Tyler Cowen, of whom I’m generally a big fan, summarizes an interesting post by Michael Mandel on recent productivity growth (the lack thereof). But he ends by trumpeting Hamilton Project analyses claiming to show that men’s earnings declined by 28 percent between 1969 and 2009. This claim, like the Mandel analyses, reinforces Cowen’s argument that we are in a Great Stagnation, but it’s not true! Stop this meme!
I’ve not had much time to blog recently, so I submitted a brief critique in the comments to the Leonhardt post that introduced the world to this unfortunate study (co-authored, unfortunately, by a fellow classmate of mine from Harvard’s inequality program) and in the comments to the Hamilton post. Here’s the basic problem: the analyses assign all nonworking men annual earnings of $0, and since labor force participation among men has declined, the result is a big drop in median earnings over time. But a lot of that decline in labor force participation is attributable to earlier retirement (they include men as old as 64), later and longer school enrollment (they include men as young as 25), rising “disability” rates (which do not correspond in any obvious way with changes in health or job demands but which do correspond with increasing generosity in disability benefits), and other factors having nothing to do with the strength of labor markets.
I re-crunched the numbers as follows. I included all men age 20 to 59 except for those who said they worked only part of the year or not at all because they were retired, going to school, in the Armed Forces, sick or disabled, or taking care of home and family. Using the inflation adjustment that the Hamilton guys likely used, I find a decline in median earnings of 9 percent, not 28.
Note, however, that comparing 1969 and 2009 holds up a likely peak year (when the business cycle was at a high) to a trough year (when it was at a low). Comparing 1969 to 2007 is apples-to-apples, and when I did that, the median was EXACTLY the same in both years (to the dollar, which is a pretty crazy coincidence). Finally, if I use the Bureau of Economic Analysis “personal consumption expenditures” deflator, which I think overstates inflation somewhat less than other commonly-used deflators, median earnings among men rose 7 percent from 1969 to 2007.
Seven percent is no great shakes, but this figure is also too small for assessing how men’s economic fortunes have changed over time. None of these analyses account for the fact that as a group, husbands reduced their hours over time in response to rising work and wages among wives. Nor do they account for the rising share of non-wage benefits in total compensation (health and retirement benefits have eaten into wages, presumably following the preferences of the median worker). Nor do they include the impact of taxes (which have declined) and tax credits (which have increased). In addition, even my figures may overstate inflation, thereby understating the earnings increase over time–inflation measurement is much more tricky when choices within categories of goods and services and retail outlets explodes and when so much of what we consume is (thanks to the inter-web . Finally, the analyses do not account for changes in the composition of the population. For instance, the fact that more men today are nonwhite and foreign-born pushes the 2009 median down, but it is likely that the typical white, nonwhite, native-born, and foreign-born men are all doing better than the trend in the overall median implies. Someday I’ll get to a full analysis.
Subject for discussion (and a future post): how are we as a nation supposed to clearly understand the state of the economy and our living standards when even moderate think tanks and researchers are so eager to hype negativity? As I’ve said before, policymakers aren’t the only people who–individually or collectively–can talk down the economy.
Crossposted at The Empiricist Strikes Back