Mitt Romney has stuck with his theme of attacking President Obama over "gutting" the work requirement for welfare. In an often-aired television commercial, Romney's claims that Obama "quietly" dropped the requirement that welfare recipients find work and instead they "just send you a check".
As Romney's ad loosely indicates, in 1996 Bill Clinton signed a law that replaced the federal program we call "welfare" (Aid for Families with Dependent Children) with "Temporary Assistance for Needy Families". The major change in this law was that recipients, with few exceptions, must find a job within two years of beginning the program, rather than remain on welfare indefinitely. Romney claims that Obama removed this provision of the law.
Of course, Romney’s claim is false. But are these ads effective politics? The answer appears to be yes – at least, if we just look at whether people believe the ads to be true.
I asked a nationally representative sample of adults if they had heard about Obama's "announcement" about welfare: 1 out of 3 respondents (31%) said they had heard and, of these, almost half (47%) had said that they thought it was a true story. So, if we believe that voters are against "gutting" the work requirement (which the Romney campaign clearly believes), then Romney has effectively conveyed this message about Obama gutting welfare to about 15% of voters.
However, once we look a little closer at these survey responses, we can see that if these ads are effective, it is most likely not because the ads are changing opinion, but rather because they are tapping into opinions that voters already hold - in this case opinions that are squarely rooted in attitudes about race.
First, it's worth noting that most people don't know that there is a work requirement. In the survey, I asked about welfare policy by giving respondents five options and asked them to choose which one described the law's work requirement provisions. Limiting the responses to white voters, fifty percent chose "don't know". Only 20% chose the correct answer of two years, which is about what we'd also expect if everyone was just guessing. In total, about 29% of persons thought there was some kind of work requirement, which is higher than the 20% that thought there was no work requirement, but still not a ringing endorsement of America's knowledge of the public policy: most people just said they didn't know or were exactly wrong (of course, all this misinformation could be because of Romney's ads, more on that in a minute).
However, these surveys are about much more than just ignorance, these answers demonstrate the influence of attitudes about race in public policy. To begin, we can look at the plot below. Here I divide up whether or not people believe this false rumor based on their level of resentment towards African Americans. This "resentment" is measured using a very common survey scale that asks whether African Americans would better off in American society if they just worked harder or, rather, if it is primarily generations of slavery and discrimination that makes it difficult for them to get ahead. Looking at this plot, you can see that persons with high levels of resentment against African Americans, indicated by the blue bars, were five times as likely to say that Obama really did "gut" the work requirement than are the person with low resentment towards African Americans, indicated by the green bars (71% to 15%).
If we further want to see the relationship between racial attitudes and welfare, we can look at the next plot, which shows whether or not people think there is a work requirement for welfare recipients, again divided into persons that have high and low resentment towards African Americans. Notice that a plurality of both high and low resentment persons know that there is a work requirement for welfare recipients, but low resentment persons are much more likely to know there is requirement than high resentment persons (58% to 43%), and that a much greater portion of high resentment persons incorrectly believe there is no work requirement (31% versus 7%).
So, persons that have high-levels of resentment against African Americans also have a tendency to think that welfare comes with no work requirement and that Obama really did gut the work requirement. One interpretation of this finding is that it logical conclusion to draw from Romney's advertisements. If Obama really did gut the work requirement, then by the time I ask these survey questions, there was no work requirement because Obama gutted it! Such reasoning makes sense, but more likely what is holding these attitudes towards work, welfare, and Obama together is the common association with African Americans – and a resentment of African Americans – rather than logical consistency.
We can see the influence of resentment, rather than logical consistency, by looking at the plot below, here I show the percentages of persons that believe Obama really did gut the work requirement divided by resentment towards African Americans and whether or not respondents think there is a work requirement for welfare recipients. As on the previous graph, large majorities of persons with high resentment towards African Americans believe that Obama gutted the work requirement, far more than the low resentment persons – more interesting is that high-resentment persons that believe there is a work requirement are actually more likely to believe that Obama gutted the work requirement than high-resentment persons that say there is no work requirement (77% versus 68%).
The take away point here is that if a white American resents African Americans than they are likely to believe the Romney ads are telling the truth whether they believe that there is a work requirement for welfare or not.
These associations hold even if I statistically control for political ideology, partisanship, income, age, and education. The single attitude that far more powerfully than any other predicts whether a person believes that there is no work requirement for welfare is resentment towards African Americans. And controlling for these same variables, resentment towards African Americans is the factor that most powerfully, by far, predicts whether a person believes Obama really did gut the work requirement.
These attitude associations should come as little surprise to those that are familiar with scholarship on racial politics in the United States, such as that of Martin Gilens, who demonstrates that Americans tend to have negative attitudes about welfare because of its association with African Americans. At the same time, Michael Tesler and David Sears, have shown that attitudes about Obama are powerfully predicted by resentment towards African Americans. It is this resentment that likely ties together attitudes about welfare and Obama.
Like most questions of complex public policy, Americans do not have the inclination or ability to pay careful attention to the nuance of the issue – instead they rely on shortcuts, which psychologists call heuristics, to guide their attitudes about the policy. Here people likely rely on the heuristic provided by their pre-existing feelings towards African Americans: they don't like welfare, they especially don't like welfare without a work requirement, they don't like Obama, and they associate African Americans with all these things – making it easy to simultaneously hold negative attitudes about all of them.
With their ads, the Romney campaign is simply "priming" the easy association some voters have between welfare and Obama and the negative feelings associated with both – unfortunately, to do this, Romney must cynically rely on the connecting thread of racism that makes this association possible.