Median family income since 1989: Is the stagnation real?

The Census Says Median Family Income Has Fallen Since 1989—Should We Believe Them

A blog about business and economics.
Sept. 18 2013 10:33 AM

Median Family Income Is Lower Today Than in 1989


Census Bureau data released Tuesday reveal the striking fact that inflation-adjusted median family income has fallen slightly since 1989. This basic result appears robust to various forms of statistical quibbling around family size and family structure.

But is it right? I'm skeptical.


Consider that since 1989 houses have gotten bigger. We own more cars than we did in 1989, and the cars are better (safer, more fuel-efficient, etc.). Obviously our entertainment options have improved in terms of better televisions, MP3 players, on-demand video, messing around on the Internet. The murder rate is lower, elementary and high school students are doing better, and we are earning bachelor's degrees at a slightly higher rate. There seems to have been a small increase in the share of the population that lacks health insurance since 1989. On the other hand, there are treatments for illness available in 2012 that weren't around in 1989. And thanks to Obamacare, the uninsurance rate is falling and should continue falling for the next several years. The food (whether judging by what's in supermarkets, by what's in restaurants, or the spread of things like farmers markets) is better in 2012 than in 1989.

So where's the offsetting decline? I suppose it is possible that the quantity and quality of apparel and furniture owned by the average American family has declined so rapidly as to offset the improvements in housing, transportation, and entertainment. But it doesn't seem to me that you hear people waxing nostalgic about the great refrigerators they used to be able to afford in 1983 and how much better they are than the crap they have to settle for today.

The census data says what it says, but I don't believe it. The moral of the story seems to me to be one about the perils of using consumer price index deflators over long time horizons. An alternative explanation, given by Scott Winship, is that progressive taxation and welfare state programs account for the discrepency between rising living standards in the real world and census statistics calling it stagnation or decline. 

Note, however, that this is not relevant to the well-known fact that growth has slowed down. The 1950s and 1960s saw a period of rapidly increasing living standards. In the 1980s and 1990s, growth slowed and mostly went to the top 10 percent of the population. Since 2000 we've witnessed an even bigger slowdown and an even greater concentration of wealth. Those are the trends, and it's clear as day. But I think the contention that the average family had a higher real standard of living in 1989 than in 2012 needs a sanity check outside the BLS/BEA/Census statistical process, and I think it flunks.

Matthew Yglesias is the executive editor of Vox and author of The Rent Is Too Damn High.

  Slate Plus
June 30 2015 5:26 PM Slate Voice: “Bitch, I’ll Tell You Why This Sentence Construction Is So Effective” Listen to Katy Waldman read about the perfect sentence construction—and what makes it it’s so powerful.