A man arrives to vote at Brooklyn Avenue Elementary School in Los Angeles on Election Day in 2016. (Frederic J. Brown/AFP/Getty Images)

With the primaries over, the focus now is the strategies each party will use to win in November. Many Democrats are pushing to mobilize the party’s base, especially people of color.

They point to races where the votes of racial and ethnic minorities appeared crucial — from Doug Jones’s surprise win in December 2017, to upset primary victories by Alexandria Ocasio-Cortez in New York and Andrew Gillum in Florida. And in some races, this strategy is clearly necessary: Beto O’Rourke’s fortunes in Texas’s Senate race largely depend on Latino turnout.

But Democrats confront a major challenge: People of color vote at a substantially lower rate than whites. In my new book, “The Turnout Gap: Race, Ethnicity, and Political Inequality in a Diversifying America,” I explain why. Below are four takeaways on racial differences in who votes.

1. Minority turnout lags behind white turnout — by a lot

For much of our nation’s history, the right to vote was granted to only white men. But eventually the Voting Rights Act of 1965, and later amendments and extensions in the 1970s and 1980s, dismantled many barriers that kept African Americans, Latinos and Asian Americans from the polls.

Despite these successes, minority turnout continues to trail that of whites. This is especially true in midterm elections, as you can see in the figure below.

In 2014, for example, 54 percent of whites reported voting, compared with 49 percent of African Americans — a gap not much different from the figures of the 1980s and 1990s. Latino and Asian turnout was even lower: Among voting-age citizens, only 33 percent of Latinos and 35 percent of Asian Americans voted. The 20-point gap between them and whites is the largest on record.

In fact, the gap is probably larger. Most measures of racial differences in turnout come from surveys such as those used in the graph above. But in my book, I show that surveys underestimate the turnout gap by as much as 10 to 15 percentage points.

2. Racial turnout differences hurt Democrats

The large turnout gap has real electoral consequences. Because whites have increasingly tended to vote for Republicans and minority groups for Democrats, turnout disparities unsurprisingly hurt Democrats at the ballot box.

But how much? Using survey data that asks nonvoters for whom they would have voted, I determined what elections would have looked like if minorities turned out at the same rate as whites. The table below shows how Democrats’ fortunes in the Senate and electoral college would have changed in this scenario.

F­or one, without the turnout gap, Democrats would have retained more Senate seats. If minority turnout had been equal to white turnout, Democrats would have held on to a majority in 2016 despite their losses in 2010 and 2014. Merrick Garland might have ended up on the Supreme Court, after all.

Closing the turnout gap would not have affected the 2008 or 2012 presidential elections very much, but it would have made an enormous difference in 2016. If minority turnout had been equal to white turnout (or equal to what it was in 2012), Hillary Clinton would have won the electoral college handily.

3. Vote suppression does not explain the turnout gap

What explains these persistent disparities in turnout between minorities and whites? Many focus on “vote suppression,” or election practices such as voter identification laws, reductions in early voting and precinct closures — all of which are often compared to Jim Crow-era restrictions. Indeed, voter suppression is implemented in places where reducing minority turnout would help Republicans.

But do these laws actually have a large impact on minority turnout? That is not clear. In “The Turnout Gap,” I show that states with these policies actually tend to have higher minority turnout, both before and after implementation. In a separate study of Texas, Michael Miller and I find that voter identification laws disproportionately affect minorities, but the effects are nowhere near large enough to explain the turnout gap.

Researchers are still working to measure the full impact of policies such as voter identification. But it is difficult to say that these policies are responsible for the turnout gap.

4. The gap is smaller in places where minority citizens control election outcomes

So what is more strongly associated with the turnout gap? There are several other theories, including whether minority candidates are running, targeted mobilization, and a sense of empowerment and efficacy. I argue that these things depend on a crucial underlying factor: whether minority voters are a large share of the population in the first place.

In places with larger numbers of minorities, the turnout gap is much smaller. For example, the maps below show that the black-white turnout gap is smaller in Southern states where the African American population is larger. The Latino-white turnout gap is smaller in the heavily Latino states of the West and Southwest.

This relationship is even stronger in congressional districts — especially because “majority-minority” districts ensure that minority groups can dominate elections.

But there are still relatively few places where minorities have such voting power. Non-Hispanic whites are about 70 percent of the voting-eligible population as well as the largest bloc of potential voters in every state except for Hawaii. And even majority-minority districts might become endangered if the Supreme Court further limits the Voting Rights Act. Moreover, many Latinos and Asian Americans are located in states such as California and New York, where Democrats can count on significant support from whites anyway.

In short, there are no easy ways to close the turnout gap between minorities and whites. This leaves Democrats far more reliant on white support than many assume. Changing this dynamic would require a stronger effort to mobilize minorities with targeted tactics, but even these tactics may not be enough. If so, U.S. elections will continue to fail to reflect the views of all Americans.

Read more:

Bernard Fraga (@blfragais assistant professor of political science at Indiana University, and author of The Turnout Gap: Race, Ethnicity, and Political Inequality in a Diversifying America (Cambridge University Press, 2018).