Exploring the agent/process problem in international negotiation

How do power asymmetries and domestic influences effect the course and outcome of international bargaining? The World Trade Organization’s (WTO) most recent round of multilateral negotiations have suggested an answer to this question in its struggle to continue the economic liberalization process. Known as the Doha Development Round (“Doha Round”), substantive talks began in November 2001 and reached a stalemate in 2008, in which rising powers and developed countries failed to settle an agreement on the scope, modalities, and aims of the negotiations on agriculture[1].

The following analysis seeks to expose the theoretical basis of the Doha Round and its failures by examining the events primarily through the lens of William Zartman’s model of process and implementation. For the purposes of this study, I will specifically examine the bargaining relationship between the developed and developing country coalitions on the issue of agriculture liberalization. Where process analysis fails to account for the intricacies and dynamics of state actors, I will insert W. M. Habeeb’s theory on issue-specific bargaining power as well as Robert Putnam’s model of two-level games. Using the Doha Round as a case study, I will contend that general theories of process as espoused by Zartman fail to capture the complexities and difficulties of multilateral negotiation in an intergovernmental context. In order to fully realize this, factors at both the structural and domestic levels must be considered.

Before delving into a more substantive analysis, it is important to first flesh out Zartman’s theoretical dictates. In what is known as the “formula-details” approach, it is held that negotiation, at its core, involves the finding of a formula that embraces the best combination of the parties’ interests which, upon discovery, requires the actors to parse out the details for implementation[2]. Zartman privileges the process over the agent as the central determinant of decision making, in which he views negotiation as a movement through a number of phases until the conditions are sufficiently “ripe” for settlement[3]. This includes the existence of an ongoing stalemate, the looming threat of catastrophe, legitimate representation, and mutual recognition that an agreement would produce the best way forward for all parties[4]. In the event that all four of these criteria are satisfied, it is assumed that the stage is set for an amicable negotiated solution. With these established, actors then seek to reach a formula that abides by their respective “minimax” threshold—that is, one that encompasses their maximum number of interests and the minimum of the other party’s required for agreement[5]. In doing so, states trade concessions until they converge upon a “win-set” that is congenial to all parties; in what is essentially a reactionary process, negotiators monitor the behavior of their opponent and counter such that it progresses discussion beyond the impasse while also being careful not to jeopardise the minimax objective[6]. This is known as the “concession-convergence” approach. Although the formula-details and concession-convergence methods are posed as separate models of conceiving negotiation, Zartman goes on to note that both are fundamentally compatible with each other and to some degree complement the other where they are found lacking[7]. Where concession-convergence examines agent-specific relations and their respective interests, an opportunity arises to impart “useful information on the relation between [the negotiators’] behavior and the process of finding a formula and implementing details”[8]. In other words, finding the best formula—and later agreeing upon the specifics—involve a process of conceding and converging and insight into their opponent’s behavior and intentions thereof.

As it relates to the Doha Round, theory meets practice most prominently at the outset, or in Zartman’s terms, the “formula-finding” phase. This is owing to the fact that negotiations simply never make it beyond this stage. Above all else, conflict arose surrounding the agriculture talks; more specifically, between the producers in the global north and south regarding improved market accessibility in the face of tariff cuts and the elimination of nontariff barriers (NTB) such as export subsidies, domestic support, and production quotas[9]. From the commencement of the Doha talks until 2005, thirteen major meetings failed to produce a resolution on agriculture as parties lacked consensus on the “modalities” (i.e. formula) and general principles for an agreement[10]. This is due to the emergence of coalitions across global north-south lines; pacts precipitated by developing countries’ (hereafter DCs) opposition to farming policies and domestic subsidies in developed export nations voiced at the 2003 Ministerial Cancun meeting[11]. Regarding the agriculture modalities, talks effectively halted following conflicting proposals on the Special Safeguards Mechanism (SSM), which would afford DCs flexibility to protect their domestic producers from an import surge by imposing temporary tariffs over a range of products[12]. In this respect, the principal actors of the DC coalition consisted of India, Brazil, and China, while, on the other hand, the United States, Japan, and the European Union  led a united front to impose new market-access obligations on developing states[13]—a demographic that comprise two-thirds of WTO members. The main points of contention had to do with the assumed increase in food prices if these proposed reforms were to be carried out—as the price rises, so too would developed country imports from DCs, where there is commonly a comparative advantage in agriculture production. Thus, talks failed to gain traction beyond a July 2004 summit. There, DC leaders were able to secure concessions in the form of a conditional termination of EU export subsidies[14], yet no substantive larger agreement could be finalized as the United States, among other developed states, refused to comply on the issue of domestic farm subsidization[15].

