Should Biden Emphasize Race or Class or Both or None of the Above?

Should the Democratic Party focus on race or class when trying to build support for new initiatives and — perhaps equally important — when seeking to achieve a durable Election Day majority?

The publication on April 26 of a scholarly paper, “Racial Equality Frames and Public Policy Support,” has stirred up a hornet’s nest among Democratic strategists and analysts.

The authors, Micah English and Joshua L. Kalla, who are both political scientists at Yale, warned proponents of liberal legislative proposals that

Despite increasing awareness of racial inequities and a greater use of progressive race framing by Democratic elites, linking public policies to race is detrimental for support of those policies.

The English-Kalla paper infuriated critics who are involved in the Race-Class Narrative Project.

The founder of the project, Ian Haney López, a law professor at Berkeley and one of the chairmen of the AFL-CIO’s Advisory Council on Racial and Economic Justice, vigorously disputes the English-Kalla thesis. In his view, “Powerful elites exploit social divisions, so no matter what our race, color or ethnicity, our best future requires building cross-racial solidarity.”

In an email, López wrote me that the English and Kalla study

seems to confirm a conclusion common among Democratic strategists since at least 1970: Democrats can maximize support among whites, without losing too much enthusiasm from voters of color, by running silent on racial justice while emphasizing class issues of concern to all racial groups. Since at least 2017, this conclusion is demonstrably wrong.

English and Kalla, for their part, surveyed 5,081 adults and asked them about six policies: increasing the minimum wage to $15; forgiving $50,000 in student loan debt; affordable housing; the Green New Deal; Medicare for All; decriminalizing marijuana and erasing prior convictions.

Participants in the survey were randomly assigned to read about these policies in a “race, class, or a class plus race frame,” English and Kalla write.

Those given information about housing policy in a “race frame” read:

A century of housing and land use policies denied Black households access to homeownership and neighborhood opportunities offered to white households. These racially discriminatory housing policies have combined to profoundly disadvantage Black households, with lasting, intergenerational impact. These intergenerational impacts go a long way toward explaining the racial disparities we see today in wealth, income and educational outcomes for Black Americans.

Those assigned to read about housing policy in a “class frame” were shown this:

Housing is the largest single expense for the average American, accounting for a third of their income. Many working-class, middle-class, and working poor Americans spend over half their pay on shelter. Twenty-one million American families — over a sixth of the United States — are considered cost-burdened, paying more for rent than they can afford. These families are paying so much in rent that they are considered at elevated risk of homelessness.

The “race and class group” read a version combining both race and class themes.

English and Kalla found that

While among Democrats both the class and the class plus race frames cause statistically significant increases in policy support, statistically indistinguishable from each other — among Republicans the class plus race frame causes a statistically significant decrease in policy support. While the race frame also has a negative effect among Republicans, it is not statistically significant.

Among independents — a key swing group both in elections and in determining the levels of support for public policies — English and Kalla found “positive effects from the class frame and negative effects from both the race and class plus race frames.”

A late February survey of 1,551 likely voters by Vox and Data for Progress produced similar results. Half the sample was asked whether it would support or oppose zoning for multiple-family housing based on the argument that

It’s a matter of racial justice. Single-family zoning requirements lock in America’s system of racial segregation, blocking Black Americans from pursuing economic opportunity and the American dream of homeownership.

The other half of the sample read that supporters of multiple-family zoning

say that this will drive economic growth as more people will be able to move to high opportunity regions with good jobs and will allow more Americans the opportunity to get affordable housing on their own, making it easier to start families.

The voters to whom the racial justice message was given were split, 44 in support, 43 in opposition, while those who were given the economic growth argument supported multiple-family zoning 47-36.

After being exposed to the economic growth message, Democrats were supportive 63-25, but less so after the racial justice message, 56-28. Republicans were opposed after hearing either message, but less so in the case of economic growth, 35-50, compared to racial justice 31-60.

López founded the Race-Class Narrative Project along with Anat Shenker-Osorio, a California-based communications consultant, and Heather McGhee, a former president of Demos, a liberal think tank and author of the recent book, “The Sum of Us: What Racism Costs Everyone and How We Can Prosper Together.”

I asked López about the English-Kalla paper. He was forthright in his emailed reply:

As my work and that of others demonstrates, the most potent political message today is one that foregrounds combating intentional divide-and-conquer racial politics by building a multiracial coalition among all racial groups. This frame performs more strongly than a class-only frame as well as a racial justice frame. It is also the sole liberal frame that consistently beats Republican dog whistling.

Shenker-Osorio faulted English and Kalla’s work for being “unsurprising”:

If you tell someone to support a policy because it will benefit a group they’re not part of, and that doesn’t work as well as telling them to support a policy they perceive will help them — this isn’t exactly shocking.

