My latest Model Politics post focuses on disparities in perceptions of national conditions and their political implications. One of the questions in a recent YouGov survey asked whether “the tax burden on middle-class Americans has increased or decreased since Barack Obama became president?” That seems like a pretty straightforward factual question, but Americans’ answers are all over the map: more than 40% of the respondents said that taxes have increased under Obama, about 20% said that they have decreased, and almost 40% said that the tax burden has remained unchanged.
Regular Monkey Cage readers (for example, here, here, and here) will not be surprised to learn that these disparate perceptions reflect a good deal of partisan bias. More than 70% of Republicans said that the tax burden has increased since Obama became president, while fewer than 20% of Democrats agreed. But what is more striking—and more consequential politically—is that most Americans, regardless of party, are simply wrong about what has happened to the tax burden over the past three years.
The fact is that the tax burden on middle-class Americans has decreased during Obama’s presidency. More than one-third of the 2009 stimulus bill consisted of tax cuts, including expanded tax credits for workers, people with children, college students, homebuyers, and the unemployed. In 2010, Obama proposed and Congress accepted a substantial temporary reduction in the payroll tax, which was recently extended through 2012. Meanwhile, the Bush-era income tax cuts were also extended through 2012. While one might quibble about whether all of this amounts to decreasing the tax burden on middle-class Americans by “a little” or “a lot,” only 20% of the public gave either of those answers.
How consequential is the American public’s phantom tax hike? The following graph shows how support for Obama in a trial heat with Mitt Romney varies with perceptions of change in the middle-class tax burden during Obama’s time as president (ranging from -100 for “decreased a lot” to +100 for “increased a lot”). The relationship is shown separately for Democrats (including “leaners”), pure Independents, and Republicans. In each case, statistical controls are included for strength of partisan identification, political ideology, education, race, and sex.
The large dot along each line shows the average perception of the tax burden and support for Obama in the corresponding partisan group. Democrats are just to the left of the zero point, indicating that they were slightly more likely to say that the tax burden went down than that it went up. However, even they are well to the right of the -50 point on the horizontal axis, which would reflect a uniform perception that the tax burden had declined “a little.” Independents are even further to the right, on average, with Republicans still further to the right.
If Obama could convince everyone in the country that the tax burden on middle-class Americans has decreased just a little, his prospects for reelection would improve significantly. Even leaving aside the unlikely steep gain among Republicans implied by the figure, my analysis suggests that simply getting Democrats and Independents to appreciate that taxes are lower now than they were under President Bush would increase Obama’s vote share by about two percentage points.
Of course, it would be rash to count on that happening. Daniel Patrick Moynihan famously said,“Everyone is entitled to his own opinions, but not to his own facts.” But Americans rely on their own facts all the time, including in the voting booth.