Interestingly, this impasse came about despite the Applied General Equilibrium (AGE) quantitative models that support the fact that absolute gains (in economic terms) were to be gleaned if a deal were to pass[16],[17]. To this end, a 2008 estimate from the World Bank purported that if terms were agreed upon that effectively lifted agricultural tariffs and NTBs then global welfare could rise by as much as $3 trillion USD[18], with the majority of which going to the developing world in the form of trade surpluses[19]. Given these circumstances, according to Zartman’s model of process, one would reasonably expect negotiations to culminate in a deal. This is by virtue of the four “ripeness” criteria having been satisfied, that is: i) the existence of an ongoing stalemate, ii) an impending threat of disaster, iii) legitimate representation, iv) and mutual recognition that an agreement would produce the best way forward.

While qualifiers (i) and (iii) are relatively self-evident, criteria (ii) and (iv) require some degree of explanation in order to make clear. With respect to (ii), the looming “disaster” scenario is one in which a negotiated solution does not emerge and thus all potential positive-sum trade bonuses are foregone. There are, for all involved parties, deleterious economic and political consequences attendant to such an outcome. For (iv), it is important to consider that unlike adjudication or coalition decision-making, according to Zartman, negotiated decision-making is systematized as a joint process between actors who realize the positive-sum nature of an agreeable solution[20]. In this case, this “win-win” quality is undeniable from the perspective of the principal actors. On one hand, the developed states (US, EU, and Japan) were reportedly most interested in the liberalization of large and middle-income developing economies for greater agricultural market access[21]. On the other hand, developing states (largely represented by the G-20), advocated for substantial reform to domestic subsidization of farming and exports in developed states following a joint US-EU proposal of August 2003 that failed to adequately address the issue[22]. As author J.J. Schott comments, “[the G-20] position is not antithetical to US interests and objectives in the Doha Round”[23]. This alludes to the fact that the parties at the bargaining table had objectives sufficiently compatible such that a deal would provide them with a solution mutually preferable to the status quo.

Therefore, from a purely high-level standpoint a deal makes sense for all parties barring any ulterior motives or relative gains considerations—factors not directly addressed in Zartman’s model. Hence, one can reasonably conclude that the Doha bargaining conditions were such that a process-centric theorist would presume the modalities to be agreed upon. However, this is not what is seen in practice. Therefore, it becomes apparent that a more comprehensive analysis must be conducted in order to fully understand why negotiations failed to produce a substantive solution. To discover why, one must look outside the limits of process theory and instead turn their attention to structural and agent-level factors.

Where process-centric models of negotiation are found most deficient is in their omission of power analysis. This oversight does not lend itself well to the study of political behavior, an endeavor attached to power dynamics by necessity. Habeeb’s study of power is particularly useful in filling this analytical gap. Whereas Zartman simply assumes symmetry between bargaining actors in terms of power, Habeeb posits an agent-centric approach to negotiation that considers power discrepancies between actors and closely examines how they are operationalized. For the purpose of this case study, it is important to distinguish between Habeeb’s conceptions of aggregate structural power versus issue-specific structural power. Aggregate structural power refers to “an actor’s resources, capabilities, and position vis-à-vis the external world as whole”[24]; this includes absolute material wealth in terms of gross domestic product, and/or broader structural power such as China’s privileged status as a hegemon given their growing diplomatic and economic strength within the international system. Conversely, issue-specific power refers to “an actor’s capabilities and position vis-à-vis another actor in terms of a specific mutual issue”[25]; in this case, it can refer to the G-20 negotiators capabilities in relation to the US on the issue of agriculture and food security.