Testing the effectiveness of messages on controversial issues, Shenker-Osorio continued, has to be done in the context of dealing with the claims of the opposition:

Politics isn’t solitaire and so in order for our attempts to persuade conflicted voters to work, they must also act as a rebuttal to what these voters hear — incessantly — from our opposition. A class-only message about, say, minimum wage, held up against a drumbeat of “immigrants are taking your jobs” or racially-coded caricatures of who is in minimum wage jobs doesn’t cut through. Neither does a message about affordable housing credits or food stamps when the opposition will just keep hammering home the notion of “lazy people” wanting “handouts.”

Unless Democrats explicitly address race, Shenker-Osorio wrote, millions of whites, flooded with Republican messages demonizing minorities, will continue to be

primed to view government as taking from “hard working people” (coded as white) and handing it to “undeserving people” (coded as Black and brown). If we do not contend with this basic fact — and today’s unrelenting race baiting from the right — then Nixon’s “Southern Strategy” will simply continue to haunt us. In other words, if the left chooses to say nothing about race, the race conversation doesn’t simply end. The only thing voters hear about the topic are the lies the right peddles to keep us from joining together to demand true progressive solutions.

The Race-Class Narrative Project, which has conducted extensive surveys and focus groups, came to the conclusion that race could effectively be addressed in carefully worded messages.

For instance:

No matter where we come from or what our color, most of us work hard for our families. But today, certain politicians and their greedy lobbyists hurt everyone by handing kickbacks to the rich, defunding our schools, and threatening our seniors with cuts to Medicare and Social Security. Then they turn around and point the finger for our hard times at poor families, Black people, and new immigrants. We need to join together with people from all walks of life to fight for our future, just like we won better wages, safer workplaces, and civil rights in our past.

The race-class project also tested the efficacy of a class only message — “We need elected leaders who will reject the divide and conquer tactics of their opponents and put the interests of working people first” — versus a race and class message that simply added the phrase “whether we are white, Black or brown” to read:

We need elected leaders who will reject the divide and conquer tactics of their opponents and put the interests of working people first, whether we are white, Black or brown.

The result? The race and class message did substantially better than the class alone message among both base Democratic voters and persuadable voters. Base Democrats approved of the class message 79-16 and approved of the race and class message 86-11. Fewer persuadable voters approved of the class message than disapproved, 42-45, while more approved than disapproved of the race and class message, 48-41.

I asked Shenker-Osorio how well she thinks Biden is doing when he talks about race:

It’s definitely hit or miss. Sometimes he uses what I shorthand as “dependent clause” messaging where you name race after you’ve laid out an economic problem or offered an economic solution — e.g. “It’s getting harder for people to make ends meet, and this impacts [X Group] in particular.” This doesn’t work. For many people of color, this feels like race is an afterthought. For many whites, it feels like a non sequitur.

At other times, Shenker-Osorio continued, Biden

does what we’ve seen work: begin by naming a shared value with deliberate reference to race, describe the problem as one of deliberate division or racial scapegoating, and then close with how the policy he is pushing will mean better well-being or justice or freedom for all working people.

In partial support of the Shenker-Osorio critique, Jake Grumbach, a political scientist at the University of Washington, emailed me to say:

The English and Kalla survey experiment was done in a particular context that did not include Republican messaging, media coverage and imagery, and other content that “real-world” politics cannot escape. If Republicans use race, whether through dog whistles or more overt racism, then it might be the case that Democratic “class only” appeals will fall flat as voters infer racial content even when Democrats don’t mention race.

Another “important piece of context,” Grumbach wrote, “is that Biden is an older white man, which, research suggests, makes his policy appeals sound more moderate to voters than the actually more moderate Obama proposals.”

There is no consensus in the world of liberal advocacy on the most effective approach to racial issues.

Celinda Lake, the Democratic pollster who conducted much of the research for the Race-Class Narrative Project, was outspoken in her criticism of the English-Kalla paper, writing in an email: “There are huge flaws in their study and therefore in their conclusions. No candidate would run on what they put forward as the ‘race’ message.”

When I asked Kalla about these criticisms, he countered:

The messages that we tested did come from the real world of politics. Our messages came from actual politicians. As we note in the paper: “To improve the external validity of these findings, we adapted the frames from real-world political sources.”

In addition, Kalla noted that the paper does not say “that ‘we cannot talk about race.’ We do not make such a blanket claim.”

Micah English described the larger purpose of her work with Kalla in an email:

We try to make two contributions to this literature in political science. First, much of the prior literature has shown how Republicans’ use of race-based messaging has been effective at decreasing support for the welfare state and other social safety net policies. We are studying something different. We are studying Democrats’ use of race-based messaging to promote their policies.

And, she continued, “we are studying a different period in American politics. Many believe that in the past few years, and especially the past year, racial attitudes have shifted leftward.” But, she wrote, “our results show that, despite these ostensible leftward shifts in racial attitudes, voters do not seem to be very positively inclined to race-neutral policies when they are described with a racial justice lens.”

In the past, English wrote, scholars studied how Republicans used racial frames to “undermine support for redistributive policies, but now Democrats have started doing the same thing — with, according to our data, the same effects.”

