Rewarding in International Law: It's About Time

Anne van Aaken and Betül Simsek's article represents a significant contribution to the literature on international legal compliance.¹ It pushes forward our understanding of the role that positive incentives play in its promotion, usefully highlights potential tradeoffs between positive and negative incentives, and identifies ways in which rewards and penalties may be employed together. While the authors suggest that a richer understanding of rewarding is useful for questions of institutional design, they purposefully focus on the compliance-side of the coin. This essay builds on their theoretical and conceptual ground-clearing to consider rewarding's implications for treaty negotiations and design. In doing so, it focuses on the role that time implicitly plays in the authors’ analysis and argues that assumptions about time horizons inflect all design calculations. For this reason, understandings of temporal dynamics should be foregrounded in both academic and policy realms.

This essay first identifies several general challenges for the inclusion of rewards in international agreements and then discusses design choices specific to the conditions under which rewarding is most likely to be effective. It concludes by reflecting on the implications of taking time seriously for the design and use of rewards as a means to strengthen international cooperation.

Time, Temporality, & Institutional Design

Governments develop, join, and implement treaties in a process that is complex, multi-staged, and often quite lengthy. These drawn-out processes can take decades, involving both incremental improvements in compliance and backsliding, with every stage shaping and being shaped by changing distributions of power, preferences, and values. Yet few studies on international law's impact theorize or explicitly model time in the compliance process, much less address how temporality assumptions factor into design considerations.² This is not to claim that time or temporal dynamics are not dealt with at all. Over two decades ago, Rosalyn Higgins reflected on several problems the concept of time posed for international legal jurisprudence.³ And theoretical and empirical approaches to institutional design and compliance do vary in the extent to which they account for time or address these dynamics. Rather, the claim is that temporal assumptions are rarely articulated as such and we have yet to assess how they inform international legal scholarship as a whole.

This is surprising given that the way in which scholars understand and represent time fundamentally shapes their claims about the effects of international law on state behavior. At the very least, every theoretical approach presumes that treaty negotiators operate with certain views on the length of their own and targeted actors’ time horizons.⁴ Designers also rely on a set of assumptions about the temporal scope of various causal processes set in motion by an international agreement. Understanding that this is intrinsic to international politics more generally, a growing literature in international relations (IR) has turned its attention to the role of time and temporal assumptions implicit in the field's major theoretical paradigms.⁵ These include the rationalist approaches from which van Aaken and Simsek draw in developing their argument, with implications for questions of both institutional design and rewarding as a compliance mechanism.

Rationalist approaches incorporate temporality in a limited manner, as they tend to focus on how discrete treaties generate stable equilibria rather than viewing institutional design or its effects as dynamic processes. While this enables scholars to derive concrete, testable, and generalizable hypotheses, process-based approaches such as historical institutionalism are more attuned to temporal dynamics and the importance of contingency and sequencing.⁶ Yet even from a rational design perspective, implicit assumptions about actors’ time horizons and the dynamics of causal processes impact considerations central to the creation of successful rewarding mechanisms. To be sure, van Aaken and Simsek indirectly point to the important role of time in distinguishing rewards from penalties. Rewards refer to anything that improves the targeted country's “value position” relative to a baseline of expectations. They argue that the reward/penalty distinction is contingent on this baseline, which depends on actors’ expectations about the future and can “shift over time.”⁷ Removal of a reward will be seen as a penalty if it has become over time part of that baseline of expectations, while removal of a long-standing penalty effectively represents rewarding.

Foregrounding time and temporality raises a number of questions for institutional designers that warrant further consideration. How do negotiators making design choices understand time? What assumptions do they hold about the temporal scope of key causal processes and mechanisms, namely those sparked by treaty provisions designed to enhance compliance, such as rewards? Can one make design choices without arbitrating between different actors’ time horizons, varying understandings of the duration of causal processes, or divergent cultural conceptions of time? In terms of rewarding, can negotiators draft treaties that effectively anticipate how temporal processes will change targeted actors’ baseline expectations over the short- and/or long-terms? For example, if states begin to view rewards “after some period of time” as entitlements (reducing their effectiveness as such),⁸ what is the temporal scope for this process and can negotiators design rewards to prevent or adapt to such an occurrence? The following section considers, albeit non-exhaustively, these issues in the context of the design of international rewards.

