(Nacho Doce/Reuters)

Much debate continues about whether support for Donald Trump has more to do with racial or economic anxiety. A key question in this debate — explored by Wonkblog’s Jeff Guo — is whether economic anxiety may actually cause racism. Guo shows, for example, that Americans who think the economy is getting worse currently score highest in racial resentment.

So which is the chicken and which is the egg? The evidence suggests that racial resentment is driving economic anxiety, not the other way around.

One reason is that perceptions of the economy are often not objective and depend on people’s political leaning. A large body of research shows that party identification strongly colors people’s beliefs about how the economy is doing. Democrats and Republicans both think that the economy is performing better when one of their own is in the White House.

Partisan identities aren’t the only thing that matters. In my book, Post-Racial or Most-Racial?, I show that racial attitudes have increasingly structured public opinion about a wide array of positions connected to Barack Obama, including subjective perceptions of objective economic conditions.

For one, racially sympathetic white Americans were far more likely than racially resentful whites to correctly conclude that the unemployment rate was declining in the year leading up to the 2012 election. Before Obama’s presidency, racial attitudes were uncorrelated with perceptions of the election-year unemployment rate.

The comparison between 2004 and 2012 is especially informative. Both George W. Bush and Obama saw the unemployment rate rise by about two percentage points at various times during their first terms in office; both presidents then presided over drops in the unemployment rate during the year leading up to their reelections (about half a point for Bush and one point for Obama).

Yet the graph below shows that racial resentment had a much different impact on perceptions of the unemployment rate in 2004 and 2012. After accounting for partisanship and ideology, racial resentment had no effect whatsoever on perceptions of the unemployment rate in 2004. But in 2012, people who expressed more racial resentment were less likely to perceive that, in fact, the unemployment rate had improved.

Analysis limited to whites only. Predicted values calculated by setting party identification and ideological self-placement to the average white respondent. (Graphic by Michael Tesler)

This suggests that the national economy’s association with Obama has made racial resentment a stronger determinant of gloomy economic perceptions than it was before his presidency. However, comparisons between 2012 and earlier years cannot conclusively resolve the chicken or egg question.

To do so, it’s important to have surveys of the exact same individuals before and after Obama became president. The 2007-2008-2012 Cooperative Campaign Analysis Project can do this by testing whether racial attitudes — measured before Obama became president — increasingly shaped economic perceptions during his presidency.

The results below show that this is precisely what happened.  Racial resentment was not related to whites’ perceptions of the economy in December 2007 after accounting for partisanship and ideology. When these same people were re-interviewed in July 2012, racial resentment was a powerful predictor of economic perceptions. Again, the greater someone’s level of racial resentment, the worse they believed the economy was doing.

Analysis limited to white panelists interviewed in both the December 2007 and July 2012 wave of the CCAP Re-Interviews. Predicted values calculated by setting party identification and ideological self-placement to the average white respondent. (Graphic by Michael Tesler)

Furthermore, additional analyses indicate that economic perceptions, whether measured in 2008 or even in 2012, did not cause people to change their underlying levels of racial resentment.

In fact, multiple studies, using several different surveys, have shown that overall levels of racial resentment were virtually unchanged by the economic crash of 2008. Some data even suggests that racial prejudice slightly declined during the height of economic collapse in the fall of 2008. The evidence is pretty clear, then, that economic concerns are not driving racial resentment in the Obama Era.

Of course, this doesn’t mean that economic anxiety has no influence on support for Trump. John Sides and I presented some preliminary evidence that economic insecurity was a factor in Trump’s rise.

Nor does it mean that racial resentment is the prime determinant of economic anxiety. It isn’t.

Nevertheless, in an era where racial attitudes have become increasingly associated with so many of the president’s positions, Obama’s race is largely responsible for the association between racial resentment and economic anxiety. And this racialized political environment undoubtedly aided Donald Trump’s rise to the top of the Republican Party.

Michael Tesler is associate professor of political science at the University of California, Irvine, and author of Post-Racial or Most-Racial? Race and Politics in the Obama Era.