Adverse Selection and Competitive Market Making: Empirical Evidence from a Limit Order Market

This article presents a new methodology for testing economicrestrictions on the price schedules offered in a limit orderbook that are based on (i) break-even conditions for marginallimit orders and (ii) rational updating conditions for orderbook revisions over time. Using order flow data from the StockholmStock Exchange, I find strong evidence of insufficient depthin the limit order books relative to the theoretical predictions.An extended model, which allows the model parameters to dependon market conditions, captures some of the systematic variationin the observed order book depth.