It only took twelve years, but the New York Times has discovered the fundamental problem with the 1996 welfare reform: It doesn’t work in a deep recession!
This should not be news. While the law was hailed as ending the “entitlement” to welfare, that did not mean that individuals lost an automatic entitlement to benefits, which they never had anyway. What it meant was a change in financing, so that in place of the old AFDC structure, under which states were automatically reimbursed by the federal government for 50% of their welfare costs (more for poorer states), states now get a fixed block grant based on their caseload in the early 1990s, when welfare caseloads were at all-time high points. The old system was flexible and based on need – when poverty and caseloads rose, so did federal spending. The post-1996 system, known as TANF, is inflexible and breaks any connection with the economic cycle.
The work requirements and time limits that got the most attention in the welfare reform debate were unrelated to this huge change. Work requirements and time limits could have been added to the older system, as many states had already done. But the Republican ideologues were eager to break the system entirely, and governors of both parties loved the block grant, because with caseloads already dropping and the economy booming, it was certain to result in far more money than they would have received under the old matching system, at least for as long as any of them were going to be in office.
That was money they could use for tax cuts and roads, but the law required them mostly to recycle it into programs for welfare recipients. As the economy boomed and teen child-bearing (a major driver of rising welfare caseloads) plummeted, states were able to throw considerable resources at the remaining welfare families, giving many of them enough of a foothold in the job market that they weren't the first-fired in the recession of 2001. To this extent, welfare reform could be considered a success, but only because its huge gamble on the economy paid off in the later 1990s.
Twelve years later, the block grant structure is showing its age and its inflexibility. Populations have grown, the recession is particularly rough for women, and states don't have any flexibility. You don't build a social safety net for boom times, you build it for rough times.
That's a point that few people wanted to talk about that at the time. Poverty, they thought, was not about the economy, it was a matter of culture and incentives. Defending the “entitlement,” meaning just the economically responsive financing system, was seen as a stale defense of the old system. (The late Senator Daniel Patrick Moynihan raised the issue, as did my boss at the time, Bill Bradley.) Reading the Times article, I couldn't help being a little annoyed, because among those who were obsessed with work requirements, time limits, and the "culture of welfare," while paying little attention to the long-term economic risk posed by changing a flexible safety net program into a block grant, was the leading reporter on welfare at the time, Jason DeParle of the New York Times. DeParle is also the author of today’s article.
That said, welfare reform has on the whole been a greater success than many of us expected, partly due to the good economy, and twelve years of success is better than a lot of programs can show. But the huge problems DeParle shows it encountering in this recession were entirely predictable, predicted, and avoidable. If we’d kept the old AFDC financing structure, added work requirements and time limits, with enough funds for training and education that few parents would ever actually reach the time limits, we could have had a welfare system over the last decade that simultaneously would have cost less in total than the block grant during the boom years, would have been just as effective at helping parents move into the workforce, and would have served the basic counter-cyclical economic function of a social safety net program that we desperately need now
-- Mark Schmitt
