The Foundation for Government Accountability (whose staff contribute to NRO on occasion) has an interesting new report on what happened to the families Kansas removed from the welfare rolls via stronger work requirements. Able-bodied-adult enrollment in the state’s cash-welfare program plunged 78 percent between 2011 and this year, compared with just 14 percent nationally.
The FGA was given detailed data on 6,090 families including more than 17,000 individuals. In the first year after being booted off welfare, the group’s wages more than doubled. For the families with a full four years of data (about 1,000), wages had tripled by the end of the period. Overall income, including wages, cash welfare, and the Earned Income Tax Credit (EITC), rose about 50 percent over four years.
These are valuable data, and it’s worth digging into them and thinking through what we can and can’t conclude.
One key question is how many people failed to find work. The authors told me via e-mail that “about 80% had some record of employment after being removed from the program, but there was some fluctuation of people moving into and out of the labor force. About 65% were working in any given year after removal.”
Another question is just how well-off these families became. At the baseline, the group made about $48 million in total income (including welfare and the EITC); after four years they made $73.4 million (a number extrapolated from the families with four years of data). Divide those numbers by 6,090, and you see that the average family income rose from about $7,900 to about $12,100 — which is roughly the poverty line for a single person, whereas the families in the study have nearly three members each on average. Using the same back-of-the-envelope math, wage income alone rose from $3,200 to $11,100.
In general, then, these folks have a long way to go to reach the middle class, even if we should be glad they’re becoming more self-sufficient.
Further, a more technical point: There is no control group here, and thus we don’t know how these families would have fared without the reform. In other words, we can’t say that the work requirements caused the improvement the study documents. Even before the federal welfare reform in 1996 there was a lot of churn on and off the rolls, and part of the 1996 reform was a five-year time limit, so it’s possible that many of these folks would have found work even without Kansas’s new, stricter policy. To use the statistical term, the authors may partly be capturing “regression to the mean”: If you start monitoring people when they have extremely low incomes, you’ll almost certainly see their incomes rise over time. This is especially likely if a recovery is underway; Kansas’s unemployment rate has fallen almost continuously since peaking in mid 2009.
One last nitpick: The extra income came from wages and wage-based tax credits, so it required these families to spend money on job-related transportation, child care, etc. That cuts into any gains to their overall financial well-being.
In short, these new data show that when families left welfare, they didn’t end up permanently mired in extreme poverty — in fact, they made more money afterward than they had before. But there are legitimate questions about whether the reform itself caused this result and whether some families were made worse off. Hopefully Kansas will continue to collect and study data on the population it removed from the rolls.