This paper empirically investigates the causal impact of avatar gender and race on transaction prices within the Non-Fungible Token (NFT) marketplace. The impetus behind this investigation is grounded in the observation that NFT investors frequently represent a younger, wealthier, and politically progressive demographic, which raises the question of whether this marketplace is less susceptible to prejudice and social bias. Contrary to such expectations, our analysis reveals pronounced gender and racial disparities in NFT prices. Specifically, female CryptoPunks avatars are transacted at a 36.8% lower price compared to their male counterparts (i.e., avatars with similar image attributes), while black CryptoPunks avatars are transacted at a 30.7% lower price compared to their white counterparts. We further identify that avatar features linked to high-tech or higher education attributes (e.g., 3D glasses, Virtual Reality glasses, Nerd glasses) can effectively counter these disparities. We verify the causal patterns revealed in the NFT dataset with a controlled experiment, where NFT buyers are randomly shown the comparable avatars that exhibit variations in race or gender, enabling us to evaluate the impact of these attributes on buyer choices and willingness-to-pay. The experiment confirms the gender (male) and race (white) price premium in the NFT market and suggests that buyers exhibit a preference for NFT avatars resembling their personal identity. We offer insights into policy strategies aimed at promoting racial and gender equity within the NFT marketplace.
Read full abstract