Nicholas Valentino, a political scientist at the University of Michigan, said the English-Kalla study “comports with a long line of work in political psychology demonstrating a gap between widely a shared principle of racial equity and resistance to policies intended to achieve it.”

From the standpoint of rhetorical strategy, Valentino continued,

there is a trade-off between persuasion and mobilization. Highlighting racial injustice may mobilize nonwhite constituencies and racially progressive whites to engage in politics more forcefully.

That anger could be crucial in motivating voters “to overcome the obstacles to voting being pursued by the G.O.P. in many states,” Valentino noted. “The downside is that policy support for racial redistribution among moderates may decline.”

Martin Gilens, a political scientist at U.C.L.A., praised English and Kalla, but was quick to add caveats:

It’s a very nice paper and solid work. Their findings suggest that even in this time of heightened public concern with racial inequities, Democrats are not likely to boost public support for progressive policies by framing them as advancing racial equality.

That said, Gilens added, “I would consider the English and Kalla results to be sobering but not, in themselves, a strong argument for Democrats to turn away from appeals to racial justice.”

Elizabeth Suhay, a political scientist at American University, captured the complexity of the debate.

“English and Kalla’s findings are compelling,” she wrote by email:

Their findings are consonant with a great deal of conventional wisdom in political science. We would expect race-focused messaging to decrease support for a policy not only because of racism in the public, but also because many Americans perceive policies directed at specific population subgroups as unfair.

Suhay also noted: “Don’t forget self-interest. A longstanding definition of politics is that it’s a contest over ‘who gets what, where, when and how.’ ”

Suhay’s caveat:

Broad public approval is not the only thing politicians care about. From a strategic perspective, they must also be responsive to activists, interest groups, and donors. Given the intense focus on racial justice among some of the most active Democrats — including but not exclusively African Americans — Biden needs to not only deliver on this issue but also to tell people about it.

Suhay went on:

They face intense demands from Democratic activists for both policy and symbolic actions that address racial inequity; however, these actions do threaten to turn off many whites, especially those without a college degree.

Biden, Suhay argues, “seems to have no choice but to find some middle road: focusing communication on how his policies benefit most Americans while also, more infrequently but unmistakably, making clear his commitment to racial equality” and, she added, “he seems to be walking the tightrope well.”

This debate is not one that lends itself to resolution. Biden and his Democratic colleagues are cross-pressured in ways that date back at least to the early 1960s, in the run-up to the enactment of the 1964 Civil Rights Act and the 1965 Voting Rights Act.

At the same time, the terrain undergoes constant change. Hispanics have eclipsed African-Americans as the nation’s largest minority, and there are signs of a slight but significant shift to the right among both groups. White Democrats have moved sharply to the left, but it’s not clear if that shift reflects an enduring commitment, a temporary response to the Trump phenomenon, or a reflection of social desirability bias. At the same time, generally speaking, residents of red states continue to fall behind as economic growth remains concentrated in deep blue regions, further adding to the volatility of voting and elections.

All of these trends, along with others I have written about in other columns, have an impact on the dynamics of race, which remain at the core of American politics. In these circumstances, the ideal form of Democratic messaging is a moving target, changing shape and substance across geography and time.

The race-versus-class debate within the Democratic Party is, perhaps, inevitable for a party with roots in the politics of the New Deal, politics that have evolved in recent decades from a preoccupation with social stratification — class — into a focus on the more specific rights of the once-marginalized: racial and ethnic minorities, women and the LGBT community.

A granular overview of racial attitudes emerges from looking at the instability of the public opinion trend line on the Black Lives Matter movement, as documented by the polling firm Civiqs.

In December 2019, according to Civiqs data, more voters, 42 percent, supported BLM than opposed it, 33 percent. Immediately after the murder of George Floyd on May 25, 2020, support shot up to 59 percent and opposition dropped to 29 percent. Since then, the difference between support and opposition has drifted back to the earlier levels, with the most recent measure on April 25 at 48 in support of BLM, with 40 opposed 40.

Last year, two psychologists, Steven O. Roberts of Stanford and Michael T. Rizzo of N.Y.U., published a trenchant paper, “The Psychology of American Racism,” declaring that “American racism is alive and well.” They based their claim on

a large body of classic and contemporary research across multiple areas of psychology (e.g., cognitive, developmental, social), as well as the broader social sciences (e.g., sociology, communication studies, public policy), and humanities (e.g., critical race studies, history, philosophy).

In fact, there is a much more accessible source to make this case: the 2020 election results. Donald Trump, running explicitly as the candidate of white America, won 74,216,154 votes. The fact that Trump was narrowly defeated in the Electoral College remains the salient point, but the burden falls on the Democratic Party to keep Trump and the Republican Party he continues to lord over from regaining the White House and Congress.

The Times is committed to publishing a diversity of letters to the editor. We’d like to hear what you think about this or any of our articles. Here are some tips. And here’s our email: letters@nytimes.com.

Follow The New York Times Opinion section on Facebook, Twitter (@NYTopinion) and Instagram.