"Average Versus Typical"
Michael Kinsley’s Politico column debuted yesterday with an argument about why I’m wrong and the rigors of “intellectual honesty” demand so much more than accuracy. And I agree with many of his points, there’s more to life than accuracy. But accuracy isn’t really as low a bar to clear as people sometimes seem to think. For example, Kinsley writes:
Here is another justification for taxing the money people leave behind when they die. According to a survey from the Federal Reserve Board, the average American household aged 65 to 74 has assets worth more than $1 million. Typically these amounts get spent down as people get older and sicker, so let’s say the second member of the typical couple dies leaving $500,000. That is far below the threshold for the estate tax. But for years, this couple has been collecting benefits from Social Security and Medicare. These are supposed to be insurance programs. Social Security protects you against the risk of being old and poor. Medicare protects you against the risk of being old and sick. Medicare operates like typical insurance: it pays to cover the costs of medical care and it pays out only if you actually are sick and suffer these costs. Social Security is different: it pays whether or not you’re actually poor.
Brad DeLong observes that the statistically “average” family is not necessarily typical and reports that the 2007 Fed Survey of Consumer Finances (PDF) says the median household in that age range had just $239,000 in assets even though the average was slightly above $1 million. The typical family is going to have less today than they had in 2007, so Kinsley is off here by a factor of four.
(I also don’t think this is a very good description of how Social Security works. If Social Security actually insured you against the risk of being poor, the senior poverty rate would be 0% rather than 13%).