Understanding the Causes and
Consequences of Non-cooperation in Politics
NADEEM HUSSAIN does the math.
Is there any rational explanation why the democratic political systems in place in the subcontinent appear to often find themselves in an equilibrium that is universally recognised as dysfunctional for the economy and polity? In this equilibrium all the major players refuse to cooperate with each other on key economic and political issues, thus failing to move the society to a known, more efficient, equilibrium where all parties are better off relative to the existing state of political affairs. Here I argue that such a rational explanation of the existing state of systematic political malaise can be constructed within a game theoretic framework. Specifically, the structure of a specific game widely known as the “Prisoners' Dilemma” is well suited for unravelling the fundamental forces underlying the political disorder.
The politicians' dilemma
Imagine an economy where two large parties dominate the entire political spectrum. Voters can only choose between these two parties. While the constitution provides for the conduct of a universally free and fair election through the institution of an independent Election Commission, both parties are endowed with enough ideological, money and muscle power to influence the election process, but the payoffs are interdependent and not necessarily certain. They thus have to be strategic in deciding how and when to use their resources to influence not only elections but also their access to the distribution of resources through the apparatus of the state.
At any given point of time, one of the two parties is in power and the other is in opposition. Assume for simplicity that both parties have equal bargaining power based on ideology, money and muscle. In running the business of the state, each party has to choose between cooperating and not cooperating with each other on any particular issue. Which strategy they end up choosing depends on the payoffs that accrue to each party individually under each of the two options -- cooperation or non-cooperation.
Consider all possible states in this game in the context of an illustrative payoff matrix presented below.
Note: In each cell the first number is opposition's payoff and the second is payoff to the party in power.
If both cooperate, they share power on the basis of free and fair elections and hence benefit equally since they have equal bargaining power by assumption. The economy functions efficiently and the sum of the payoffs to the two parties is maximised in this state. If the party in power does not cooperate while the opposition cooperates, the party in power gains more while the opposition loses some of its payoffs relative to when both cooperate. Non-cooperation gives the party in power an unfair advantage in gaining and maintaining access to power and the rents that come with it. Non-cooperation enables them to use, for example, repressive tactics to mute the strength of the opposition. The payoffs are symmetrically opposite when the party in power cooperates but the opposition does not. In this case the opposition takes undue advantage of the cooperative stance of the party in power through mudslinging and undermining the government's good work. When both do not cooperate, the functioning of the economy is disrupted. Payoffs to both are therefore lower and the sum of the payoffs to the two parties is minimised. Since both parties have the same strength, the payoffs are same to each.
Which of the possible four states in this game can emerge as equilibrium? According to Wikipedia, a set of strategies is (Nash) equilibrium if no player can do better by unilaterally changing his or her strategy. Imagine that each player is told the strategies of the others. Suppose then that each player asks himself or herself: "Knowing the strategy of the other player, and treating the strategy of the other player as set in stone, can I benefit by changing my strategy?" If any player answers "Yes", then that strategy is not a (Nash) equilibrium. But if every player prefers not to switch (or is indifferent between switching and not) then the set of strategy is (Nash) equilibrium. Thus, each strategy in equilibrium is a best response to all other strategies in that equilibrium. Such equilibrium may sometimes appear non-rational from a third-person perspective because it may happen that the equilibrium is not Pareto optimal. In other words, it may not fully exploit opportunities for making one party better off without hurting the other party. The equilibrium may also have non-rational consequences in sequential games because players may "threaten" each other with non-rational moves. For such games the subgame perfect (Nash) equilibrium is more meaningful as a tool of analysis. A '''subgame perfect equilibrium''' means that if the players played any smaller game that consisted of only one part of the larger game and their behaviour represents a Nash equilibrium of that smaller game, then their behaviour is a subgame perfect equilibrium of the larger game1.