Designing Effective Rewards

For a number of reasons, it is difficult to envision how certain types of rewards could be intentionally or explicitly designed ex ante. Of course, certain types of penalties pose the same obstacles and thus the issue is not unique to rewarding. While this does not detract from van Aaken and Simsek's argument as such, it is worth unpacking when considering institutional design. The authors distinguish internal rewards—benefits derived from the treaty bargain itself—from external rewards—those in addition to the treaty-based benefits. It seems fairly straightforward to incorporate most internal rewards into the substantive terms of an agreement. For example, the provision of epidemiological and statistical services found in the Constitution of the World Health Organization,⁹ or the Montreal Protocol's conditioning of trading privileges on membership and its creation of a multilateral fund for technical and financial assistance.¹⁰ In contrast, some external rewards prove challenging to build into regime design without converting them into internal rewards, while other types could plausibly be designed in such a way that retains their external nature.

External rewards that appear difficult to incorporate explicitly include rewarding through side payments not part of the treaty bargain, certain reputational rewards, and external rewards provided by non-state third parties. For instance, designing an agreement that guarantees tangible external rewards supplied by private, non-party actors seems a challenging task, although there may be ways to indirectly incentivize their provision. To develop a hypothetical discussed by the authors,¹¹ a protocol to the Convention on International Trade in Endangered Species of Wild Fauna and Flora (CITES)¹² could be designed that provides a framework and incentives for member states to develop public-private partnerships with the tourism industry (something they already do independently of the Convention). Under the auspices of CITES, these partnerships would be required to manage parks consistent with Convention obligations, for which they could receive certification or public praise, thereby increasing the probability that states accrue higher income from park tourism. The reward—boosted tourism and income—remains external, in that it is not part of the negotiated treaty benefit, but its provision is encouraged by treaty design. In terms of temporal dynamics, this suggests that the success of certain types of rewards will depend on institutional effects that may or may not transpire after a treaty becomes operational.

To give another example, treaty drafters could explicitly mandate international civil servants or experts—such as human rights treaty bodies—to praise states with good compliance records as one means to generate the (potential) external reward of a positive reputation.¹³ Doing so would certainly complicate treaty drafting, as the reward's credibility requires institutional mechanisms to ensure that third-party agents delegated the task of praising follow their mandate. Research demonstrates that the extent to which treaty bodies engage in praising has varied considerably across committees and over time, which suggests that reward effectiveness will be shaped by developments in regime practices over time.¹⁴

More generally, we know that international institutions and regime practices frequently develop in ways that are unintended and often suboptimal, due to path dependence, agency slack, or bureaucratic pathologies.¹⁵ Rational design scholarship tells us that high levels of uncertainty about such “future states of the world” push states to negotiate greater treaty flexibilities through, inter alia, imprecise language, finite durations, and renegotiation provisions.¹⁶ To minimize these concerns, then, some types of rewards will be more effective if treaties incorporate flexibility mechanisms that enhance learning over time and make recalibration possible, such as “trial-and-error experiments”¹⁷ combined with procedures to redesign institutional practices. To return to the example of human rights treaty bodies, negotiators could draft provisions that permit states to reshape emergent institutional practices in order to correct deviant or non-use of praising by committee members.

Enhancing rewarding through institutional design further involves a delicate balancing of—and sometimes tradeoff between—flexibility and precision. One way to avoid the need for greater oversight and course-correction mechanisms, for instance, would be to design precise evaluation standards and to delineate the conditions under which third-party agents bestow a reputational reward. The accreditation process for National Human Rights Institutions (NHRIs) carried out by a Subcommittee of the Global Alliance of National Human Rights Institutions illustrates this approach. The Subcommittee—composed of NHRI representatives—grades NHRIs according to their compliance with standards elaborated in the Paris Principles.¹⁸ Notably, these principles are unique in terms of soft law instruments, in that they were drafted almost exclusively by independent bodies (NHRIs) and not state actors.

