From the time of the Constitution’s ratification until Joe Biden’s junior year of college, the United States was governed by a de-jure racial caste system that denied most Black Americans political power and economic opportunity. For this reason, America’s working class is much less white than its rich. Which means that any economic policy that takes resources away from America’s economic elite — and gives them to its laboring masses — will also, on net, transfer resources from white Americans to Black ones.

Historically, Republicans have taken pains to emphasize this fact, while Democrats have attempted to downplay it. The logic of these tactics was straightforward: The U.S. electorate is both overwhelmingly white and overwhelmingly non-rich. If non-rich white voters believe that redistributive policies take money from white people like them and give it racial others, then it will be easier for the anti-redistribution party to prevail in elections; if such voters believe that redistributive policies take money from the rich and give it to working people like them, then the pro-redistribution party will have the upper hand. Thus, in 2009, Glenn Beck derided the Affordable Care Act as “the beginning of reparations,” while Barack Obama insisted it was a means of providing health care to “middle-class Americans.”

In recent years, however, some Democrats have come to emphasize — and, occasionally, even exaggerate — the racial implications of race-neutral, redistributive policies. In the early days of the 2020 primary, multiple Democratic candidates attempted to rebrand their ideas for delivering affordable housing, baby bonds, and middle-class tax cuts as forms of “reparations.” More commonly, Democrats will make a point of spotlighting the racial-justice implications of progressive economic reforms, from universal health care to the minimum wage.

This development interested Yale political scientists Micah English and Joshua Kalla. A large body of poli-sci research had affirmed the wisdom of the major parties’ traditional messaging strategies: emphasizing that universal programs disproportionately benefit racial minorities decreases support for those programs. But perhaps times had changed. After all, the Black Lives Matter movement had mounted the most prominent challenge to racial inequality in the U.S. in a generation, and the racial attitudes of white Democrats had liberalized in response.

To test this possibility, English and Kalla took a set of race-neutral economic policies — a $15 minimum wage, student-debt forgiveness, zoning reform, the Green New Deal, and Medicare for All — and then presented them to voters using three different arguments: one that emphasized the policies’ class implications, one that emphasized their impact on racial justice, and one that emphasized both. To enhance the validity of their experiment, they used Democratic politicians’ actual rhetoric for each of these arguments.

Their findings confirmed conventional wisdom: The class message tended to modestly increase support for progressive policies, while arguments involving race slightly reduced support for them, though not by a statistically significant margin.