International collective action problems of internationally excludable goods

In international relations, many issue-areas pose international collective action problems, where states have incentives to free ride on beneficial international outcomes by avoiding the costs associated with their production. The precise nature of the international collective action problem depends on the particular issue-area. Although open international markets are sometimes referred to as international public goods, states that free ride by adopting protectionist policies can be excluded from the market's benefits by restricting their trading opportunities. Because an international public good is non-excludable (and non-rival), the politics of open international markets are therefore the politics of an internationally excludable good.

International exclusion is necessary because of the domestic incentives that narrow economic interests create for state policymakers. According to Ricardian theories of comparative advantage, unilateral free trade allows states to maximize their aggregate consumption. However, trade openness has distributional implications, because import-competing sectors undergo adjustment costs associated with changes in relative prices. Narrow interests tend to be politically powerful compared with diffuse interests such as consumers, and have incentives to transfer income to themselves regardless of costs imposed on the domestic public good in the form of overall economic efficiency (Schattschneider Reference Schattschneider1935; Olson Reference Olson1965, Reference Olson1982, 44).Footnote 1 The problems of open international markets are therefore the problems of an issue-area that simultaneously concerns an excludable good in international terms and a public good at the domestic level.

While treaties by themselves are insufficient to explain the solution of international collective action problems because of states’ ability to act unilaterally and selectively contrary to treaty commitments (Carr Reference Carr1940; Weiler Reference Weiler1985), current scholarship provides two generalizable explanations for the provision and maintenance of open international markets.

The first, hegemonic stability, is modeled on the collective action theory concept of the ‘privileged group’, where a large state with sufficient size and self-interest unilaterally provides or organizes the international market as an international public good (Olson Reference Olson1965, 49–50; Olson and Zeckhauser Reference Olson and Zeckhauser.1966; Kindleberger Reference Kindleberger1973; Keohane Reference Keohane1980).Footnote 2 The second, specific (or contingent) reciprocity within a Prisoners’ Dilemma model, yields cooperative outcomes through repeated interaction, a low discount rate, monitoring, and trigger strategies of interstate exclusion (Axelrod Reference Axelrod1984; Keohane Reference Keohane1984, esp. 77–78 on interstate exclusion; and 54, 103, 211 on domestic interests underlying the international Prisoners’ Dilemma). Because export opportunities in this system are dependent on permitting imports, state policy is determined by the balance of, and struggle between, different narrow interests in light of international reciprocity incentives (Gilligan Reference Gilligan1997). On this logic, with international institutions providing monitoring and states themselves punishing free riders (Garrett Reference Garrett1992, 535), international institutionalist scholars argued that rational, egoist states could sustain cooperative equilibria through repeated interaction and trigger strategies.Footnote 3

The classic example of a trade regime operating on these principles was the General Agreements on Tariffs and Trade (GATT) (Keohane Reference Keohane2005, xi), now reconstituted, with a stronger dispute settlement system, as the World Trade Organization (WTO). Within the WTO, protectionist free riding is punished not only by reputational costs, but also by the ‘suspension of equivalent concessions’ (i.e. retaliatory trade sanctions) by other WTO members.Footnote 4 Variation in outcomes within WTO-style trade regimes is explained by, for example, differences in industry type, retaliatory potential, and domestic institutions (e.g. Bown and Blonigen Reference Bown and Blonigen2003; Davis and Shirato Reference Davis and Shirato.2007; Rickard Reference Rickard2010; C. Davis unpublished).

There are of course limits to the organization of international markets on a specific reciprocity basis, and these limits can provide a baseline against which outcomes from alternative approaches can be compared. First, as trade openness in this model is driven by retaliatory exclusion, the degree of potential punishment determines the depth of an international trade regime. Explanations for demanding adjustments therefore require punishment strategies – such as the ‘grim trigger’ strategy of permanent punishment for the smallest amount of free riding – of such severity and duration that that states may lack forward-looking incentives to apply them (e.g. Downs and Rocke Reference Downs and Rocke1995; Downs et al. Reference Downs and Rocke1996). Tit-for-tat strategies based on retaliatory exclusion approximately equivalent to the degree of free riding do not suffer from these credibility problems but in turn cannot sustain such demanding obligations. In addition, because punishment strategies only impact on state behavior to the degree that states incorporate future consequences in their current decision making, the depth of an open international market is limited by states’ discount rates.