Turn back to the payoff matrix now. Each party in deciding its own strategy will have to make an assumption about what the other party is likely to do. Consider the party in power. Its payoff is 7 if it cooperates and the opposition also cooperates. Its payoff rises to 8 if it does not cooperate while the opposition cooperates. Thus, given that the opposition cooperates, the optimal strategy for the party in power is to not cooperate. Now suppose the opposition does not cooperate. In that case, if the party in power cooperates, its payoff is 4 and the payoff rises to 5 if it does not cooperate. Hence, the optimal strategy for the party in power is again to not cooperate if the opposition is not cooperating. The same conclusion holds for the opposition since the payoff matrix is symmetric. Thus, irrespective of the strategy chosen by the other party, the optimal strategy for each player is to not cooperate. Non-cooperation is thus a dominant strategy and constitutes a Nash equilibrium. It is not a socially optimal equilibrium since the sum of the payoffs to both parties is higher in every other state. The non-cooperation equilibrium is stable because no party has any incentive to change their strategy from this state. Both parties are individually worse off if they unilaterally change their strategy. In fact the party that does not change its strategy benefits even more if it stays put while the other party changes its strategy. Note also that even if somehow -- say through a third party mediated coalition -- the two parties initially end up cooperating with each other, this cannot be a stable equilibrium. Each party can improve its payoff by moving from cooperation to non-cooperation while the other stays put.
Lack of foresight and trust results in non-cooperative behaviour
Note that the game described above is not actually a one shot game. It is a repeated game. In a one shot game, the players choose their strategies, there is an outcome and there are payoffs, and that's that. In contrast, a repeated game is played over and over. The players choose their strategies, there is an outcome and there are payoffs. Then they do it again and again and again. The game might repeat n times, where n is known or not known beforehand or it might repeat an infinite number of times. In the case in point, our politicians' dilemma is a repeated game and the players do not know n, but think that n might be large. The question then is why don't the players choose “Cooperate,” knowing it may cost in the short run (the current game), but believing that if they choose “Cooperate,” the other player will be more likely to also choose “Cooperate” in future plays of the game. Similarly, if one party chooses “non cooperation” in the current game, they may fear that the other player will punish them by not cooperating in the future. Under certain conditions -- if future payoffs matter enough -- cooperate, cooperate is an equilibrium in the repeated prisoners' dilemma. Thus we may see cooperation in situations like the prisoners' dilemma, where simple game theory indicates we should see non-cooperation. This is not because people are good-hearted or virtuous, but because of a dynamic social contract: “Let's cooperate now and get good payoffs; for if we don't, in future we will punish each other and get bad payoffs.” The key is that future payoffs matter enough to both parties.
Alas, foresight into such enlightened self-interest does not always happen in reality. Instead, players resort to retaliatory repeated game strategies affecting their choices within a game, contingent on what has happened in prior periods in the game. They adopt “tit for tat.” The tit for tat repeated game strategy works like this. In the first period of the game, the player chooses “Cooperate.” In any subsequent period, the player looks at his opponent's action in the previous period of the game. If the opponent chose “Cooperate” in the previous period, then the player chooses “Cooperate” in the current period; if the opponent chose “non-cooperate” in the previous period, then the player chooses “non-cooperate” in the current period. In short, the player matches what his opponent did in the last period of the game. This kind of repeated game strategy might be described as “crazy” or “tough,” but it might also be very effective. Under certain conditions, it can be shown that if player 1 plays “tit for tat,” there may be equilibrium in which both players are choosing “Cooperate” most of the time. It may sometimes be in the interest of people to have reputations as being “crazy” or “tough,” in order to induce beneficial changes in the behaviour of others.
