All Commentary
Wednesday, March 6, 2019

Why Child Poverty Plunged in the US over the Last Two Decades

Without implementing a single reform, we might be closer to achieving the project’s stated goal than we even realize.


For the last almost two years, a team commissioned by the National Academies of Sciences, Engineering, and Medicine (NAS) has worked on what they call “an agenda to reduce the number of children in poverty by half in 10 years.” That’s an ambitious and worthy goal, but I found their final report something of a mixed bag: Clearly this was the work of thoughtful scholars who recognize that our safety net already contributes to a large reduction in child poverty, but their report shows we need to do more to get better poverty measurements into the mainstream, and credit the work-first reforms of the 1990s for our success fighting poverty in the past.

Supplemental Poverty Measures vs Consumption

First, what I liked: The best thing about the NAS report is its acknowledgment of the progress we’ve made fighting child poverty. I’m thrilled that they seem to have moved on from the official poverty measure (OPM) as an indication of how widespread child poverty is in our country. The OPM, as the report acknowledges, does not do a good job of capturing who is really in poverty—only who does not have a lot of earned income.

Consumption poverty does the best job of telling us who is suffering.

The report instead uses a version of the Supplemental Poverty Measure (SPM) that accounts for some transfer programs and even how the benefits of some of those programs can be under-reported. This “adjusted SPM” shows some real progress, culminating in an all-time low child poverty rate of 13 percent in 2015, down about 15 points since the mid-1990s.

Of course—and here’s my first criticism—it would have been nice for the authors to be less dismissive of consumption poverty as the bar against which we measure progress. Consumption poverty, which accounts for transfers as well as under-reporting of income and benefits, does the best job of telling us who is suffering (an all-time low of 2.8 percent of Americans in 2017). That would seem to be the major concern of a report that spends quite a lot of time on the various negative consequences associated with childhood poverty.

But even if we simply accept the SPM with adjustments, we see progress, as with the numbers below, from 2015:

And if we think about 2016 to the present, a time period that has seen healthy economic growth and a continuously robust safety net, we can reasonably extrapolate that things have gotten even better. By my estimate, we are likely at or below 12.5 percent SPM with the report’s preferred adjustments—that’s a reduction by half since 1994, and by more than a quarter in the last seven years.

So without implementing a single reform, on the power of lots of people getting to work and wages rising, we might be closer to achieving the project’s stated goal than we even realize.

Inaccurate Diagnosis 

Here’s what I didn’t like. The report’s authors seem to refuse to acknowledge what got us to this point: The success of welfare reform—the undeniable truth of the 1990s-trendline below—with a focus on work-first public assistance combined with a healthy economy.

The report comes close to giving welfare reform its due credit, but stops just short:

The decline in two-parent family structure is the single biggest factor associated with the increase in child (official) poverty between the mid-1970s and the early 1990s. However, child poverty has fallen since the early 1990s, despite continuing increases in single parenthood. This more recent decline in child poverty is most strongly associated with increases in maternal employment. (Conclusion 4-2, Emphasis added)

The report deserves credit for implying that a return to a norm of two-parent homes would serve America’s children well. But it should credit the work-first reforms of the ’90s (starting with state reforms in the early part of the decade and then federal reform in 1996) for those increases in maternal employment. No one denies that many single mothers moved to employment and earnings because of welfare reform, and the report admits that it made a serious dent in child poverty.

Unfortunately, it seems as though the fear of advocating for more work-first reforms—like the work requirements under debate for SNAP and other programs—won out over the obvious lesson this all points to. Sadly, as Matt Weidinger noted, the report ultimately recommends more transfer payments rather than more work.

The report ends up being an admirable attempt to tell us where we are. It does not do as well in telling us how we got here, and that detracts from its ability to tell us where to go next.

This article is reprinted with permission from The American Enterprise Institute.


  • Robert Doar is the Morgridge Fellow in Poverty Studies at the American Enterprise Institute (AEI), where his research focuses on how improved federal and state antipoverty policies and safety net programs can reduce poverty, connect individuals to work, strengthen families, and increase opportunities for low-income Americans and their children.

    Mr. Doar has a bachelor’s degree in history from Princeton University and has testified numerous times before Congress, and his writing has appeared in The Wall Street Journal, USA Today, The Hill, and National Review, among other publications.