Second, a related point, specific reciprocity regimes do not entirely escape the limits on international organization driven by states’ unequal size. The inability of small or poor states to effectively use dispute settlement mechanisms based on withdrawal of reciprocal trade concessions is frequently noted (e.g. Simma Reference Simma1994b, 233; Palmeter Reference Palmeter2000–01, 472–74; Lawrence Reference Lawrence2003, 7). In the absence of enforcement by ‘third-party’ states (Maggi Reference Maggi1999), specific reciprocity mechanisms do not provide effective means of ensuring an open market between large and small states.

A third limitation is that the operation of WTO-style retaliatory sanctions disrupts the international market (e.g. Charnovitz Reference Charnovitz2001, 810–11; Lawrence Reference Lawrence2003, 1–4). Prisoners’ Dilemmas ‘solved’ by tit-for-tat strategies between individuals may not be disrupted by defection and punishment, as the cooperative outcome may be stable. Trade agreements enforced by ‘suspension of equivalent concessions’ mechanisms, however, tend to involve considerable market disorder. Even with ex ante unanimity among states, as circumstances change, narrow economic interests facing adjustment costs may successfully demand selective protectionism even if export opportunities will be lost.Footnote 5 This in turn may prompt trading partners to withdraw ‘equivalent’ trading opportunities, further disturbing the international market for firms in both states. A WTO dispute between the EU and United States over trade in bananas, for example, led to increased tariffs on handbags, coffeemakers and batteries, while tariffs on cashmere and pecorino cheese were threatened but not ultimately imposed (Devereaux et al. Reference Devereaux and Lawrence2006, 129). Ex post retaliatory enforcement mechanisms are therefore double edged, because trade scholars find them both highly inefficient and hard to eliminate (e.g. Lawrence Reference Lawrence2003).

Open international markets without exclusion: encompassing domestic political institutions

It is impossible to overstate the centrality of the above two models in scholarly work on international organization. Both theoretical discussions and empirical investigations frequently rely on causal mechanisms derived from either the hegemonic stability/‘privileged group’ model or – even more so – from the specific reciprocity/‘Prisoners’ Dilemma’ model (e.g. Dai Reference Dai2005; McGillivray and Smith Reference McGillivray and Smith.2008; Stone et al. Reference Stone and Slantchev2008; Johns and Rosendorff Reference Johns and Rosendorff.2009; Rector Reference Rector2009). As Dai writes: ‘Since the mid-1980s work on international institutions has been largely influenced by one simple model: the repeated Prisoner's Dilemma’ (Dai Reference Dai2002, 405).

These explanations have attracted scholarly support for at least three related reasons. First, they are positivist and rationalist explanations. Second, they benefit from ‘intellectual arbitrage’ with the wider social sciences (Lake Reference Lake1993, 468; e.g. Ostrom Reference Ostrom1997). Third, they provide explanations for important phenomena, such as prominent international regimes and the relative openness of the global economy. However, there remain important scenarios that neither of these models is equipped to explain. These include deep international markets not driven by the availability of extreme (or implausible) interstate sanctions; international organization in costly issue-areas on an equal basis between large and small states; and international organization involving costs falling on a diversity of narrow interests, which does not require the operation of market-disrupting interstate sanctioning mechanisms.

This paper puts forward an alternative collective action theory-based explanation for open international markets without relaxing the assumption of rational, egoist states, coexisting in anarchy, whose constituents include a variety of rational, egoist interest groups facing a stream of costly adjustments from changes in relative prices. A common feature of the current pair of explanations is that their solution to the possibility of states’ free riding on open international markets is indirect. Each assumes that states will intermittently sacrifice domestic public goods such as economic efficiency or state reputation in order to provide private gains to narrow economic interests. Open markets are therefore maintained by incentives provided by the behavior of other states, particularly the practice of interstate exclusion following the logic of various solutions to Prisoners’ Dilemma problems.

