Marketplace Lending, Information Aggregation and Liquidity

We address a puzzle whereby lending marketplaces, aimed at directly connecting retail lenders and borrowers, retreat from auctions and take on the role of price setting and credit allocation, despite evidence that retail investors possess valuable soft and nonstandard information. Our analysis uses a unique data set on 7,455 auctions and 34 million bids, from the leading British peer-to-business platform. We find that the main problem of the platform was its vulnerability to liquidity shocks, resulting in sizable deviations from information eciency. These increased over time due to a growing role played by non-crowd players, particularly large investors and algorithms.