Prisoners’ dilemma in Madagascar
Using game theory to explain the Madagascar cycle
In the game theory literature, the Prisoners’ Dilemma is used to illustrate or to explain instances in which two parties do not cooperate, even when it is in their best interests to do so. It is best understood using a simple numerical example, in which two persons (let’s call them Rakoto and Rasoa) can either (i) cooperate, or (ii) act unilaterally. The payoff for each action is known by both players (see Table 1):
· If both cooperate, each gets 10 Ariary.
· If Rakoto is willing to cooperate, but Rasoa choses to act unilaterally, then Rakoto gets nothing while Rasoa reaps 15 Ariary. That is, Rasoa can take advantage of Rakoto’s good will.
· If, on the other hand, Rasoa wants to cooperate, but Rakoto prefers to act unilaterally, then Rasoa gets nothing while Rakoto collects 15 Ariary.
· Lastly, if both choose to act unilaterally, they each get a mere 1 Ariary.
In this example, the best outcome is for the two players to cooperate—both will earn 10 Ariary. This is achievable if both Rakoto and Rasoa cared about the joint welfare, and if they trust each other. The dilemma comes from the fact that they may not trust each other and that they only care about their own individual payoff. Even if they both initially commit to cooperate, there is a strong incentive to renege: If Rasoa cooperates, Rakoto can take advantage of her goodwill and get 15 Ariary by acting unilaterally (compared to 10 Ariary if he cooperates); if Rasoa acts unilaterally, Rakoto gets 1 Ariary by also acting unilaterally (compared to nothing, if he cooperates). Using the same logic, Rasoa’s rational response to any action by Rakoto would be to act unilaterally. If one player believes that the other one will act unilaterally, his or her best option will be to do the same. Thus, in the absence of mutual trust or a enforcement mechanism, both players will renege on their initial commitment to cooperate and will act unilaterally —both will get 1 Ariary. Failure to coordinate their actions leaves them with an inferior payoff even though a superior payoff is clearly within reach.
The Prisoners’ dilemma has been used to model many real world situations involving cooperative behavior. It provides a novel approach for explaining Madagascar’s failure to reach a sustainable and inclusive growth trajectory, and for understanding the Malagasy cycle of political crises.
As is often said, Madagascar is a country that is extremely rich … in terms of potential. There is no doubt that if all Malagasy parties cooperate, then economic development that is broadly shared is achievable (shown by the “shared prosperity” cell in Table 2). Indeed, there have been episodes, where many (not to say most) political players have shown a lot of good will and willingness to cooperate (or at least to remain constructive when they are in the opposition). These have resulted in high growth rates, at least in the short term.
Unfortunately, these growth spurts tend to be short-lived given the “winner-takes-all-and-keeps-all” mentality that is prevalent among Malagasy politicians. Soon after the benefits of cooperation become palpable, a party (the ruling one) will deviate from the collaborative equilibrium, and will start acting unilaterally. Political and economic power will become centralized and concentrated. Political adversaries will be harassed and excluded, as had happened in 1991, 2002, and 2009. While the growth rate may remain high, the resulting prosperity will not be broadly shared. Such “non-inclusive growth” is not sustainable given that, sooner or latter, the other (excluded) parties will also stop cooperating.
This is the typical narrative of the “Madagascar Cycle”: just as a growth is about to take off, a political crisis occurs. Simply put, the lack of cooperation and the prevailing mistrust among the Malagasy political elites tend to shift Madagascar to an inferior payoff equilibrium (i.e. the “stagnation” cell in Table 2). It is key to note here that, in the Malagasy context, “cooperating” could simply mean refraining from deliberately hurting one’s opponents, sabotaging each other’s projects, or adopting predatory behavior. Also, cooperation among the different parties does not necessarily mean an absence of an opposition; it suggests a constructive relationship between the ruling party and the opposition.
Going forward, it is crucial that the new Rajaonarimampianina administration understands this framework and avoid the trap of the inferior equilibrium of the “Madagascar’s dilemma.” Currently, they are still in a “honeymoon period”—most people are willing to collaborate with them, or at least to grant them the benefit of the doubt. This will not last for long. They need to quickly capitalize on that show of goodwill to build the foundation for genuine and sustained cooperation among all parties. This will go a long way towards delivering a solid growth rate that will benefit the vast majority of the 22 million Malagasy citizens.
More importantly, the new administration needs to send some credible signals that they will avoid the temptation of acting unilaterally by adopting opportunistic predatory attitudes once the “peace dividends” become clear. They should demonstrate leadership and resolve in establishing a climate of mutual trust and cooperation among Malagasy political actors, in the spirit of national reconciliation. It is the right thing to do. And this is the time to do it.