However, the application of Prisoners’ Dilemma models to the politics of international trade is accurate only in certain circumstances. The Prisoners’ Dilemma model correctly captures incentives for protection and the possibilities of political deals for mutual reduction in tariff barriers. In so doing, it represents the politics of trade within the trading states at what might be called a ‘common sense’ level of aggregation, where making export opportunities contingent on import-openness is sufficient to ensure mutual tariff reduction. But decision making within states varies in its aggregation of winners and losers from trade policy according to the prevailing political system. It is not the issue-area of international organization – trade and its adjustment costs – which makes trade politics a Prisoners’ Dilemma, but the relative privileging of various interests in authoritative national decision-making arrangements. The variation in trade politics that can result from variation in domestic political systems is not theoretically limited to being slightly more or slightly less cooperative within a Prisoners’ Dilemma model. States’ preferences could equally well lead them to reject any lowering of trade barriers on particular interests even if contingent export opportunities were lost, or, even in the presence of adjustment costs, support unilateral trade openness regardless of export opportunities. In the latter case – this is the central point – such a state would, entirely for rational and self-interested reasons, be making unilateral contributions to the solution of international collective action problems, by accepting political costs that reduce the need for reciprocal organization of an open market.

For the individual in a Prisoners’ Dilemma, a cooperative solution can be achieved through repeated interaction and trigger strategies. This is the solution that ‘Axelrod–Keohane’ approaches adopt to explain international trade organization and its limits. For states placed in such a scenario through domestic incentives derived from narrow economic interests, however, changing those domestic incentives that give rise to such interstate behavior provides an alternative solution. This is not a new solution to the Prisoners’ Dilemma itself, but to the Prisoners’ Dilemma-type behavior of states where this is derived from the internal influence of powerful narrow interests. This possibility has perhaps been obscured by direct comparisons of Prisoners’ Dilemma situations between individuals, on the one hand, and between states, on the other (e.g. Keohane Reference Keohane1984, 94 (footnote), 106; Downs and Jones Reference Downs and Jones.2002, S95).

Not all organized groups face the same incentives to seek inefficient income transfers regardless of the costs to society as a whole. Groups representing large constituencies – ‘encompassing groups’ in collective action terminology – receive a substantial share of the gains from collective benefits such as overall economic growth, and suffer a substantial share of the costs of policies that contribute to overall inefficiencies (Olson Reference Olson1982, 48). Encompassing groups are both inclusive of a wide range of diverse constituents and possess the ability to authoritatively coordinate policy across issues as these apply to particular constituents (Jankowski Reference Jankowski1988). Cohesive large constituency organizations have powerful incentives to accept costs on particular constituents in order to provide public goods to their members as a whole, even without external sanctions. These encompassing incentives are inseparably part of the same theory of collective action as the incentives for narrow economic interest groups (as emphasized in Olson Reference Olson1986).

Outside of international relations, the most important literature on the explanatory power of encompassing institutions addresses labor market politics (Olson Reference Olson1982, 90–91), where the encompassing group concept provided microfoundations for the study of corporatist organizations. Where union federations authoritatively represent large numbers of workers, they have incentives to ‘self-regulate’ their behavior and accept costs on particular constituents, for example, by preventing strikes in certain firms or sectors (Garrett and Lange Reference Garrett and Lange.1985, 795; also, e.g. Schmitter Reference Schmitter1981; Calmfors and Driffill Reference Calmfors and Driffill.1988; Cameron Reference Cameron1988, esp. 575ff; Driffill Reference Driffill2006). They do this in order to provide public goods – such as low inflation and high economic growth – for their constituents as a whole.

Here, however, we are concerned with the political institutions within states that make trade policy decisions. Where political institutions are responsive only to narrow interests, policymakers may have incentives to adopt protectionist trade policies unless restrained by the behavior of other states. This could occur because political institutions are undemocratic, or because democratic political institutions are responsive to special interests or small groups of voters. However, where domestic political institutions are responsive to larger constituencies, of consumers and a wide range of firms with different interests, and possess the ability to authoritatively coordinate policy across diverse issues, policymakers have incentives to adopt free-trade policies.

