Exchange Systems and International Comovement of Return and Liquidity

There is substantial evidence of commonality or comovement in the liquidity, order flow and returns of different assets. Studies have shown that such comovement cannot be completely explained by comovement in fundamental characteristics. These studies focus on the role of frictions in explaining comovement unrelated to fundamentals. In this study we propose to examine the role played by trading technology, specifically the practice of market operators selling their trading technology to other exchanges.

Over the last few years, algorithmic trading has become a dominant feature of trading in several asset markets such as equities and FX. Institutional investors and vendors of trading algorithms face high fixed costs to develop such algorithms for a particular trading platform. On the other hand, it is comparatively inexpensive to deploy such algorithms in multiple jurisdictions if exchanges use the same trading software. If the same investors use the same algorithms on different trading venues, this may lead to a high level of co-movement in returns, liquidity, and order flow, even in the absence of an economically fundamental reason.

Trading systems of dominant market operators have been increasingly employed by other exchanges in recent years. For example, NASDAQ’s INET technology is used in the Nordic and Baltic markets, and Deutsche Börse's Xetra system is used in Ireland, Austria, and Shanghai. We plan to exploit such changes to answer the question whether the use of a common trading platform leads to an increase in excess comovement. Our study is econometrically well identified because changes in trading systems clearly are not motivated by an anticipation of an increase in comovement. Moreover, exchanges without a change in trading system can serve as controls.

Comovement is economically important to investors and issuers because an increase in comovement unrelated to fundamental values diminishes the benefits of diversification, enhances investors’ systematic risk, and leads to an increase in the firms’ cost of capital. The literature has thus far ignored the role trading venues themselves may play in explaining excess co-movement. This is particularly important in the context of an increasing use of the same trading systems in different international markets.

In order to obtain measures of liquidity and order flow in international equity markets, we plan to rely on high-frequency data from Thomson Reuters Tick History.