Mitt Romney and his campaign ads have been slamming Barack Obama over his administration's treatment of welfare recipients. A recent spot accused Obama of "gutting welfare reform," and Romney said the president had done so to "shore up his base." Fact-checkers have already debunked the first part of these accusations. A few statistics will debunk the second.
In fiscal year 2009, during the depths of the recession, about 2 million families received welfare benefits from the Temporary Assistance for Needy Families program. In these families, 85% of adult recipients were single parents; the rest were married. I did the math, and I found out that the total number of adults in these families had to be less than 2.2 million. In other words, a maximum of about 3 percent of Barack Obama's 69.5 million votes in 2008 came from people on welfare -- and that's assuming every single adult on welfare voted, and they all voted for Obama.
What are the chances that they all voted for Obama? Well, I can't find any exit polls that asked voters whether they were on welfare. We can try to estimate how people voted using other statistics, though. Here's how: look at the total number of adults who receive benefits, and use a combination of exit polls and a broad-based social survey to try to estimate how they voted in 2008.
We do know that almost all black voters voted for Obama. If welfare recipients were no different, then Obama could have counted on a little more than a third of welfare recipients' votes. That's a maximum of about 1 percent of his base.
For people of other races -- which didn't vote so uniformly for Obama -- we might get better estimates by looking at the General Social Survey, which asks respondents about their incomes and voting histories. In the 2010 survey, I looked at non-black respondents with children whose families earned less than $20,000 a year, which is a rough approximation of the eligibility standard for welfare. Of course, many families who fit this description don't receive welfare, and the survey's results are based on a sample of just a couple thousand people. But these data might still give some good hints about welfare recipients' political preferences.
According to the data, about three quarters of these poor, non-black adults voted in the presidential election. Among the voters, 50% went for Obama. So among non-black welfare recipients, perhaps three in eight might be part of Obama's base. That share would amount to roughly 0.75% of his overall base.
We can use the same methodology to check the earlier supposition that all black welfare recipients did indeed vote. In fact, only about 75% of the blacks with children and family incomes under $20,000 did vote in 2008. If that also held true for welfare recipients, they could have made up no more than 0.75% of Obama's overall base.
Adding the figures in the last two paragraphs together, I got my final estimate: welfare recipients probably made up no more than 1.5% of Obama's overall voter base in 2008.
Because Mitt Romney has said that he's "not concerned about the very poor," Obama probably doesn't have to do much to ensure that these people stay in his corner. So why would the president change legislation just to please such a tiny portion of his base, when they're likely to vote for him again, anyway? It just doesn't add up.