Presumptive Democratic nominee Joe Biden leads President Trump by more than 20 points on race and policing, which have become top issues on the minds of voters amid nationwide protests against police brutality.

The latest Harvard CAPS-Harris Poll finds that 61 percent of voters say Biden would be better at solving issues of race and policing, compared to 39 percent who said Trump.

Sixty-two percent of voters say Biden would do a better job of bringing the country together following weeks of protests over the death of George Floyd while in police custody. 

And 53 percent of voters say Biden would do better at establishing law and order, compared to 47 percent for Trump.

Race relations is now tied with the economy as the second most important issue for voters, behind the coronavirus.

Trump’s job approval rating on race and policing is at 43 percent positive, the same as his overall job approval rating in the poll. Fifty-three percent of voters oppose Trump’s handling of the protests, although a majority, 57 percent said the president was right to call the national guard in to deal with protesters in Washington, D.C.

“Biden did very well on being someone who could bring the country together compared to President Trump and so Trump’s overall response, despite support for the guard and even the military, has fallen short as his numbers have definitely declined,” said Mark Penn, polling director for the Harvard CAPS-Harris survey.

The poll found deep racial divisions over policing in the U.S. but there is broad support for both the police and the Black Lives Matter movement.

Sixty percent have a favorable view of the police, while 55 percent view Black Lives Matter favorably.

A plurality, 45 percent, say they personally support the aims and policies of Black Lives Matter, compared to 33 percent who oppose them and 22 percent who don’t know.

Nearly two-thirds of voters say they believe the police mostly operate fairly and are unbiased, including 73 percent of white people. But 70 percent of black people said police are biased and operate unfairly.

Fifty-six percent said police misconduct against black people is fairly common. Whites are evenly split on that question, but 85 percent of black Americans said it is common.

A strong majority, 72 percent, said the police should not be defunded. Forty-six percent said police funding should remain the same, 28 percent said funding should be increased and 26 percent said it should be decreased.

A plurality of voters, 47 percent, said the police should undergo major reforms or restructuring, including 53 percent of white people and 80 percent of black people. Forty-six percent overall said the police only need minor restructuring or improved training.

There is overwhelming support for all police to wear body cameras, for a national registry of disciplined officers, to ban the use of chokeholds, and to strip the police of immunity from civil suits.

Fifty-eight percent of voters said racism is systematic and pervasive in the United States.

“The poll shows tremendous amount of support for both black lives matter and the police,” Penn said. “They want police reform, not defunding. They support broad police reforms but don’t want to cut the number of police. They recognize significant police misconduct and systemic racism and also supported bringing in the national guard.”

A plurality of voters, 47 percent, said the response to riots and looting in major cities has been too soft. Twenty-nine percent said the response has been adequate and 23 percent said it has been too harsh.

Sixty-seven percent said protests are an appropriate response to Floyd’s death.

Fifty-two percent said they would support the U.S. deploying the military to U.S. cities to control protests that get out of control.

The Harvard CAPS/Harris Poll online survey of 1,886 registered voters was conducted June 17-18. It is a collaboration of the Center for American Political Studies at Harvard University and The Harris Poll. The Hill will be working with Harvard CAPS/Harris Poll throughout 2020.

Full poll results will be posted online later this week. The survey is an online sample drawn from the Harris Panel and weighted to reflect known demographics. As a representative online sample, it does not report a probability confidence interval.