Assuming that sufficient numbers of eligible voters vote, and that structural conditions allow voters to make their decision on the basis of public goods, rather than quid pro quo clientelistic exchange (Lyne Reference Lyne2008), then whether policymakers will have incentives to provide local/sectorally targeted or broad national public goods depends on the design of democratic institutions. Three institutional arrangements commonly associated with incentives to provide broad public goods are: parliamentary rather than separation-of-powers legislative institutions; proportional representation rather than first-past-the-post electoral systems; and unitary rather than federal state structures (Gerring and Thacker Reference Gerring and Thacker.2008).Footnote 6 Parliamentary legislative arrangements encourage the production of national public goods, because separation-of-powers systems offer more opportunities for powerful narrow interests to influence issue-specific policy outcomes than cohesive cabinet government (e.g. Bagehot Reference Bagehot1966 [1867]; Cox Reference Cox1987). Proportional representation (PR) electoral systems, particularly closed-list PR systems, encourage the production of national public goods because single-member districts and candidate-centered systems tend to privilege localist tendencies generally and a relatively small group of constituencies and their interests in particular (e.g. Mill Reference Mill1993 [1861]; Carey and Shugart Reference Carey and Shugart.1995; Heitshusen et al. Reference Heitshusen and Young2005). Finally, unitary state structures encourage the production of national public goods, because federal structures tend to provide incentives for territorially fragmented policymaking and outcomes (e.g. Pressman and Wildavsky Reference Pressman and Wildavsky.1973; Wibbels Reference Wibbels2000).

This scholarship provides a basis for measuring the level of ‘encompassingness’ in particular states and thus a point of departure for empirical tests of the relationship between encompassing domestic political institutions and outcomes in international organization.Footnote 7 Variation in these variables – democratic institutions, structural conditions that limit clientelistic voting, parliamentary regimes, unitary state structures, and closed-list PR – can be used to explain variation in specific reciprocity-related trade policy outcomes within international trade regimes, as well as variation in the use of specific reciprocity-related mechanisms across international markets, including, notably, the possibility of international markets that operate without any use of reciprocity mechanisms at all.

Recent research provides considerable evidence to support the claim that variation in trade policy behavior can be explained by variation in the level of inclusiveness and cross-issue coordination of domestic political institutions. For example: democratic GATT/WTO members elected by majoritarian rules and/or single-member districts are more likely to violate GATT/WTO agreements than those elected via proportional electoral rules or multi-member districts (Rickard Reference Rickard2010); centralized political parties with stable linkages to large groups of the electorate are more likely to choose freer trade policies (Hankla Reference Hankla2006); delegation of trade policy authority to presidents – that is, politicians with a national constituency – in Latin America is significantly related to trade liberalization (Nielson Reference Nielson2003); and cross-sector variation in trade barriers in 65 countries can be explained by the interaction between industries’ collective action problems and diverse electoral institutions (Kono Reference Kono2009). These findings serve to illustrate that the theoretical claim advanced here has traction in empirical research projects.

This application of the ‘encompassing group’ concept to the international collective action problems of open international markets builds on several scholarly literatures that have used similar causal mechanisms, even when not drawing explicitly on the ‘encompassing’ terminology from collective action theory. There is of course a considerable literature connecting international economic policies and other state behavior to domestic political institutions (e.g. Ruggie Reference Ruggie1982; Katzenstein Reference Katzenstein1985; Doyle Reference Doyle1986; Rogowski Reference Rogowski1987; Alt and Gilligan Reference Alt and Gilligan.1994; Brawley Reference Brawley1994; Bueno de Mesquita et al. Reference Bueno de Mesquita and Morrow1999; Kant Reference Kant2003 [1795]; Milner and Kubota Reference Milner and Kubota.2005). There are also recent contributions that have used domestic interest politics or institutions to explain variation in states’ willingness to accept costly adjustments, while remaining within the specific reciprocity/Prisoners’ Dilemma approach to international organization (e.g. Dai Reference Dai2005; McGillivray and Smith Reference McGillivray and Smith.2008; Johns and Rosendorff Reference Johns and Rosendorff.2009; Rickard Reference Rickard2010; C. Davis unpublished). The argument presented here has similarities with ‘joint product’ models of political outcomes (Broz Reference Broz1999), and, at the most abstract level, with the ‘state-society’ approach to international relations (also referred to as ‘preference-based’ or ‘liberal’), which emphasizes that variation in political and social organization across states determines the scope of international collective action problems (Moravcsik Reference Moravcsik1997; Lake and Powell Reference Lake and Powell1999).

While the existing literature on domestic political institutions in international political economy has tended to focus on at-the-margin effects on unilateral state behavior, such as trade openness, the effects that these sort of variations can have on international organization, particularly international organization between several states that each possess encompassing domestic political institutions, have not fully been appreciated. These are discussed in the next section.

