Why do cooperative ventures between some countries work, and not others? Current literature identifies lack of common threats, base identities, and overlapping goals as dilemmas that make cooperation among certain sovereign states unlikely. Indeed, if cooperative ventures are viewed from a Darwin-Waltz perspective, weak institutions, norms, and domestic regimes explain the difficulties countries face in collaborating. Adding to these traditional explanations, in this post I want to use game theory to introduce the concept of a ‘supervising agent’ to identify why regional may be unsustainable in the long run.
In game theory, sustaining cooperation relies on two fundamental assumptions. The first is about punishment: even with repeated interaction over infinite N periods, a defector member state that does not play by the ‘rules of the game’ will be removed from the union, thereby hindering access to competitive advantages, be they technical learning through information exchange, energy development, increased trade and industrialisation, and so on. The second assumption is about preferences: it is assumed that member states strictly prefer returns from belonging to a regional group than they do to functioning independently.
To add to the analysis of non-cooperation, I want to introduce the concept a ‘supervising agent’. Supervising agents serve the role of distributing rewards from the bloc to members, and in doing so they both reward and punish. Each member, irrespective of membership to a wider group (i.e. membership of a wider identity group), makes its decision with imperfect information (not knowing what the other members have decided). The supervising agent, on the other hand, moves afterward with complete and perfect information.
The portion of total benefits available to the supervising agent depends on the number ‘punished’ for defecting (defection in the form of trade protectionism, for example). Punishment means that the supervising agent withholds rewards from members. The payoff function for the supervising agent is characterized as:
To explain in more detail: is the portion of benefits; x is the number of members sanctioned for rebelling against rules set by the bloc; and N is the total number of bloc member countries. A supervising agent’s payoff is affected by the actions taken by other member states in the regime. As the proportion of member countries that play by the rules of the game rises, the payoff to the supervising agent grows. In this way, the supervising agent has a stake in the total benefits created by the regime, plus the special role as a hegemon, sanctioning members that do not cooperate.
The stream of benefits produced by the bloc, as well as the costs of defecting, influence member states’ payoffs. Both of these parameters are seen in the payoff function for member states at any t iteration of the game. The behaviour of other members affects the stream of benefits produced by the bloc, but unlike the supervising agent, they do not exercise sanctions against defecting members.
These payoffs are presented as:
Given these payoff functions, member states utilize two different trigger strategies. The first, the grim trigger (g-trigger), means that a member supports the regime until another member rebels, or the supervising agent fails to reward after cooperation, such that:
g-trigger t = 0: support
t > 0: support if b( and all agents, a, are rewarded, i, for all t*<t,
Alternatively, members will utilize a strategic trigger wherein they defect when it is in their best interest for that iteration of the game, such that:
s-trigger t = 0: support
t > 0: support if b( and the leader rewards member, i , on all
t*<t, defect otherwise.
With the s-trigger strategy, members will support the union as long as the benefits of supporting the bloc exceed the benefits of defecting. Given these strategies, cooperation between different countries can occur without the supervising. My point is, however, that such cooperation is difficult to sustain in the long run. There is always an incentive to defect (within member country groups, because of underlying identity conflicts). Such a defection, in turn, can set off other members’ defections, resulting in widespread defection.
As such, a supervising agent can play a critical role in sustaining cooperation. Threats of fellow members can sustain cooperation (that is, the grim trigger strategy), but the rewards and punishments of the supervising agent play a large role. It is interesting that the supervising agent observes each member’s strategy choice, , for each iteration, t, of the game, rewarding and punishing each member separately. In this manner, the supervising agent establishes an incentive system to influence the actions of members of the regime. A supervising agent serves the role of distributing residuals produced by the union and sanctioning all rebelling member countries.
Therefore, the supervising agent follows a M-trigger strategy whereby he creates an incentive system by rewarding and punishing members:
M-trigger t = 0: reward all members, i
t > 0: reward member i if =1 on all t* < t, punish defectors
Under such conditions, and with such leadership capabilities, support for the union can be enforced through punishment, meaning leadership in the form of the supervising agent will play a more significant role in attaining cooperation. As long as two conditions persist (a low probability of the game ending, and mutual expectations about reciprocity) we can expect to see support for cooperative ventures. If these conditions are called into question, the supervising agent can serve to facilitate cooperation by enforcing compliance.
For example, Ayoob (1985:447) discusses the need for a ‘pivotal power’ within the regional grouping of the South Asian Association of Regional Cooperation (SAARC), which will ‘set limits’ which neither the pivotal power nor its partners may stray away. According to Paul (2013:120), the presence of India’s dominance within the South Asian region is clear: it contains 74% of the region’s population, 75% of its GDP, 79% of its trade, and caters to 81% of the region’s Foreign Direct Investment (FDI) flow
However, I question this assertion because Khosla (2007) argues that a necessary precondition for regional cooperation is an acceptance by member states of each other’s sovereignty and legitimacy. Pakistani opinion regarding the Kashmir issue, for example, is a questioning of India’s sovereignty and Khosla’s preconditions regarding ‘pluralist security communities’. This leads to a nebulous nature of a pivotal power, or a ‘supervising agent’, resulting in a lack of a mutual dispute settlement system.
Thus in order to increase the success of SAARC as a regime, it is necessary to realign the balance of power in this regional grouping. This would be achieved by assigning a functional power the power to punish non-cooperation. The lack of a supervising agent, coupled with the inability to fulfil necessary preconditions regarding successful regional cooperation: whether these being Khosla’s (2007) (a rejection of war, a mutual dispute system, a shared culture and common threats) Ayoob’s (1985) (ideological and political affinity) and Bandara and Yu’s (2003) (geographical advantages, trade complementarity, lack of political tension, and differentiated economic structures) means that member states are acclimatized to defect, making the regional entity captive to on-going disputes.
Ayoob (1985) “The Primacy of the Political: South Asian Regional Cooperation (SARC) in Comparative Perspective”. Asian Survey, 25(4), pp. 443-457.
Bandara, J. S., & Yu, W. (2003). How desirable is the South Asian Free Trade Area? A quantitative economic assessment. The World Economy, 26(9), pp. 1293-1323.
Paul, S. (2013). Role of SAARC in strengthening the relationship between India and Pakistan: A critical analysis. Asian Journal of Research in Social Sciences and Humanities, 3(1), pp. 114-125.
United Nations Department of Economic and Social Affairs: Population statistics. Projections available online: http://esa.un.org/unpd/ppp/index.htm.
United Nations Development Program: Indices and Data available online: http://hdr.undp.org/en/statistics/.
Waltz, K. (1959) “Man, the State, and War” Colombia University Press.
 [This means that competition selects outcomes according to their consequences, and falls under the grander theory of ‘sociobiology’. I use it as a reference in this essay because both Waltz (‘Man, the State, and War) and Darwin argue that structure affects agents’ behaviour through (1) competition and (2) socialization. The premise both theorists employ make international structures as mechanistic and unsocial as possible. We can link this to two books – “Social theory and International Politics by Alexander Wendt: Cambridge University Press, 1999: 101-109 and ‘Realism and World Politics’ by Ken Booth: Taylor and Francis, 2011: 326-329]
 These statistics are consistent with data from the UN Department of Economic and Social Affairs.