WikiSummary, the Social Science Summary Database

Keohane: After hegemony

From WikiSummary, the Free Social Science Summary Database

This summary needs formatting (i.e. "wikification"). Can you help us improve it? (Formatting help.) Please volunteer.

Keohane. 1984. After hegemony: Cooperation and discord in the world political economy. Princeton: Princeton University Press.



Realists have argued that rational choice leads to either hegemony or conflict. Institutionalists have sought to explain cooperation based on idealism, or something less than rationalism. Keohane seeks to demonstrate that rational choice (realists' premises) predicts cooperation and the establishment of institutions.

Examples of how this works:

  • The prisoner's dilemma is the paradigmatic example of why cooperation should break down. But in reality, states are playing iterated PD (b/c they are negotiating on several issues at once, and they cooperate repeatedly over time). Iteration makes cooperation not only possible, but likely under certain circumstances (i.e. if there are some states playing tit-for-tat).
  • Olson argued that his collective action problems would not be severe when there is only a smal number of parties that must cooperate. In international relations, this is almost always true.
  • Olson further argued that selective incentives matter. Keohane points out that most international organizations provide club goods, not public goods.


Rational choice premises tempt us to make three risky extensions. First, we should not assume that all behavior is voluntary; the constraints of rationalism are real. Second, we should not assume that all behavior is anomic. We do not all exist as separate individuals. Consider how the PD game changes if both players are in the Mafia; defection means death. Only anomic players (the kind we assume in rational choice) would mutually defect. Third, rational individualism does not imply egoism. Utility maximization can include altruism and ethics.


Y: We overcome the three political market failures (see below).

X: We establish institutions

But since this is a functional argument, our expectation of Y becomes X, and X becomes Y, like this:

Y: We establish institutions (to overcome these market failures)

X: Inverting Coase, Coase's three assumptions become our three variables: "problems of property rights, uncertainty, and transaction costs."

Causation: a functional story (see ch 5). This is not a Darwinistic-functionalism, but a "rational anticipation" functionalism; actors form institutions based on functions they expect the institution to serve.

X1: property rights (legal liability). Although states value their sovereignty too much to make strong legal institutions, these institutions do serve purposes related to law. If nothing else, they create "conventions" (coordination). At other times, they serve as "quasi-agreements" of the type that bind firms in a cartel; legally unenforceable, but informative.

X2: transactions costs. Regimes have several effects here. Besides lowering the transactions costs of negotiating, they also change the transactions costs of taking certain actions: "International regimes reduce transaction costs of legitimate bargains and increase them for illegitimate ones." (For example, GATT [now the WTO] raised the costs to states of raising a tariff: other states could retaliate.) Also, regimes bring many negotiations into a single forum, making the negotiation over side payments and multiple issues easier.

X3: Uncertainty and information. Drawing on Akerlof, Keohane notes that "The literature on market failure elaborates on its most fundamental contention--that, in the absence of appropriate institutions, some mutually advantageous bargains will not be made because of uncertainty--by pointing to three particularly important sources of difficulty: asymmetrical information; moral hazard; and irresponsibility." ASYMMETRIC INFO (the lemons problem): Regimes both provide information and facilitate signaling. It also helps if states have open governments. MORAL HAZARD: Institutions can change incentives in bad ways, just as the Federal Reserve removes Citibank's incentives not to make risky loans. IRRESPONSIBILITY (adverse selection): states may commit to more than they can do (especially if circumstances change). Just as banks check your credit rating, states know that those most anxious to cooperate are those with the least to give and the most to take.

  • Regimes help overcome information problems. They allow states to cooperate when, even though they want to, they might not otherwise do so due to "political market failure."


Regimes can't enforce their rules. Why do states comply most of the time? There are two possible arguments:

  • (1) States value the institution enough to cooperate even when they prefer not to (in a single instance) so as to maintain the institution. Williamson has shown that oligopolists do so; establishing cooperation is costly, but once it's set up, they find it better to maintain it.
  • (2) "Networks of issues and regimes": think tit-for-tat. States know that failure to comply now may lead to retaliation by others playing tit-for-tat. Moreover, states involved in iterated interactions worry about reputation, as it affects opportunities for future cooperation. All we need assume for these mechanisms to work is that (a) states value future interactions and (b) states keep track of who reneges (they monitor one another).