Discussion

Carol Osler wondered about the limits of the benefits of fragmentation involving uninformed and informed traders. She suggested that, if one were to take Blume's argument to the logical extreme, one might think that there would be only uninformed traders in a single market.

Bob Litan asked about the direction in which industry structure worldwide is headed: will there be consolidation, with perhaps two exchanges in the United States, one in Europe, one in Japan, and one or two in Asia, or will there be a proliferation of various kinds of exchanges and constant churn in the market? One participant responded that firms like the New York Stock Exchange and Nasdaq have an entrenched position and are very good leaders: they will co-opt a lot of the fragmentation and buy out any market that is successful. Therefore, there is little likelihood of fragmentation with, say, ten markets each having 10 percent; rather, there probably will be a lot of markets with 1 percent of shares and a few markets with very large shares, such as is the case with the New York Stock Exchange right now.

Blume stated that, in the past, Paul Mahoney questioned the inherent contradiction in the U.S. system of securities regulation. He asserted that the mandatory disclosure system seems to be based on the proposition that retail investors ought to trade based on firm-specific information. However, regulation of market structure seems to be based on the proposition that it should be very inexpensivefor retail investors to trade on the basis of that information. Blume responded that the goal of securities regulation is to create an even playing field. However, since a very small percentage of retail customers trades a lot, a very small segment of people who probably are churning their accounts so much that they are going to be losing money is being protected. Nonetheless, it would be politically unacceptable to forbid retail traders from investing directly.

Dan Tarullo questioned whether the different market structures would have a significant impact on the efficiency of capital flows in capital markets, given that the principal issue is pricing and who gets the best price. Blume responded that he does not believe that a best price is definable. The issues confronting regulators in the future are global and systemic in nature, like prevention of fraud on an international basis, and that is where the Securities and Exchange Commission should be putting its efforts.

On a separate subject, Jim Cochrane pointed to the high costs, in the wake of the events of September 11, of maintaining simultaneous "hot sites" in different urban population centers on different power grids. He added that many companies, especially in lower Manhattan, already are facing insurance bills that probably will take them out of business. Nonetheless, Cochrane noted that the New York Stock Exchange is beginning a process of massive global repositioning, reorganization, and redeployment of its business to take account of the kind of contingencies that September 11 made all too evident.

Bob Litan commented that if Cochrane was right about redundancy cost being built into the system, then that has an implication for market structure. Since smaller exchanges cannot afford redundancy costs, and traders want to go to places where there is redundancy, concentration will increase. He wanted to know a ballpark figure for the costs of building redundancy in systems.

Benn Steil answered that the cost of redundancy on exchanges differs dramatically depending on the market structure. He stated that the estimated cost of building a backup flow for a system like the New York Stock Exchange would be about $150 million, but for a system like Island, which is a continuous, electronic order-matching system based on personal computers, the cost would be only a few million dollars. He added that the Vertex Exchange backup system for London and Zurich cost a few million dollars, nowhere close to $150 million. Cochrane agreed that it is very cheap, as countries around the world have found, to build...