Another Reason to Think Hard About 1937-1938
Derek Thompson:
The Scariest Unemployment Graph I've Seen Yet: The median duration of unemployment is higher today than any time in the last 50 years. That's an understatement. It is more than twice as high today than any time in the last 50 years. OK, you're saying, but what does this mean? Does it mean we must increase the duration of unemployment benefits to protect this new class of unemployed, or does it mean we need to stop subsidizing joblessness? Does it mean we need to expand federal retraining programs, or does it mean federal retraining programs aren't working? Does it mean we need more stimulus, more state aid, more infrastructure projects, more public works ... or does it mean it's time to stop everything, stand back and let business be business?
You're going to find smart people make a case for all six of the above public policy directions. (I tend to side with the first of each coupling.) It's hard to know for sure how to design public policy for historically unique crises precisely because they are historical orphans, without precedent to show us the right way from the wrong.
Actually we do have a precedent. So far the rule "don't do what people did in the decade after 1929" has served us rather well. Which is why those arguing for a withdrawal of stimulus should think hard about 1937 and 1938
