Network Replication of Inequality in Medical Crowdsourced Funding

Abstract

In 2018, over 250,000 American families found themselves unable to pay for medical care and turned to the online “crowdfunding” service GoFundMe to raise money online. The $650 million dollars raised from these medical campaigns appear to have filled a sizable hole in the American social safety net. Yet crowdfunding is at heart a network process, and a large body of research shows that social networks can reproduce inequality. In this paper I show that medical crowdfunding replicates patterns of racial, ethnic, and geographic income stratification in ways that are consistent with network theory. Using 2,618 GoFundMe campaigns hand-coded for perceived race and ethnicity of the recipient, I show that Black and Hispanic beneficiaries receive substantially less money via their networks than White and Asian beneficiaries. Hierarchical linear models show that social network access via online sharing does not vary by race and ethnicity. However, network mobilization, measured in terms of the number and size of donations, varies substantially and produces unequal returns to campaigns. Variations in the number of donations can largely be explained by differences in estimated network financial capacity, but variations in donation size are not fully accounted for even in models including proxies for network income. Estimates of donor race and ethnicity indicate that donors of all races and ethnicities tend to give White recipients the largest donations and Black recipients the smallest. Overall, I demonstrate that the use of “crowd insurance” in place of sufficient medical insurance reproduces existing patterns of inequality.