A story of missed opportunities of political cooperation
Unfortunately that does not necessarily happen all the time. Tit for tat does not always result in healthy cooperation. Cooperation can fail because of a breakdown of trust even in a repeated game. In our illustration above, cooperation is perceived as a weakness and hence taken advantage of. Such a breakdown of trust may have historic roots. Consider the Bangladesh example. BNP and AL joined hands together to topple military autocracy in the early 1990s. BNP won the 1991 election following which the BNP-AL cooperation ended right away. Who triggered the latter is moot. However, in the way the political game in Bangladesh has panned out, time seems to have played a crucial role. It shaped up as, what in game theory is called, sequential move games or sequential games. BNP won the 1991 elections followed by AL victory in 1996 followed again by BNP victory in 2006 and AL victory in 2008. In sequential move games, the conventional wisdom is that there is a first-mover advantage. A first move sets the tone for the rest of the game, and the first mover can create the kind of play that she or he wishes. The first mover advantage seems to have been wasted in Bangladesh democratic political game. BNP refused to meaningfully share power with AL and the AL refused to cooperate with BNP from the very outset of electoral democracy in the 90s. The farcical February 1996 elections set the wrong tone. BNP had to accede to the demand for holding elections under a mutually acceptable composition of the caretaker government. The AL won the 1996 elections. Subsequently, the AL government's peaceful handover of power to a properly constituted caretaker government in 2001 was not reciprocated by the BNP government in 2006 when it appointed as Chief Adviser one of its own and made several other blatant attempts to influence the election outcome. While the AL did not tinker with the composition of the 2001 caretaker government, it did leave an administrative apparatus that, if not reshuffled by the CTG, could have influenced the election outcome. The difference between AL and BNP in twisting the rules was tactical, not intent. All these resulted in a huge loss of trust. The rest is history. The loss of trust may have induced the current AL government to punish BNP for failing to cooperate sequentially when AL itself established a precedent of some cooperation in 2001. These two parties have been the two most powerful political forces in the last two decades. When in government none of the two parties have shown much semblance of cooperative considerations for each other. Arguably, this non-cooperative stance has hardened over time.
Hopes not lost yet
There are many subgames within the larger political game where cooperation does happen. One such subgame is participation in elections. Both parties have participated in the last four elections. But it took some doing each time to build trust in the election process. At the end all parties agreed to participate only under a third party caretaker arrangement chosen on the basis of mutual consensus of the two parties. When in power, each party attempted to constitute its own brand of caretaker government. When such attempts rose to extremes, both parties lost control over the game as in 1/11/2007. The strategies played by the two parties in late 2006 resemble the game of ''chicken''. The principle of the game is that while each player prefers not to yield to the other, the worst possible outcome occurs when both players do not yield. This results in a situation where each party, in attempting to secure their best outcome, risks the worst and loses to a third force (1/11). Interestingly, the two parties rejoined forces, albeit implicitly, against this third force after 1/11. This shows that when there is a credible third party threat, cooperation between the two parties becomes an option that could serve well the self-interest of each party. Such cooperation has indeed helped the two parties regain control over the game and keep it within themselves. Following that the two parties again have succumbed to the temptations of non-cooperation.
The good news is there are some subgames, particularly in the area of economic and social policy, business and legislature, where the relevant representatives of the two parties do cooperate implicitly or explicitly. Both parties have moved forward the structural and social policy reform agenda pursued by the previous government with some exceptions here and there. These include trade reforms, fiscal reforms, public financial management, de-regulation, energy, infrastructure, education and environment. Another significant example of explicit cooperation is participation in the deliberations of the parliamentary sub-committees. There are social forces such as matrimonial or ancestral relationships which unite some members of the two parties outside politics. In such situations the two parties have adopted the “live and let live” stance. Unfortunately, scaling up such cooperation in the larger political stage has proven to be well nigh impossible so far. The country appears to be heading towards another political impasse in 2013.