International organization between states with encompassing domestic political institutions

The implications for international organization of states with encompassing domestic political institutions can be most clearly seen by comparison with outcomes based on specific reciprocity solutions within the Prisoners’ Dilemma model. As noted above, this approach contains several well-known limitations: the depth of a trade regime is limited by the availability of interstate punishment and states’ discount rates; small or poor states cannot effectively use reciprocity-based dispute settlement mechanisms against powerful trading partners; and the application of ex post trade sanctions as enforcement mechanisms damages the operation of international markets.

However, none of these limits apply in the same way to the relations between states with encompassing domestic political institutions. Consider a hypothetical comparison between two zero-tariff international markets. In one, narrow constituencies intermittently possess the political influence to create unilateral protectionist rules regardless of costs imposed on domestic public goods. In the other, internal political institutions can be characterized as encompassing in each state. Assume in both cases that the market produces a stream of changes in relative prices that impose severe adjustment costs on narrow economic interests. In both cases, there are powerful incentives to maintain an open international market. In the former scenario, however, demands for defection from narrow economic interests result in a stream of disruptive restrictions in trade opportunities, both protectionist and retaliatory, in a manner familiar to observers of the WTO regime. In the latter, the states with encompassing domestic political institutions would have incentives to consistently accept these adjustment costs, producing highly reliable interstate trading opportunities. These distinct international equilibria derive from variation in equilibria in domestic politics.Footnote 8

States with encompassing domestic political institutions also have incentives to submit to a continuous stream of costs on a diversity of narrow organized interests – certainly a deep international market – even without savage maximum or enduring interstate punishments, so long as the demands for protectionism derive only from narrow interests. Furthermore, such states have incentives to enjoy the benefits of trade openness in every period, and thus, again, the discount rate variable would not limit the depth of an international trade regime.

Similarly, where the benefits of restricting imports from a poor or weak state fall solely to a narrow economic interest in a large or powerful state, a state possessing encompassing domestic political institutions has incentives to impose the cost on the narrow interest in order to benefit its broader constituency, even if it has greater retaliatory potential than its trading partner. Thus, international organization between states with encompassing domestic political institutions can manage a greater degree of diversity in state size and power than international organization based on specific reciprocity between states with powerful narrow interest groups, and this result does not rely on any incentives derived from the behavior of ‘third-party’ states.

Even where national policymakers have incentives to prioritize these domestic public goods, this policy choice may strain their relationship with their diverse constituents, who would be obliged to pursue their interests under powerful constraints. Such states’ acceptance of a stream of significant changes in relative prices might appear routine and reliable when looked at from the international perspective, but the relationship between decision makers and their varied domestic constituents could nonetheless be as tense and difficult as the relationship between corporatist union federation leaders and firm-level workers in hard economic times. The organization of international markets outlined here is not derived from a universal harmony of interests, nor from any assumption that organizing an open international market is in any sense a ‘trivial’ activity, but from the specification of a mechanism for addressing the relevant collective action problems that differs from those emphasized in current scholarship.

Implications for research on international organization

International organization based on encompassing domestic political institutions has been presented here as a clear alternative to existing explanations. However, different mechanisms underlying international organization in costly issue-areas can be both substitutes and complements – as has long been recognized in the discussion of hegemonic stability and specific reciprocity mechanisms. Incentives to support open international markets emphasized by these different mechanisms may mutually reinforce each other or undermine each other's effects. This discussion therefore underlines the importance of incorporating measures of domestic political institutions in the investigation of specific reciprocity trade regime outcomes, as the work of many scholars explaining variation in outcomes within the WTO now demonstrates. Indeed, the most common effect of this variation in domestic institutions may be to alter the severity of interstate Prisoners’ Dilemma problems, rather than to eliminate them entirely.

One distinctive possibility derived from integration of international reciprocity and domestic institutional incentives would be a treaty regime where the essential part of the bargain would be a mutual rejection of specific reciprocity mechanisms. Such a regime might entirely eliminate the use of ‘suspension of equivalent concessions’ enforcement mechanisms (with all their inequalities and market-disrupting disadvantages) in favor of a mutual commitment to complete-and-without-exception compliance with regime obligations. Because this regime would be accepted on a mutual basis, the outcome would rest in part on reciprocity/retaliation incentives: this system of attenuated or diffuse reciprocity would rest not on ‘norms of obligation’, but on incentives to avoid triggering trading partners’ reversion to WTO-style reciprocity behaviors. At the same time, this outcome would also require sufficiently encompassing domestic political institutions to provide incentives for national policymakers to behave in ways compatible with the international bargain. As this discussion shows, combinations of international reciprocity and domestic institutional incentives may potentially vary the severity of Prisoners’ Dilemma problems between states in qualitatively important ways that go beyond at-the-margin variation in outcomes within a WTO-style regime.

