Measuring NFT creators’ contributions to market price and liquidity

Abstract

This article studies the relationship between creator-related cues and market outcomes – price, time to sale, and non-sale – of non-fungible tokens (NFTs) in a leading marketplace. We first extract textual and visual indicators and summarize them into cognitive and affective composites using principal component analysis. We then estimate hedonic regression and duration models with creator-level random effects and recover creator-related components using empirical Bayes shrinkage. These components provide a descriptive decomposition of market outcomes into variation linked to observable asset cues and residual variation systematically associated with creators. We find substantial heterogeneity in creator-related components for both price and liquidity, while simple social-media metrics account for only a small share of that heterogeneity. We also model non-sale probability and show that creators’ social media activity is modestly associated with sale failure. Methodologically, the paper offers a transparent approach to mapping creator-related heterogeneity when creator metadata and standard brand-equity measures are limited.