One Security, Many Markets: Determining the Contributions to Price Discovery

When homogeneous or closely-linked securities trade in multiple markets, it is often of interest to determine where price discovery (the incorporation of new information) occurs. This article suggests an econometric approach based on an implicit unobservable efficient price common to all markets. The information share associated with a particular market is defined as the proportional contribution of that market's innovations to the innovation in the common efficient price. Applied to quotes for the thirty Dow stocks, the technique suggests that the preponderance of the price discovery takes place at the New York Stock Exchange (NYSE) (a median 92.7 percent information share). ALTHOUGH MOST OF THE CLASSIC paradigms in market microstructure concern a security that trades in a single centralized market, such situations are becoming increasingly rare in practice. Fragmentation, the dispersal of trading in a security to multiple sites, has emerged as a dominant institutional trend. This process is of concern to financial economists and regulators because price information and price discovery (the impounding of new information into the security price), arguably the most important products of a security market, have many attributes of public goods. As the process of fragmentation accelerates, it may be important to determine where the price information and price discovery are being produced.1 This paper suggests a practical econometric approach to this problem. The technique is illustrated with an application to U.S. equity markets. Many stocks listed on the New York Stock Exchange (NYSE) trade concurrently on the regional exchanges and the National Association of Securities Dealers' National Market System (NASD NMS). Since a share of IBM stock is the same security whether purchased on the Midwest or Pacific Exchange, this is a particularly clear instance of multiple markets. Nevertheless, the expres* Stern School of Business, New York University. For comments on an earlier draft, I am indebted to the editor, the anonymous referees, Gail Belonsky, Frederick H. deB. Harris, Gautam Kaul, A. Craig MacKinlay, Ananth Madhavan, Terry Marsh, Maureen O'Hara, and presentation audiences at Duke University, University of Iowa, University of Missouri, Washington University, and the American Finance Association. The author is a former Visiting Academic Economist at the New York Stock Exchange. All errors are my own responsibility. 1 Economic and institutional issues related to fragmentation are discussed by Cohen, et al. (1982), Cohen, Conroy, and Maier (1985), Mendelson (1987) and Harris (1993). For current institutional and regulatory perspectives, see Shapiro (1993) and the U.S. Securities and Exchange Commission's Market 2000 Study (U.S. SEC (1994)).