As for conditions conducive to successful rewarding, three raise interesting design issues when considered together. First, effective rewards must be designed to align with what targeted countries value and need at a given moment in time. This suggests that compliance rewards—those offered to incentivize states to bring their practices (back into) line with treaty obligations—will be more difficult to explicitly incorporate into a treaty compared with entry rewards—those promised for ratifying a treaty. The former are more likely to be employed in the distant future, at a time when targeted actors’ values may differ from those known at the time of treaty design. In other words, estimations of value shift considerably over time. Assuming that actors have fixed needs and values seems highly implausible and limits rewarding's effectiveness. From a rational design perspective, the contingency of value on some unknown point in time—about which there is considerable uncertainty—will lead states to design more flexible treaties. That is, negotiators will draft vague (imprecise) compliance rewards that can be calibrated to a particular moment in time (when compliance is achieved). Such flexibility would also permit customization to the targeted actor, given that “for some national leaders tangible rewards are of great value” while “for others the need for symbols of high status [is] more important.”¹⁹

There are, however, downsides to greater flexibility in hard law agreements. Renegotiation of treaty terms is costly and creates “hold up” opportunities; as more states join, revising institutions becomes even more difficult.²⁰ The possibility of changing the terms of a treaty bargain may further reduce the credibility of the promised reward, a second condition conducive to rewarding's effectiveness. And flexibilities such as imprecision are in tension with a third condition for success: the communication of explicit, clear, and specific rewards. To be effective, treaty language stipulating rewards must exhibit high levels of precision, which reduces one form of agreement flexibility. Vaguely specified rewards may increase the ability to promise something a state values at the time of compliance, but the authors suggest they are less effective because they “may lead the target to conclude that the promisor has little power to deliver.”²¹ Further exacerbating the situation, clear specification of a reward may even hinder cooperation. IR scholars drawing from research on time horizons in psychology and behavioral economics have demonstrated that forcing negotiators to focus on “exacting questions of institutional design” and “nettlesome details” actually thwarts reaching agreement in the first place.²²

Conclusion

This essay has suggested that foregrounding issues of contingency and temporal dynamics helps scholars better understand processes central to the politics and practice of international law and the institutional design of rewards. Yet surprisingly, politics are noticeably absent from van Aaken and Simsek's account. One unaddressed issue thus concerns the role that relative state and non-state power plays in their argument, and how power asymmetries and the relative influence of different international institutions might change the design calculations considered here. As noted by Erik Voeten, “it has become increasingly obvious that we do not live in a world where institutions were designed to best solve strategic cooperation problems. The design of institutions is deeply political, and so are the challenges to institutions.”²³ To gain a fuller understanding of how questions of power and distributive politics interact with institutional design requires explicit consideration of time and temporal dynamics. This could entail combining the authors’ rationalist and behavioralist analysis of rewarding with historical institutionalism's emphasis on temporal processes, contingency, and path dependent phenomena.²⁴ Mark Copelovitch and Tonya Putnam, for example, have examined how pre-existing contexts change rational design expectations for coping with uncertainty.²⁵ Drawing collectively from rationalist, historical institutionalist, behavioral, and constructivist approaches, Jean Galbraith shows how temporary regimes can have a significant influence on the design of permanent institutions.²⁶

Formal institutions or organizations created by treaty-based international law are sticky and often struggle to adapt to changing relative power, actor preferences, or values that shape what is perceived as a reward and what stings like a penalty. If we take seriously issues of distributive politics and temporal dynamics when thinking about how to design effective rewards, soft law or hybrid arrangements that incorporate mechanisms of experimental governance, learning, and recalibration will be necessary in many contexts.²⁷ To be sure, this will not remove power asymmetries from the equation, but it will enable both scholars and policy-makers to grapple more fully with their consequences. In a world characterized by shifting distributions of power and sites of political contestation, the challenge is well worth the reward.