I’m delighted to be having this conversation with Scott, who is second to none in the thought and nuance he brings to discussions of America’s economic conditions. The good news is that we generally agree on the data itself (as a rule of thumb, I go with whatever Scott says!). But we diverge sharply on how to make sense of it and what it is telling us. My view, as I describe in The Once and Future Worker, is that American workers — specifically, those without a college degree — are facing a deepening crisis, for which primary responsibility lies with a couple generations’ worth of public policy that disregarded labor-market health. I see that segment of the labor market as fundamentally broken, and believe that repairing it should be our focus.
Let’s start from the wage data that Scott analyzes. The analysis reminds me of a skillful cross-examination by a defense attorney. A scrupulously honest attorney, to be clear. But it’s as if current economic conditions are his client and the only task is to bring forward every exculpatory line of argument and piece of evidence. So while the standard Bureau of Labor Statistics data look alarming (e.g., median earnings of a high-school graduate down 15 percent between 1979 and 2017), Scott introduces a long series of adjustments: a different price deflator, 20th and 50th percentile instead of high-school graduate, add in benefits, disaggregate by race to account for immigration, and so on. Each of these is a valid critique and a worthwhile lens to look through. But there are plenty of critiques and adjustments one might propose in the opposite direction: rising commute times, higher income volatility, artificial increases at each percentile caused by labor-force dropout at the bottom, the low marginal value of expensive health benefits, higher payroll taxes, and so forth.
To paint the most accurate picture of economic conditions, rather than the one friendliest to the status quo, our efforts at recalibration have to be balanced. (It would be rather strange if the standard government measure of wage trends were the most pessimistic one possible.) Of course, it’s always possible that some enormous overlooked factor is determinative. But that’s not the case here. Scott’s adjustments, all in one direction, still bring us to just a 10 percent compensation increase across 40 years for the median male, during a period in which GDP per capita more than doubled. We should also look beyond aggregate data and recognize that the picture may be much worse for certain places and groups. Take young men — of particular relevance for family formation and social stability. The Census Bureau reports that “in 1975, only 25 percent of men, aged 25 to 34, had incomes of less than $30,000 per year [in 2015 dollars]. By 2016, that share rose to 41 percent of young men.”