When accounting for the disparities between the Doha Round negotiators in terms of issue-specific structural power, a clearer picture develops as to why the talks resulted in stalemate. Although in aggregate economic terms the combined power of the G-20 developing states competes with the United States and the European Union, there is a stark asymmetry of influence in the agricultural sector. In 2011, the US and EU stood as the top two global foodstuff exporters with a combined $234 billion (USD) in exports—over twice the combined amount of Brazil, China, and Argentina, the top three exporters of the G-20 developing states[26]. Further, between 2002 and 2004, the developed economies of the OECD implemented price-distorting measures on agriculture in the form of tariffs over five times higher than any other sector on average—amounting to a total of $254 billion in producer support[27]. Despite having a clear advantage in material power, this has not been able to translate into effective bargaining leverage. Thus, this stands as evidence of an “actualization problem” in framing power as a possession of material capabilities[28]. On the particular issue of agriculture liberalization, however, developed states stand to gain a considerable amount more than developing states do. According to models by the Global Trade Analysis Project (GTAP), in gross terms, high-income states were projected to gain 41.6 billion dollars in agriculture liberalization compared to 11.9 billion in the developing world[29]. In per capita terms, this amounts to $40 per person in developed states versus $2.54 in developing[30]. This reveals the fact that liberalization is simply worth more to developed states in regard to net economic benefit. The asymmetrical interdependence, then, can be used by developing states to prolong the bargaining process and negotiate a better deal under the assumption that high-income states would be more likely to shift their win-set to secure an agreement. Thus, both coalitions are found in a bargaining position in which they are pitted against the other; on one side, a faction of developing states with an issue-specific relative advantage, and, on the other, a bloc of developed states unwilling to concede by virtue of their perceived aggregate structural advantage.

While one may emphasize that relative bargaining power privileges one actor over another in absolute terms (i.e. country A has a food surplus, thus is less “dependent” on an agriculture deal than impoverished countries B and C), it is important to assess power dynamics not in relation to vulnerability but instead by examining where exactly a state’s power emanates from with respect to the particular issue[31]. In the case of the Doha talks, while the US may have a relative advantage over developing states in terms of having greater available substitutes for trade, this assumes that government preferences are static, fixed, and easily ascribed—the same assumptive fault that Zartman’s formula-details approach commits. In reality, however, state interests are subject to the influence of both the suppliers and the consumers (e.g. American wheat farmers and Indian wheat consumers)[32]. This demonstrates that when assessing power symmetries, it is essential that one bear in mind the non-fixed nature of state preferences.

In accounting for the influence of domestic political forces, Putnam uses the metaphor of a two-level game. Rejecting the notion that states behave as cohesive, unitary actors, Putnam espouses the idea that, at the national level, political pressure groups seek to influence national governments by lobbying them to adopt preferable policies. At the international level, governments seek to maximize the interests of their domestic constituents while at the same time “minimiz[ing] the adverse consequences of foreign developments”[33]. This gives rise to an interplay between domestic pressure groups and national governments in coming to an agreeable position in international negotiations—when the win-sets of both game players overlap they form an agreement zone in which a negotiated settlement can take place. In assessing the Doha talks, one must consider the various two-level games being played on both sides of the negotiation.

The United States, on one side, are beholden to a considerably influential farm lobby, headed by the American Farm Bureau Federation that staunchly advocates in favor of both the existing agriculture subsidies and greater foreign market access[34]. This is reinforced by the Grossman-Helman econometric model that “clearly establish[es] that interest-group money bends agricultural policy in the U.S.”[35]. Thus, it is entirely plausible that domestic forces minimized the United States’ acceptable win-set such that negotiations were bound to break down. That is, it is reasonable to assume that the audience costs were so high for US negotiators that their hand was effectively forced into adopting an aggressive protectionist bargaining position. This becomes an acceptable position to take when one considers the vast influence the Farm Bureau, among other stakeholders, possess in maintaining the political viability of US officials by way of campaign contributions[36]. Further, the argument can also be made that the United States’ decentralized federal framework provides an unpalatable political landscape for implementing cuts to domestic subsidies that enjoy popular and state-level support. For instance, the legislatures of Iowa, Illinois, Texas, and other farm-intensive states may reject the terms of an unfavorable agreement to such an extent that it becomes politically non-viable, thus causing the deal to fail. This provides a further degree of complexity to the United States’ domestic game table. Neither of these arguments can be said of certain developing states, such as China, that occupy the other side of the negotiating table. Given their heavily centralized, if not authoritarian, policy-making mechanisms it makes sense to assume that minimal domestic opposition would stand in the way of a deal’s implementation. For this reason, it can be aptly concluded that prominent Doha round actors such as the United States may have had their ability to negotiate effectively impeded by influential domestic level forces. This is due to their minimization of the US’ win-set such that it could not move forward with an agreement zone that overlapped their opponent’s.

In conclusion, it is clear that the process-centric literature on bargaining assumes governments act as unitary actors, in perfect symmetry with one another, with easily ascribable values and preferences; however, this reduces the study of negotiation to an overly simplistic formula. In order to form a more complete understanding, an agent-specific analysis must complement the study of process. To this end, the works of both Habeeb and Putnam effectively patch the holes found in Zartman’s formula-details approach. When addressing for issue-specific power dynamics, and the influences of domestic-level forces specific to the bargaining agent, one realizes why negotiations may have failed to produce an agreeable outcome. This is in contrast to Zartman’s model which leaves little room for explanation as to why talks came to nothing. Thus, future research must shift attention to the agent and not the process of negotiation, as it is impossible to have a clear understanding of the latter without the former.


