Macroeconomic Cooperation
Sustenance

Sustainability is a key question in the design of rules for international economic policy cooperation. Cooperation may be unsustainable because an incentive may emerge for one or more countries to renege on the internationally agreed set of rules, and to pursue individually designed, non-cooperative policies. The existence of such an incentive to renege will, over time, lead to a breakdown of cooperation. Moreover, if participants foresee this, they may be unwilling to conclude any cooperative agreement in the first place. If policy coordination is to be a realistic policy option, it is essential to discover how cooperative rules can be made sustainable.

The sustainability of cooperative policies can be enhanced by supplementing them with an appropriate threat strategy to be carried out against countries which renege. An ideal threat strategy will punish a reneging country, making any possible non- cooperative strategy less attractive to that country and thereby sustaining cooperative behaviour in all circumstances. To be effective, however, the country which is threatened must believe that the threat will be carried out. The threat must therefore be one that the punishing countries can carry out without inflicting undue damage on themselves. It is therefore important to investigate whether there exist plausible and credible threat strategies that sustain cooperative forms of behaviour in the international economy.

In Discussion Paper No. 102, International Macroeconomics Programme Co-Director David Currie and Paul Levine investigate the question of sustainability in a two-country world. They model policy-making as a 'supergame', that is as a game played repeatedly over time and not on a single occasion. The authors examine whether there is any simple threat strategy which will sustain policy cooperation in this model. One possible threat is that of choosing a non-cooperative strategy oneself if the other country reneges. Currie and Levine consider a number of such threat strategies involving the so-called 'Nash' equilibrium, in which the country chooses that strategy which maximizes its own objective function, taking the other country's strategy as given (with the other country behaving similarly). Currie and Levine find that such threat strategies are credible, i.e. the country which is tempted to renege believes that the threat will be carried out. Currie and Levine find that the threat is effective in sustaining cooperative behaviour in their model. A country which is tempted to renege not only believes that the other country will carry out the threat, but also finds that the threat is sufficiently powerful to prevent it from reneging because it will be worse off if it does so.

Currie and Levine also emphasize the importance not only of relationships between governments but also relationships between governments and the private sector. A fully optimal policy rule is often considered to lack credibility because it is time- inconsistent: the mere announcement of government policy intentions changes private sector behaviour, which creates an incentive for the government to renege on its policy announcement and hence its commitment to the private sector. Currie and Levine have explored this question for a single economy in Discussion Paper No. 94 (reported in Bulletin No. 14). There, they found that the presence of continuing shocks to the economy, which make the policy game a repeated one, may well remove the temptation for the government to renege on its policy announcements. If the government reneges on its announcements today, its reputation will suffer, and this may reduce its ability to deal with shocks to the economy in the future. Concern for reputation may therefore render the fully optimal rule time-consistent and thus credible and sustainable, provided that the government does not discount these future shocks too heavily.

Currie and Levine find this earlier result confirmed in their two-country model. Concern for reputation renders fully optimal policy credible, and this holds both for commitments made by the two governments to each other, and for commitments made to the private sector. If one government reneges on the private sector, this may also lead to a breakdown of cooperation between the two governments. Since a breakdown of cooperation imposes additional costs, this provides an extra incentive not to renege on the private sector.

Currie and Levine also find that the effect on a government's 'reputation' of reneging is crucial to the gains that can be made from cooperation. In models where there are no 'reputation effects', policy must be time-consistent to be credible: the gains from cooperation are small or even negative. But in contrast, when fully optimal policies are pursued in models where reneging affects reputations, the gains from cooperation are large relative to the non-cooperative outcome. An interesting finding is that the worst equilibrium is the non-cooperative one in which reputations matter. These results lead Currie and Levine to conclude that in an interdependent world, government reputations may prove beneficial if governments cooperate but damaging if they do not.


The Sustainability of Optimal Cooperative Macroeconomic Policies in a Two-Country World
Paul Levine and David Currie

Discussion Paper No. 102, April 1986 (IM)