| Abstract: | People rapidly and subconsciously process information from facial images. On sharing economy platforms, facial cues can provide a useful supplement to other information provided by reputation systems. Previous small-scale, rater-informed studies examining trust and attractiveness based on facial features on Airbnb found mixed support for impacts on pricing. We re-examine their impact using deep learning to classify host faces for an extensive data set of Airbnb accommodation in 10 US cities (n = 78,215). Together, trust and attractiveness contribute to almost a 5% increase in prices for Airbnb accommodation. We also test Gray's theory of motivation via the examination of pricing for different types of accommodation, finding that trust is more important in situations of smaller accommodation shared with strangers. The paper concludes with limitations and implications for research and practice. |