[1] Francois, J., van Meijl, H., van Tongeren, F., & Evenett, S. J. (2005). Trade Liberalization in the Doha Development Round. Economic Policy, 20(42), 351–391

[2] Zartman, W. (1977). Negotiation as a Joint Decision-Making Process. Journal of Conflict Resolution, Vol. 21. 635.

[3] Hampson, F. O. (1995). Multilateral Negotiations: Lessons from Arms Control, Trade, and the Environment. John Hopkins University Press. 16.

[4] Ibid.

[5] Zartman. 636.

[6] Ibid. 625.

[7] Ibid. 637.

[8] Ibid.

[9] Hufbauer, G.C., Schott, J.J., & Wong, W.F. (2010). Figuring out the Doha round. Policy Analysis in International Economics, Vol. 91. 4.

[10] Bouet, A., Bureau, J.C., Decreux, Y., & Jean, S. (2005). Multilateral agricultural trade liberalization: The contrasting fortunes of developing countries in the Doha round. The World Economy, Vol. 28(9). 1329.

[11] Ibid.

[12] Hufbauer, et. al. 24.

[13] Schwab, S.C. (2011). After Doha: Why the negotiations are doomed and what we should do about it. Foreign Affairs, Vol. 90(3). 107.

[14] Bouet, et. al. 1329.

[15] Elliot, K.A (2007). “Agriculture and the Doha Round”. Center for Global Development (CGD/IIE Brief). 2.

[16] Hertel T., B.M. Hoekman, & W. Martin. (2003). “Agricultural Negotiations in the Context of a Broader Round: A Developing Country Perspective”. Chapter 6, In (Eds. P.L. Kennedy and W.W. Koo) Agricultural Trade Policies in the New Millenium. Haworth Press, Inc. New York.

[17] Goldin I., Knudsen, O., & van der Mensbrugghe, D. (2003). “For Whom the Bell Tolls: Incomplete Trade Liberalization and Developing Countries”. Presented at the Trade Consortium, University of Calabria, University of California joint meeting, Capri, June 23-27, 2003.

[18] Anderson, K., & Lomborg, B. (2008). Free Trade, Free Labor, Free Growth. Project Syndicate.

[19] Anderson, K., & Martin, W. (2006). “Agriculture, Trade, and the Doha Agenda.” In (Eds. Kym Anderson and Will Martin) Agricultural Trade Reform and the Doha Development Agenda. London and Washington: Palgrave Macmillan and the World Bank. 13.

[20] Zartman, 622.

[21] Schott, J. J. (2004). “Reviving the Doha Round”. Peterson Institute for International Economics. Retrieved from http://www.iie.com/publications/papers/paper.cfm?ResearchID=207#3

[22] Ibid.

[23] Ibid. 6.

[24] Habeeb, W.M. (1988). Power and negotiation. In Power and Tactics in International Negotiation. John Hopkins Press. 17.

[25] Ibid. 19.

[26] European Commission (2013). “Agricultural trade in 2013: EU gains in commodity exports”. Agriculture and Rural Development Report. Retrieved from http://ec.europa.eu/agriculture/trade-analysis/map/2014-1_en.pdf

[27] Elliot, 3.

[28] Bow, B. (February 2016). POLI 3581 Lecture. Dalhousie University.

[29] Ackerman, F. (2005). The shrinking gains from trade: A critical assessment of Doha round projections. Global Development and Environment Institute, Working Paper No. 05-01. 5 (Table 1).

[30] Ibid.

[31] Wagner, R. H. (1988). Economic interdependence, bargaining power, and political influence. International Organization, Vol. 42(3). 464.

[32] Ibid.

[33] Putnam, R. D. (1988). Diplomacy and domestic politics: The logic of two-level games. International Organization, Vol. 42(3). 434.

[34] Griswold, D. (2007). “Should the United States cut its farm subsidies?” CATO Institute. Retrieved from http://www.cato.org/publications/commentary/should-united-states-cut-its-farm-subsidies

[35] Gawande, K. & Hoekman, B. (2006). “Lobbying and Agricultural Trade Policy in the United States”. World Bank Policy Research Working Paper 3819. 28.

[36] Ibid.

Exploring the agent/process problem in international negotiation

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s