How Support for Black Lives Matter Impacts Consumer Responses on Social Media

We scrutinize the direct and moderated impact of brands’ support for Black Lives Matter (BLM) on consumer responses. Our empirical strategy exploits Blackout Tuesday as a natural experiment in which BLM support occurred on Instagram (treated platform) but not on Twitter (control platform) to perform a within-brand cross-platform difference-in-differences (DID) analysis. We also combine econometric models with machine learning techniques to analyze the unstructured data of the social media content. Based on a unique multi-industry, multi-year, and multi-platform dataset of 435 major brands and 396,988 social media posts, we find a negative impact of BLM support on consumer responses such as followers and likes. Furthermore, our analyses uncover a multifaceted set of heterogeneous DID effects across brands: 1) While lone-wolf BLM support leads to negligible effects, large-scale BLM support from many brands can lead to strong negative effects, i.e., the bandwagon effect. 2) Posting self-promotional content exacerbates the negative effects of BLM support. 3) Historical prosocial posting on social media attenuates the negative effects. 4) Brands with socially-oriented missions suffer less from the negative effects. 5) Customers’ political affiliation also matters: the negative effects of BLM support are amplified/attenuated for brands with mostly Republican/Democratic customers. And 6) slacktivism (showing BLM support in words but without financial donations) can mitigate the negative effects for brands with mostly Republican consumers but amplify the negative effects for brands with mostly Democratic consumers