No shame being a chicken
The two parties have adopted the policy known as 'brinkmanship'. This is a policy adapted from a sport called 'Chicken!' played by some youthful degenerates. It is played by choosing a long straight road with a white line down the middle and starting two fast cars towards each other from opposite ends. Each car is expected to keep its wheels on one side of the white line. As they approach each other, mutual destruction becomes more and more imminent. If one of them swerves from the white line before the other, the other, as he passes, shouts 'Chicken!” making the one who has swerved become an object of contempt. As played by raucous boys, this game is considered decadent and immoral, though only the lives of the players are risked. But when the game is played by eminent leaders, who risk not only their own interests but those millions of fellow citizens', it is thought on both sides that the statesmen on one side are displaying a high degree of wisdom and courage, and only the leader on the other side are reprehensible. This is illogical. Both are to blame for playing a dangerous game. The game may be played without misfortune a few times, but sooner or later it may reach a point where loss of face is perceived as more dreadful than mutual exit as happened in 1/11. The moment may come when neither side can swallow the mocking cry of 'Chicken!' from the other side. When that occurs, the leaders of both sides will plunge the country into chaos. Brinkmanship involves an element of uncontrollable risk: even if all players act rationally in the face of risk, uncontrollable events can still trigger a catastrophic outcome.2
Threats and counter threats perpetuate the race to the bottom
One specific form of brinkmanship is threats. A threat is an announcement made by a player at the beginning of a sequential move game, indicating that at some stage, at some point in time, he will depart from what is rational in order to punish the other player. I won't do what you think I ought to do. For instance the current opposition could announce a stance signaling we will take us all down if the government does not restore the system that existed when they won the last election. Obviously, if the government believes the threat, it should restore the CTG system. But in a sequential move game like this, if the players carry through on its threat, they may end up worse off. So threats like this may seem not too credible. Of course life is more complicated when games are played over and over, or when parties play games with different partners, and develop reputations that spread out to other players. If a game is played over and over between two players, an aggressive player may carry out threats in the initial games, so that his playing partner comes to believe that he will carry out his threats, no matter how self-destructive they may be. In this case, his partner becomes trained to give in to his threats. Or, if he plays with many different players who talk to each other, an aggressive player may want one player to see that he carries out his threats, so that word gets around. With more than one such aggressive player, this could potentially ignite and perpetuate a race to the bottom.
Consequences of asymmetric distribution of political power
The discussion above assumes that the two parties are equally powerful. What if they are not? This by no means renders the power balance inconsequential. Here lies the dilemma for the weaker party. Power asymmetry gives the stronger party greater capability to go it alone. This affects the respective costs of agreement failure, which enables the stronger party to hold out longer and ultimately gives it greater sway to obtain its preferred outcome. The weaker actor faces a conundrum: Unwilling to accept inequitable 'take-it-or-leave-it' policy solutions, yet unable to balance against the other actor in using force, what is the weaker actor to do to avoid perpetually obtaining only minor concessions in cooperation? One answer is soft balancing in contrast to traditional hard balancing. Soft balancing is a strategic effort by a weaker actor to increase influence vis-a-vis a stronger actor via non-violent means. Soft balancing does not seek a permanent confrontational stance. Rather, the goal is to nudge the other party back to a cooperative framework of mutually acceptable concessions.
There are at least two types of soft balancing strategies. One set of strategies might be called 'levelling strategies.' Parties use levelling strategies to restructure the situation either by seeking constraints on the alternatives of stronger parties, or by increasing the alternatives available to themselves. Thus, the effort to pursue an increasingly institutionalised multi-party system imposes constraints on dominant parties by stressing collective problem-solving and equality before the law. The resultant third party support increases its alternatives and subsequently its influence in relation to the dominant party. The weaker parties even seek alliances with other even more weak parties to increase the cost of going it alone for the stronger party.
The other option is strategic non-cooperation. This is when a weak party seeks to increase future influence vis-a-vis a strong party by deliberately rejecting inequitable cooperation. While walking away from gains is perplexing in the short run, it makes sense if the bargaining outcome is highly asymmetrical. This is not necessarily because of the traditional relative gains concerns. Rather, parties calculate that their reputation as a weak negotiator will erode future bargaining power and subsequently future share of absolute gains. Strategic non-cooperation is therefore a signal of resolve to increase influence in future decision making. As such, it is a soft balancing tool for weaker parties to leverage their influence vis-a-vis stronger states.
Is strategic non-cooperation rational and effective?
