Information technology and securities markets: Feasibility and desirability of alternative electronic trading systems

Valuing the benefits and forecasting the impact of strategic systems is difficult. Uncertainties include the likelihood and the pace of adoption, and the magnitude of the actual advantages to be realized through automation. The thesis examines the possibility that today\u27s securities markets will be displaced by lower-cost electronic trading systems, and demonstrates two experimental methodologies for evaluating the prospects for an important strategic system. Information technology (IT) currently exists that could support global 24-hour trading via linked trader workstations. Yet for an electronic trading system to form a liquid and widely used market, traders would need to make a coordinated transition to this alternative way of trading. Furthermore, electronically-mediated markets represent a vastly different market architecture from that currently in place. The changed design may not be favorable under all trading conditions. Using two methodologies--experimental economics and computer simulation--we test necessary conditions for an automated trading system to become adopted. Traders in a series of experiments demonstrate an ability to make a transition to a more attractive market. The simulation results indicate that the role of designated market maker (such as the New York Stock Exchange specialist), which is generally eliminated in electronic trading systems, provides significant market quality benefits. The thesis calls into question the maintained belief that the dominance and historic liquidity advantage of established exchanges protects those markets from the competitive threat of potentially superior electronic trading alternatives. Furthermore the thesis challenges the notion that screen-based electronic markets with no formal dealer-intermediary roles provide a superior market mechanism to the specialist-auction market in place on the floor-based exchanges in the U.S. The results yield policy implications for regulators, new market designers, and exchange officials. Principally, improved designs for trading systems are needed, and regulatory intervention regarding market design choices is not called for; when sufficiently attractive trading systems are available, they will be capable of attracting trading volume away from established exchanges. The results enable a sound forecast to be made of a strategic IT innovation, and help to explain the durability of manual trading practices