Comments Guidelines

All comments are pre-moderated. No spam, slurs, personal attacks, or foul language will be allowed.

Tuesday, January 20, 2015

What is Noah thinking?

Noah Smith put up a post Sunday purporting to show that things aren't so bad for the middle class. Then he immediately shows us a chart of median household income. Stop right there. As I have argued before, this is always going to give you a rosier picture than reality. We need to look at individual data, aggregated weekly (because average hours per week have fallen for non-supervisory workers), to know what's going on.

Because the individual real weekly wage is still below 1972 levels, households have had to compensate by having more incomes and going into debt. They have traded time and debt for current consumption. This is not an improvement in the middle class lifestyle. Commenter Richard Serlin points out that we also need to consider risk as well as average incomes, and he is right. The middle class is less secure than it was in 1972.

Noah has lots of interesting things to say, and you should check out his blog if you haven't already. But this is an error on his part, and I don't understand what he's thinking.


  1. "A rising tide lifts all boats. Too bad you don't own one."

  2. I disagree. The labor force participation rate is only 62.7% (both sexes, between the ages of 16 and 64). In 1972, that was about 60%.

    There just aren't that many more incomes per family than there were in 1972. Whatever rise there is in the labor force participation rate is offset by the increase is single-parent households. The median household size has dropped from 3.1 in 1970 to 2.6 in 2007. And that's not just fewer kids, but fewer married parents.

    If you tax something, you get less of it. If you subsidize it, you get more of it. Well, we subsidized single mothers, so now we have more of them. Brilliant.

    1. Your comment and Noah's analysis makes me wonder what, if any, is the right way to account for the decline in household size. Median household size for the country as a whole may be different from the average size of the median income household. Another reason why I think it's better to start with individual wages.

      Also, since there is often considered to be economies of scale in consumption, then perhaps the adjustment shouldn't be, as Noah does, simply 3.1/2.6. Here is an OECD document on various ways of calculating "household equivalencies": The takeaway is if we buy this analysis, the adjustment also becomes smaller.

    2. I believe one common way is to use the square root of the household size.