Market-making with Search and Information Frictions∗

We develop a dynamic model of trading through market-makers that incorporates two canonical sources of illiquidity: trading (or search) frictions and asymmetric information. We show that conventional predictions, derived from models that study either of these frictions in isolation, do not necessarily hold when both frictions are present. These results have important implications for technological innovations and regulatory initiatives that aim to reduce trading and/or information frictions. A key result is that reducing trading frictions slows down the rate at which market-makers learn about asset quality, amplifying the effects of adverse selection and ultimately leading to more illiquidity, as measured by, e.g., bid-ask spreads.