比较不同市场中的搜索摩擦和中介摩擦
Comparing search andintermediation frictionsacross markets by Gabor Pinter, Semih Üslü and Jean-Charles Wijnandts Monetary and Economic Department August2025 JEL classification: D40, G10, G11, L10Keywords: search frictions, market power, governmentbonds, corporate bonds, over-the-counter markets BISWorking Papers are written by members of the Monetary and EconomicDepartment of the Bank for International Settlements, and from time to time by othereconomists, and are published by the Bank. The papers are on subjects of topicalinterest and are technical in character. The views expressed in them are those of theirauthors and not necessarily the views of the BIS. This publication is available on the BIS website (www.bis.org). ©Bank for International Settlements 2025. All rights reserved. Brief excerpts may bereproduced or translated provided the source is stated. Comparing Search and Intermediation FrictionsAcross Markets∗ Gábor Pintér†Semih Üslü‡Jean-Charles Wijnandts§ Abstract We develop a two-asset search-and-bargaining model of OTC trading to estimate frictionsand welfare losses in the UK government and corporate bond markets.Using transaction-level data and a matched client sample, we find that both trading delays and intermediationfrictions are more pronounced in corporate bonds. Welfare losses due to these frictions are2.4% in government bonds and 5.0% in corporate bonds—driven primarily by trading delays.Using data from the COVID-19 crisis, we find that these losses might more than doubleduring turbulent times, revealing the fragility of the OTC market structure. JEL Classification:D40, G10, G11, L10Keywords: Search Frictions, Market Power, Government Bonds, Corporate Bonds, Over-the-Counter Markets “... liquidity—the ability to trade when and where you want to trade,in significant size and without much cost.” —Larry Harris, Trading & Electronic Markets:What Investment Professionals Need to Know (2015, p. 1) 1Introduction How large are trading frictions in over-the-counter (OTC) financial markets? How much do thesefrictions and the associated welfare losses vary across markets and across normal and turbulenttimes? What is the role of search frictions facing clients and of intermediaries’ market power inexplaining these cross-market differences? To provide quantitative answers to these questions, weestimate a novel dynamic structural model of OTC market liquidity for the corporate as well asgovernment bond markets—two of the most important markets in the financial system. We find that search frictions in the UK government bond market are quantitatively modest,with the estimated welfare loss amounting to 2.38% relative to a frictionless benchmark.Incontrast, search frictions in the UK corporate bond market are more pronounced, generating awelfare loss of 4.95%. Notably, the majority of welfare losses in both markets stem from searchdelays, with intermediation frictions contributing negligibly to the welfare loss in the governmentbond market and accounting for only 0.10% in the corporate bond market. In total, the welfareimpact of OTC trading frictions is 2.38% in the government bond market and 5.05% in thecorporate bond market under normal market conditions. When we re-estimate the model usingdata from the COVID-19 crisis, the welfare losses rise sharply to 3.63% in the government bondmarket and 11.35% in the corporate bond market.1 This increase—particularly in the role ofintermediation frictions in the corporate bond market—highlights the fragility of the OTC marketstructure in times of stress. To arrive at these estimates, our paper makes a contribution to both the theoretical and empir-ical literatures on OTC markets. Our empirical analysis uses a non-anonymous transaction-leveldataset which covers close to the universe of all secondary market trades in the UK governmentand corporate bond markets. Importantly, we are able to identify a common set of clients who areactively trading in both markets and who drive the majority of the trading volume in the client-dealer segment of both markets. This unique feature of the data allows us to exploitcross-marketdifferences in all three main dimensions of market liquidity:trading frequency, trade size, andtrade cost. Comparing trading frictions across markets is a hard task because client compositionis endogenous to the given market in question. Our approach of keeping the set of clients fixed inthe two markets goes a long way in addressing these selection issues, allowing for a comparison of the trading frictions that thesame clientsface in two different markets. The theoretical contribution of the paper is to develop a structural model of OTC trading withtwo-sided search and bilateral bargaining that can be estimated using transaction-level data withclient identities.Building on the stationary version of Lagos and Rocheteau (2009), we extendthe framework along three key dimensions to better capture the realities of fixed-income markets.Fir
