The long debate over whether America has gotten more economically unequal in the last few decades is over; all but the most recalcitrant acknowledge it. (As a recent New York Times story reported, sharp-eyed salesmen have acted on this reality, increasingly marketing to the top few percent.) The economic argument has now shifted to whether average Americans have nonetheless done alright even as the rich have become super-rich. Here one detects a subtle difference in vocabulary. Defenders of the broadening inequality insist that average family incomes have been nonetheless increasing. They have. Critics of the broadening inequality insist that earnings have been flat or dropping. They have—for men.

Average American families have been holding steady or getting a bit more affluent because wives have been working more hours to make up for the stagnation in men’s earnings. (I discuss the numbers below.) Now comes a new study that shows the gender gap in another dimension: job tenure.

For years, experts have debated whether American workers were losing their job security. (The usual measure of job security is how long workers stay in a specific job.) Despite widespread concerns, there does not seem to have been a general reduction in job tenure. Now, Matissa N. Hollister and Kristin E. Smith show, in the latest American Sociological Review, that women’s increasing ties to their jobs have masked men’s decreasing ties to theirs—male job insecurity.


The gender explanation for the right vs. left disagreement about income vs. earnings is clear. Between 1987, the earliest year for the data series we need, and 2007 (both non-recession years prior to the 2008 collapse), the median income of American households did indeed grow, in inflation-adjusted dollars, by 10.4% (source here). Over two decades, that is anemic but still growing. However, for single-earner families, largely male-headed, median income grew only 8.0% between 1987 and 2007, while for two-earner families, it grew 22.1 percent. That is, average families prospered, albeit modestly, on wives’ labor.

Another take is to compare the earnings of men and women who were employed full-time. Between the same two years, 1987 and 2007, men’s median weekly earnings, adjusted for inflation, dropped 3%, while women’s increased 11% (Current Population Survey, series IDs LEU0252882800 and LEU0252881900). If one widens the window, comparing 1980 and 2007, the gender difference in the earnings trend is yet greater (a drop of 2 percent for men and a growth of 23 percent for women).


Hollister and Smith look at how long 30- to 50-year-old respondents to the Current Population Survey had been in their current jobs at the time of the interview. Around 1980, the average tenure was about 7.5 years and around 2010 it was about the same. However: For men, the number of years at the current job dropped across those three decades from an average of almost 9 years to about 7.5 years; for women, the average grew from under 6 years to about 7 years. Looking yet more closely, the authors identify the key group: married mothers. In 1980, they had been at their jobs an average of 5 years; in 2010, an average of over 6.5 years. Married moms increasingly stayed on the job.

Focusing still more, Hollister and Smith find that the growth in moms’ tenure largely took place before the mid-1990s. (It does appear that the surge of mothers into the labor force, what I called in an earlier post the “big change” of the last half-century, peaked about then—see also this column.) The authors show that the economy’s shift toward less stable kinds of jobs and industries explains part of what happened, although men’s time on the job seemed to keep dropping even accounting for those developments. Hollister and Smith concentrate on the years before the Great Recession, but they note that the recession seems to have increased job tenure, especially for women. They report “that wives of husbands who stopped working during the recession increased their labor supply, which may be a contributing factor in women” staying with the same employer. Still, over the last generation, wives have found stronger footing in the workplace and many husbands have lost theirs.

(Could it be that the women who entered the labor force pushed out the men? Though perhaps a contributor, the big take-off in wives joining the labor force, 1950 to the 1970s, coincided with good employment times for husbands as well. Something else accounts for the men’s fate since the 1970s.)


Hollister and Smith tell much of their story in terms of the feminist shift that led women to be more career-oriented and thus stay on the job longer. That is certainly a major part of the 1980-2010 story. But many mothers who go to work or who work longer hours or who get back to work as soon as possible after a birth are doing so because their husbands (or unmarried partners) are scuffling. Unlike apologists for the post-1960s economic policies would have it, growing inequality is not raising all boats. Average American families moved up a bit, slowly, in large measure because wives worked more than before and, for many, more than they wished.