Research Proposal Shorts IV: Equities Markets and War

Note: Fourth in an occasional series of short research proposals on various topics in international relations.

Fearon (1995, "Rationalist Explanations for War," International Organization 49:379-414) presents a number of interesting arguments to support the rationalist explanation for why state leaders initiate wars. While providing evidence from formal modeling and, to a lesser extent, case studies, he provides a number of logically consistent reasons for this phenomenon. Nonetheless, further information can likely be gleaned from empirical analyses of available data.

The component of interest for this proposed research is the method of signaling. Here, the proposal departs from the rational unitary actor model in proposing that military signaling can also be accomplished through by observing the actions of non-state actors, specifically, the activities of equities markets. In their most elemental sense, equities markets serve as a purveyor of almost perfect information because they force participants to reveal their true preferences for prices of the equities being traded. Recently, this phenomenon has been expanded to the more general concept of markets which trade, not equities, but information which similarly reveals the hidden information of participants.

Observations of the predictive power of information markets has been documented in popular works such as The Wisdom of Crowds (Surowiecki 2004) and is only recently entering the realm of academic literature through journals such as the Journal of Prediction Markets, which began in 2007. Economists such as Manski (2005, "Interpreting the Predictions of Prediction Markets. NBER Working Paper No. W10359) and Gjerstad (2004, "Market Dynamics in Edgeworth Exchange," Eller College Working Paper No. 04-03) have analyzed the mechanisms at work in information markets and have found mixed results. Nonetheless, the relative newness of this field makes it ripe for application to theoretical explanations for war.

This research proposes combining readily available stock market data with data from the Inter- and Extra-State War data available from the Correlates of War project. Although market data is available for trading floors worldwide, this research will focus only on US equities markets, due to their size and breadth of trade. The New York Stock Exchange, for example, is the world’s largest stock exchange in the world by dollar volume. Both the Inter- and Extra-State War data are chosen because they track wars along multiple dimensions of interest, including not only the duration and intensity of war, but also the presence of outside intervention and non-state actors.

The hypothesis of this proposed research is that stock markets will exhibit fluctuations in equities with interests, such as natural resource pipelines or major manufacturing facilities, in the regions or states which eventually go to war. Stock market data would therefore be lagged by a certain time period of six months against Correlates of War data. Theory does not suggest whether the sign would be positive or negative, as equities traders could either fear future losses or anticipate gains from conflict. Nonetheless, the vehicle of equities markets could serve as another signal for potential conflict for states to consider in their expected value calculations.