Research suggests that the resurgence of drug-induced homicide (DIH) law popularity in the United States is a means of differentially targeting people of color for drug enforcement in a context where it is increasingly unpopular to criminalize people who use drugs. Using a survey experiment on a nationally representative sample of US adults varying the race/ethnicity of a male PWSD in a mock press release (White, Black, Latino, unnamed race), we examined the relationship between racialized narratives of PWSD and support for DIH laws. Respondents rated the extent to which they support the law, the extent to which the PWSD should receive extra jail/prison time, whether they expect the law to deter drug sale, and whether they believe the law is morally justified. We used linear probability and ordered logit models to identify treatment effects while controlling for demographic factors like gender, education, and political party affiliation.

Compared to respondents shown any other PWSD, those shown a Latino PWSD were more likely to support the law (Latino-unnamed b=0.042, p<0.10; Latino-white b=0.053, p < 0.05; Latino-Black b=0.051, p<0.05) and to believe that the law is morally justified (Latino-unnamed race 0.038, p<0.10; Latino-white 0.048, p<0.05, Latino-Black 0.051, p<0.05), but not to believe that it will deter sale. They also supported higher levels of punishment (ordered logit: Latino-unnamed OR=1.18, p<0.10; Latino-white OR=1.33, p<0.05; Latino-Black OR=1.28, p<0.05). Among white, non-Hispanic respondents, we likewise saw positive differences in law support (Latino-White b=0.046, p<0.10, Latino-Black b=0.055, p<0.05), belief that the law is morally justified (Latino-Black 0.044, p<0.10), and punishment level (ordered logit: Latino-White OR=1.31, p<0.05; Latino-Black OR=1.22, p <0.10). Among Hispanic respondents, negative differences were detected in law support (Latino-Black b=-0.236, p<0.10), belief that the law will deter sale (Latino-unnamed b=-0.348, p<0.01), belief that it is morally justified (Latino-unnamed b=-0.224, p<0.10; Latino-Black b=-0.231, p<0.10), and punishment level (ordered logit Latino-unnamed OR = 0.388, p<0.10). This suggests that findings were likely driven by white, non-Hispanic respondents, as Black non-Hispanic respondents showed few statistically significant differences, and Hispanic respondents showed opposite-signed results.