By Shom Mazumder (@shom_mazumder)
These days protests seem like the new normal. One particular set of protests that has caught the public’s attention is the Black Lives Matter (BLM) movement, which seeks to dismantle institutional racism and state violence against people of color.
Did BLM protests lead to more racial liberalization in the United States or did it lead to a backlash against African Americans?
Using the release of the 2018 round of the Cooperative Congressional Election Study (CCES) combined with data on locations of BLM protests in the wake of the police killings of Eric Garner and Michael Brown--both unarmed, black civilians--I show that BLM protests did meaningfully reduce whites’ racial prejudice against African Americans.
To do so, I compare whites from counties that experienced protests before and after 2014 (when the large wave of BLM protests in solidarity with Ferguson and New York began) to whites from counties that did not experience protests over this time period. Prior to these protests, whites seemed to have followed the same trend. After 2014, we can see that whites from protest counties reduced their level of racial resentment toward blacks more than what we would have expected given the trend.
In additional analyses, I find that these results aren’t driven by cities or prior factors associated with the growth of the BLM movement.
I also find that this racial liberalization is driven by young whites. Older whites (greater than 50) on the other hand tend to become slightly more racially resentful in response to BLM protests.
Did these protests also matter for elections?
Not only did BLM protests change hearts, but they also changed minds at the voting booth. Counties that experienced protests experienced less of decline in Democratic Party presidential vote shares relative to non-protest counties between 2012 and 2016. My estimates indicate that Democrats received about a 4-6 percentage point boost in response to BLM protests.
Overall, the evidence seems to be clear: black protests matter.