Strategic non-cooperation is only rational if there is a reasonable chance that it will enhance your position in subsequent bargaining. Does the strong party have any reasons to compromise? There are at least two differing lines of argument on this. The classical realist argument is that such an expectation is illogical. If one actor is dictating the terms of cooperation to another, it is most likely that there is an asymmetry in the capabilities and resources of the parties that allows the stronger actor to impose solutions on the weaker actor. A pure capabilities-based analysis therefore predicts that a stronger actor determines the final outcome, since she faces the same incentive of optimising her long-term gains, and greater resources should enable her to hold out longer. This is consistent with the argument that power is often the determining factor in distributional bargaining. In this narrow 'power as capabilities' analysis, strategic non-cooperation is an irrational optimising strategy.
This is the paradox. A strong party pushes the weaker party around exactly because it believes it can get away with it. However, this is the very same reason that the weaker party actually has an incentive to fight back. It does not want to be stuck indefinitely in such an asymmetric bargaining relationship. Therefore, traditional power asymmetry is not always the determining factor in negotiations. When power is conceptualised not only at the aggregate level, but also at the issue-specific and behavioural level, then strategic non-cooperation may be both necessary and an entirely logical soft balancing strategy.
The stronger actor may not always win
Indeed, there are several reasons why the stronger actor may not win. A pure power-based analysis of influence fails to take into account the countervailing dialectical forces generated by these asymmetrical structural relations. A preponderance of objective capabilities may be countered by an asymmetry of opposing desires, as when the weaker party desires its freedom from domination more than the stronger party is bent on dominating it. This willingness to accept economic punishment must be taken into account in assessing the stability of the dependence relationship.
Another key factor in determining outcomes in asymmetrical bargaining is unequal cost from the failure to agree. While stronger parties frequently have better alternatives, they may also have more at stake. The respective costs of non-cooperation therefore depend on each actor's vulnerabilities and sensitivities to non-cooperation over time. This is also consistent with Shapley's (1953) original game theoretic finding that it is not sufficient merely to compare payoffs from cooperation. Rather, asymmetrical costs of non-cooperation drive asymmetrical distributions of gains. Thus, the alternative to cooperation is important. It is not necessarily the case that superiority in hard power correlates with lower cost of alternatives. Consequently, it is not given that strategic non-cooperation will produce a perpetual stalemate, or that the stronger actor will never relent.
Parties seeking to align divergent policy preferences may face one of two basic situations. At one extreme, their initial policy preferences may be truly incompatible. This is a situation of deadlock, and without side-payments, only some form of coercion can bring about policy alignment. Alternatively, although the parties' policy preferences differ, there may be a zone of possible agreement where policy alignment, even if not at either party's ideal point, will still yield net benefits to each party. If the parties manage to align their policies, this is cooperation. If they do not, this is non-cooperation. It matters whether the rift, or for that matter any conflict, is a case of deadlock or non-cooperation. Deadlock assures a grim future because it implies fundamentally irreconcilable preferences. Non-cooperation, however, retains potential and prescriptions for achieving cooperation.
Bangladesh's 64 billion dollar question
When non-cooperation is associated with perceived political gains by the key political players, the outcome not surprisingly is non-cooperation making everyone worse off, assuming equal bargaining power. In this equilibrium, any unilateral effort to cooperate makes the non-cooperating party better off and the cooperating party worse off. Consequently, no single player has an incentive to initiate cooperation. What is the way out? The solution lies in the design of an institutional structure that changes the perceived payoffs to non-cooperation by building trust on all sides. India is a good example. Politics in India is no less dysfunctional than in other countries in the region. But there are certain institutions such as the Election Commission, the CIB, the parliament and the judiciary which are trusted by the two major parties. These institutions have created the conditions for the political parties to seek to cooperate with each other in holding free and fair elections and for delivering the most basic services. In countries such as Bangladesh and Pakistan, politicisation of these institutions has resulted in almost irreparable loss of trust. At the end of the day, sustained political will to depoliticise these institutions to restore trust will be needed. Where and how this political will come from is our 64 billion dollar question.
1. An Introduction to Game Theory, Osborne, M.J., Oxford University Press, USA, 2004.
2. Dixit and Nalebuff (1991) pp. 205222.
Nadeem Hussain is a Masters' student at Brandeis University, Boston, USA.