When the Census Bureau’s closely watched income, poverty and health insurance results for 2016 came out last week, those of us who track such numbers were glad to see the results. Poverty fell, middle-class incomes rose and the percentage of Americans without health coverage fell to its lowest level on record. As I stressed last week, these results are a function of an economic expansion that’s finally reaching more people, along with the coverage impact of the Affordable Care Act. The 2015 results were similarly positive, though unusually low inflation helped boost real incomes that year.
So, as David Brooks announced in the New York Times the other day, is the economy basically fine and all that stuff about stagnant middle-class earnings, poverty rates unresponsive to growth and the downsides of globalization just old news? Based on these results, Brooks calls for a “resurgence of the moderates,” suggesting we eschew the populist left and right with their (very different) ideas about how to achieve more favorable outcomes for the working class (or, on the far right, the white working class).
Not so fast on the rethink. Two years does not a new economic narrative make.
First, let’s put these recent gains in historical context. Look at the figure below showing the trajectory of the median income for working-age families — household head younger than 65 — over the past 30 years (the break in 2013 is due to a change in the survey that captured more income). The strong gains over the past few years took the median income of these working-age families back to where it was before the recession, which was about $2,300, or 3.4 percent, lower than it was in 2000.
Second, surprisingly, at first blush, the census data showed no real gains last year in the median earnings of full-time, full-year men or women. One reason for this is that as the job market improved last year, more low-earning part-timers moved into full-time work and that pulls down the median growth figures.
That said, there is a serious long-term stagnation problem in the earnings of middle-wage men, as shown in the next figure. Remarkably, the real median annual earnings of men — and remember, these are guys that are solidly attached to the job market — have been essentially flat since the early 1970s. Again, these are not the same guys over time, although measures that control for such changes reveal similar outcomes. There’s simply no escaping the fact that for decades now, the job market for many men simply isn’t presenting them with the opportunities they need to get ahead.
Yes, the loss of manufacturing employment, with its higher than average pay for blue-collar workers, is implicated, but that’s clearly not the whole story. Real hourly compensation — wages plus benefits — for blue-collar, manufacturing workers is up only 5 percent since 1979, about 0.1 percent per year. Here, too, along with manufacturing job losses, which accelerated post-2000, stagnant hourly earnings are also implicated.
The story for women is more complicated, as their annual earnings climbed pretty steadily up until about 15 years ago (see previous figure). Their progress, helped along by the increased demand for services and women’s education and occupational upgrading, clearly helped close part of the gender wage gap, which in these data went from 60 percent in 1960 to 80 percent last year. Still, in recent years, the median woman caught the median man’s wage disease, and the progress in gap closure had significantly slowed.
Finally, recent research tracking earnings across men and women’s prime working years — 25 to 55 — showed that even if you follow the same people, the figures end up similar to those above. Comparing earlier to later cohorts reveals that the lifetime earnings of the median man fell between 10 percent and 19 percent since the mid-1960s. In fact, the earnings of most men — the bottom 75 percent (ranked by earnings) — over their prime worker years were stagnant across cohorts that entered the job market over these years. Women’s cohorts, conversely, achieved large gains, between 22 and 33 at the median, but because their growth comes off such a low earnings base, their gains do not fully offset men’s losses.
These trends in these figures helped birth the populism upon the land today, and the past few years don’t change that (there’s also a telling urban/rural divide in the census data that feeds into today’s politics). Operationally, the populism is a demand for policies directly targeting those hurt by these long-term trends, in contrast to indirect, often-failed solutions that moderate, establishment politicians sold to working-class families, like trade deals, balanced budgets and financial deregulation. Better than anyone, except maybe Sen. Bernie Sanders (I-Vt.), Trump sniffed this out and effectively exploited it during his campaign.
That’s not to dismiss the lessons of the past few years, which provide an important reminder of the potent combination of markets, as low unemployment pushes back on income stagnation, and policy, as the ACA partially addresses the market’s failure to provide affordable health coverage. But if you think the people behind the figures above are assuaged by a couple of good years, you need to first stop kidding yourself and, second, dig a lot deeper into the structural changes driving these trends.