Last month’s Democratic and Republican National conventions showed just how different the two parties have become on matters of race and ethnicity.
At the DNC, they praised Black Lives Matter, unauthorized immigrants, and Muslims in front an audience of delegates that was equally divided between whites and people of color. One week earlier, speakers at the RNC criticized these same groups in front of an audience of delegates that was 94 percent white.
After observing the conventions’ stark racial and ethnic differences, Bill Maher suggested that the old adage, “it’s the economy, stupid” may not apply to this presidential election. Instead, Maher said, “it’s race, stupid.”
Maher may be onto to something.
Hispanics and Asian Americans were the only demographic groups who supported Barack Obama more in 2012 than they did in 2008 — even though the economic fundamentals favored the Democrats more in 2008 than in 2012. Recent polling suggests that Hillary Clinton will perform even better among non-whites than did Obama did — even though she has the challenge winning a third consecutive Democratic term at a time of slow economic growth.
One reason is this under-appreciated fact: economic voting appears to be less prevalent among racial and ethnic minorities. Although support for the incumbent party’s presidential candidate is strongly correlated with election-year changes in gross domestic product (GDP) and real disposable income (RDI) per capita, the strength of this correlation varies by race and ethnicity.
Consider this graph:
Graph by Michael Tesler. Note: Graphs plot the residuals from a first-stage regression of incumbent vote share on president’s party and if the incumbent party is running for its third consecutive term or more. Source: Exit Polls.
The upper-left hand panel shows that growth in RDI is correlated with white Americans’ votes in presidential elections since 1976. After accounting for the president’s party and the number of terms that party has spent in the White House, each one percent increase in election-year RDI is associated with nearly a three-point increase in the incumbent party’s vote share among whites.
But there is a much weaker relationship between economic conditions and presidential vote share among African-Americans, Hispanics, and Asian-Americans. Election-year changes in RDI explain none of the variation in how these minority groups vote.
Of course, these results are based on a small sample of just 10 presidential elections conducted since the exit polls reported vote choice for whites, blacks, and Hispanics, and 6 elections for Asian-Americans. And exit polls don’t always provide reliable estimates for small, geographically concentrated groups like Latinos and Asian Americans.
So we can turn to other data sources, in particular to surveys that interviewed large samples of non-white voters: the 2012 American National Election Study, which included oversamples of black and Latino respondents; and the YouGov/Economist Poll, which has asked more than 2,000 African-Americans and 1,400 Hispanics about a Clinton vs. Trump match-up since January.
The graphs below show the correlation between how whites, blacks, and Hispanics thought the economy was doing and how much they supported Obama in 2012 and Clinton in 2016.
Graph by Michael Tesler. Numbers at the bottom represent the sample size for each category, with sample sizes for whites on the bottom, blacks on top, and Latinos in the middle.
Not surprisingly, positive perceptions of the economy were strongly correlated with white respondents’ support for the Democratic candidates. But that correlation was considerably weaker for Latinos and especially African-Americans. Less than 20 percent of whites who thought the economy was getting worse supported Obama and Clinton in 2012 and 2016. Meanwhile, nearly half of Latinos and 85 percent of African-Americans who thought the economy was getting worse still supported the Democratic candidates.
These differences between whites and non-whites are even starker after accounting for the well-established finding that partisans evaluate the economy much more favorably when one of their own is in the White House. Thus we can be a little bit more confident that economic voting is stronger for whites than non-whites.
To be sure, non-whites are not the only ones who seem to rely less on economic conditions when voting for president. Partisans and politically knowledgeable Americans are also less likely to vote based on short-term changes in the economy, compared to true independents and less knowledgeable voters.
But unlike partisans, ideologues, and the politically knowledgeable, non-whites are rapidly growing as a percent of the population. If they continue to be reliably Democratic regardless of economic conditions, then the economy would become less important in presidential elections
Of course, this does not mean that we’re inexorably headed toward a permanent Democratic majority and a world in which the economy no longer affects presidential elections. A concerted effort by the GOP to reach out to minorities might boost the party’s non-white vote share in presidential elections — especially in election years where the economy favors the Republicans. And an economic collapse on a Democrat’s watch could easily swing the presidential pendulum back to the right. Increased minority support for Democrats in a diversifying electorate could also be offset by a surge in white support for Republicans.
But there are no signs of this happening in the 2016 race. Donald Trump was not significantly outpolling Mitt Romney’s 60 percent share of the two-party white vote even before his poll numbers cratered last week. Without that boost in white support needed to cancel out Trump’s historic unpopularity among the expanding minority electorate, Hillary Clinton is now a strong favorite to overcome a slow-growing economy and win a third consecutive Democratic term in the White House.
Michael Tesler is associate professor of political science at UC Irvine and author of Post-Racial or Most-Racial? Race and Politics in the Obama Era.