I develop a search‐and‐bargaining model of endogenous intermediation in over‐the‐counter markets. Unlike the existing work, my model allows for rich investor heterogeneity in three simultaneous dimensions: preferences, inventories, and meeting rates. By comparing trading‐volume patterns that arise in my model and are observed in practice, I argue that the heterogeneity in meeting rates is the main driver of intermediation patterns. I find that investors with higher meeting rates (i.e., fast investors) are less averse to holding inventories and more attracted to cash earnings, which makes the model corroborate a number of stylized facts that do not emerge from existing models: (i) fast investors provide intermediation by charging aspeed premium, and (ii) fast investors hold more extreme inventories. Then, I use the model to study the effect of trading frictions on the supply and price of liquidity. On social welfare, I show that the interaction of meeting rate heterogeneity with optimal inventory management makes the equilibrium inefficient. I provide a financial transaction tax/subsidy scheme that corrects this inefficiency, in which fast investors cross‐subsidize slow investors.

UK government bond yields tend to drift upwards before scheduled news such as monetary policy announcements and labour market data releases. This effect is particularly pronounced during periods of UK bond issuance and is linked to higher term premia. Financial intermediary constraints play a role as dealers avoid accumulating inventory in pre-news windows following issuance. The composition of liquidity providers also shifts: hedge funds buy a large share of the bond issuance outside pre-news windows, but more passive investors – such as foreign central banks and pension funds – provide liquidity in pre-news windows. We outline a simple model to rationalize these findings.

Should regulators encourage the migration of trade from over-the-counter (OTC) to centralized markets? To address this question, we study a model in which banks make costly decisions to participate in an OTC market, a centralized market, or both markets at the same time. Banks differ in their ability to take large positions, what we call their trading capacity. In equilibrium, intermediate-capacity banks find it optimal to participate in the centralized market. In contrast, low- and high-capacity banks find it optimal to participate in the OTC market, due to an endogenous complementarity. Namely, low-capacity banks receive worse terms of trade than in the centralized market but better risk sharing, thanks to the intermediation services offered by high-capacity banks. High-capacity banks receive worse risk sharing than in the centralized market, but profit from the provision of intermediation services to low-capacity banks. While the social optimum has qualitatively similar participation patterns, it prescribes that more customers migrate to the centralized market, and that more dealers enter the OTC market.