Bargaining and Fighting in Hard Times*

Those inside and, especially, those controlling an organization — be it a firm, a committee, a cabinet ministry, or the state — often have better information about the organization’s resources and activities than outsiders do. This informational asymmetry may lead to costly, inefficient conflict between insiders and outsiders over the allocation of those resources. We formalize this conflict as a signaling game in which an incumbent government and an opposing faction — possibly a rebel group — vie for control of the state and the accompanying spoils. In order to avoid a challenge, the government must buy off or coopt the opposition by offering a share of the pie. But the size of the pie is private information. The government knows how large the spoils are but the opposition only has a rough idea about the spoils (e.g., oil prices are high or low, the economy is booming or in crisis). The unique perfect Bayesian equilibrium satisfying a common restriction on off-the-equilibrium-path beliefs (Cho and Kreps’ D1) is fully separating but inefficient with the probability of breakdown increasing as times become harder. We generalize the government-opposition game to a larger class of “coercive” signaling games which exhibit the same equilibrium behavior and include models of war and litigation. * For helpful comments, criticisms and discussion, we thank... † Haas School of Business, UC Berkeley, [email protected] ‡ Department of Political Science, UC Berkeley, [email protected] Bargaining and Fighting in Hard or Uncertain Times [A note to the reader: This is a very preliminary and rough paper pulled together while in the midst of a major revision. The body of the paper sketches many of the main arguments and central points and, we hope, is clear enough to be worth reading. The appendix is still “under construction.”] The more formal appendix Those inside and, especially, those controlling an organization — be it a firm, a committee, a cabinet ministry, or the state — often have better information about the organization’s resources and activities than outsiders do. This informational asymmetry may lead to costly, inefficient conflict between insiders and outsiders over the allocation of those resources. In resource-rich, developing countries, for example, the government frequently has private information about the revenues those resources bring. This lack of transparency is believed to facilitate corruption, make conflict more likely, and has led to international efforts to promote greater transparency, e.g., the Extractive Industries Transparency Initiative (Economist 2005, DFID a, nd.; DFID b, nd). We study the effects of this asymmetry in a model in which an incumbent government and an opposition faction vie for control of the state and the spoils that come with it. The government knows how large the “pie” is but the opposition only has a rough idea of the size of the pie. The opposition knows, for example, whether times are “good” or “bad” (e.g., oil prices are high or low, the economy is booming or in recession) and therefore whether the pie is on average large or small. But the opposing faction is unsure precisely how large the spoils are in good times or how small they are in bad times. This informational asymmetry casts a shadow over the government’s efforts to buy off or coopt the opposing faction and thereby prevent a costly conflict — a civil war — over the spoils. The government in the model makes a take-it-or-leave-offer which the opposition can accept or reject by fighting. Accepting ends the game in the agreed division of the spoils. Fighting destroys a fraction of pie with the winner getting what remains. The opposition faces a simple but vexing strategic problem. If the opposing faction