borrowing against our future with stimulus spending and so on might feel good today - and some of this surely has to be done - but we should recognize that we will pay the price in higher structural unemployment every future month we are not in recession, and probably make it harder to react when new difficulties arise.