Did Welfare Reform Work?

Fifteen years after we passed welfare reform, did it work?  Ezra Klein takes on this question today.

If welfare reform was meant to cut the rolls, then it definitely worked.  And if it was meant to give states the flexibility to cut their spending on the program, it definitely worked. . . If you think the point of the program is to help the poor, then no, welfare reform is not working.

As Jake Blumgartwrites at The American Prospect, the reformed program "has failed to cushion the neediest through recessions. While in 2009 the food-stamp program responded to the increased need for government assistance, growing by 57 percent, the number of TANF caseloads merely inched upward...At the heart of the worst recession in 80 years, TANF funds only reached 4.5 million families, or 28 percent of those living in poverty. By contrast, in 1995, the old welfare system covered 13.5 million families, or 75 percent of those living in poverty."

Another possible definition of "working" is that the program has helped or forced a lot of low-income Americans, and particularly single mothers, find jobs. In the late-1990s, when the labor market was very tight, there's strong evidence that welfare reform was helpful in pushing people into the job market. In the Aughts and, in particular, since the recession has hit, it's a lot less clear that welfare reform is increasing employment rather than simply limiting support for the unemployed.
Thumbnail image for tanf benefits drop.jpg
Ezra includes this graph from the Center on Budget and Policy Priorities as evidence that welfare reform has not "worked" in any real sense.  As you can see, the percentage of poor families receiving TANF (the successor to AFDC, aka "welfare") has fallen dramatically since welfare reform was enacted.

But I'm not sure why this is supposed to be an indictment of the system.  Why is it a problem that fewer poor families are enrolled in a program that is only open to people who aren't working?  The American Prospect and the Center on Budget and Policy Priorities can't possibly be lamenting the fact that we no longer have more than 70% of our poorest families on a program that has unemployment as a prerequisite.  But the way this graph is used makes it sound like they consider this regrettable.

Ezra mentions that reform moved many welfare mothers into jobs, but I think he gives this short shrift.  Leave aside the tiresome bourgeois morality which wants to see people trying to support themselves before they turn to the generosity of their neighbors.  People are not made better off by a program that encourages them not to work--as AFDC indisputably did, given the decline in the rolls.

Don't get me wrong: it's entirely understandable that people would prefer to collect welfare rather than work long hours at an unpleasant low-wage job.  But someone who collects welfare today rather than go to work for $7 an hour is very likely to be collecting welfare ten years from now, when it will still be a rather joyless existence hemmed in by lack of money and the whims of the bureaucracy.  Someone who is working at anything has their feet on a path that might actually lead somewhere.  As anyone who has suffered through a long spell of unemployment can attest, it's hard to get back into the workforce if you've been out of it a while.  Harder still if you were never really in it, developing basic skills like showing up on time every day and handling difficult customers.

Welfare enabled people to make bad long-term decisions that were rational short-term choices.  Welfare reform changed that.  That's good news.

Of course, it's bad news that the mothers who went out to work didn't all gain the comfortable middle class existence we'd ultimately like for them.  But there was still a noticeable decline in the number of poor families that persisted even into the early years of the Great Recession:


This looks like a modest but real success to me at weaning families from welfare dependency.  Even at the nadir of the worst recession in eighty years, the percentage of families in poverty--as well as the percentage of families on TANF--was below pre-reform levels.  Unless you really think that these families would be better off spending the rest of their lives on the dole, this seems like a real achievement.

There's another reason that progressives should celebrate: changing the structure of welfare has eroded much of the opposition to it.  As long as people felt like welfare was a way for people to simply live off of tax dollars without working, there was bound to be a lot of opposition to the program.  Restructuring it as temporary assistance for those who are overwhelmed by unexpected circumstances has essentially whittled that opposition down to nothing.  When was the last time that welfare came up in an election?

Sure, maybe progressives would prefer that a generous system of benefits for anyone who wanted them was the uncontroversial norm--but that doesn't really seem very realistic in a pluralistic and fairly conservative country like America.  By ending welfare as we knew it, Clinton preserved the safety net for people who really can't cope.  If he hadn't, welfare mothers would now be competing with retirees for money in the Great Deficit Reduction Olympics.  And I think we all know who would have lost that race.

Update: several commenters think I should have included the two sentences now at the end of the Ezra Klein quote, which I initially left out because I was already in danger of grabbing the whole post.  They think it changes my post.  I disagree, because my point remains the same: Ezra is giving short shrift to the succesful drive to move people into work. But I can also see why people felt like my clip was misleading, so I've added it, and sorry, Ezra, for grabbing nearly your whole post.

Now about those sentences . . .

After saying "If you think the point of the program is to help the poor, then no, welfare reform is not working", Ezra acknowledges that it was nice that people moved into work in 2000s, but dismisses this achievement because the trend did not continue to steadily decline towards zero. This is pretty much the standard progressive line on welfare reform--it only looked like it was working because of the awesome Clinton economy--and it's not correct.

It's not, in fact, in question whether we produced a permanent change; we did.  There was a substantial structural decline in the percentage of families in poverty which persisted into the aughts.  I could have included the percentage of female headed families in poverty, or children in poverty, and they would have shown the same trend: all of them clearly inflected downwards around welfare reform.  All ticked up during the 2001 recession, but clearly settled at a level much lower than their pre-reform average.  I find this hard--actually, impossible--to square with Klein's assertion that if you think the purpose of reform was to help needy families, then no, it hasn't worked.

Dismissing the achievements of welfare reform because the poverty rate didn't decline towards zero makes no sense to me.  While it would be nice if it had happened, no one really expected it to.  The fact that a miracle failed to materialize is hardly a searing indictment of reform.  You can argue that the decline in the poverty rate was assisted by other reforms like boosting the earned income tax credit, and I completely agree.  But boosting the EITC does nothing to help people who aren't earning income.  If we hadn't done welfare reform, "not earning income" would still describe the majority of poor families.