There is of course much more to the politics of open international markets than the mechanisms that underlie the solution to international collective action problems derived from powerful narrow interests. Other important factors include: the design of international institutions, including monitoring mechanisms (e.g. Koremenos et al. Reference Koremenos and Lipson2001; Dai Reference Dai2002); the legalization of international politics (e.g. Goldstein et al. Reference Goldstein and Kahler2001); the relationship of trade and compensation systems (e.g. Cameron Reference Cameron1978; Katzenstein Reference Katzenstein1985); and concerns over ‘relative gains’ (e.g. Stein Reference Stein1984; Grieco Reference Grieco1988). Much of the current scholarship on these topics rests, implicitly or explicitly, on hegemonic stability or specific reciprocity mechanisms (as well as ‘coordination game’ incentives; e.g. Stein Reference Stein1982), but these and similar topics can equally be addressed in relation to states with encompassing domestic institutions. Three possible implications of the encompassing incentives model for the study of international organization are worth briefly mentioning.

A first implication is that states have incentives to choose trading partners on the basis of their domestic political structure. In comparative terms, one of the distinctions of international organization based on WTO-style reciprocity is that it constitutes a ‘come one, come all’ macrostructure that facilitates international organization between states with a variety of domestic political structures, so that states can participate without much prior scrutiny beyond a suitable exchange of trading opportunities (to be sure, other issues may interfere). On the other hand, where states seek to mutually agree to eliminate the use of ex post retaliation mechanisms from a trade regime, they will have incentives to screen potential participants on the basis of apparently non-trade-related domestic political institutions (e.g. Koremenos et al. Reference Koremenos and Lipson2001 on screening). After all, the reliability of exchange in such a system is dependent on domestic political arrangements. The choice of international rules in particular issue-areas, and the way in which states approach screening of possible adherents to trade regimes, may therefore interact significantly with the domestic political structure of participating states.

This discussion also suggests the possibility of unexpected levels of state obedience to international obligations. International relations scholarship generally finds that the effectiveness of international treaties is derivative of whether states have forward-looking self-interested incentives to fulfill their obligations. Reinterpretations of commitments by institutions addressing ‘incomplete contracts’ may therefore lead to treaty violations, if the outcome would impose costs on powerful narrow interests (e.g. Milgrom and Roberts Reference Milgrom and Roberts.1992, 129–40).Footnote 9 If, however, the trade policy of a state is influenced by sufficiently encompassing domestic institutions, then policymakers may have incentives to accept costly reinterpretations of commitments, even if clearly outside original agreements, where the new policies benefit the larger domestic constituency. Trade agreements designed to address narrow sectors can therefore become de facto wider-ranging regimes. This is a rather startling outcome for international relations scholarship, which frequently emphasizes that states’ egoist behavior is responsible for the fragility of treaty agreements, but it is equally the result of the rational, egoist decision making of states in anarchy. Note that this explanation for ‘over-obedience’ to treaty obligations does not rely on ‘logics of appropriateness’, the ‘legitimacy’ of international law, the ability of law to ‘mask’ distributional conflicts, or the influence of epistemic communities (Haas Reference Haas1989; March and Olsen Reference March and Olsen.1989; Franck Reference Franck1990; Burley [Slaughter] and Mattli Reference Burley [Slaughter] and Mattli.1993).

Third, this approach to international trade organization may also contribute to the understanding of international organization in other issue-areas. When considering its broader applicability, it is important to note that although this model may, in the trade issue-area, solve collective action problems of an internationally excludable good, it does not rely on international exclusion as its mechanism. This model therefore provides a potential explanation for international organization even where mechanisms of interstate exclusion do not provide an alternative solution.

As a result, this approach to international organization may be relevant to the politics of international human rights regimes. Although states can have an interest in the protection of human rights within other states, a widely accepted distinction of international human rights regimes is the difficulty and sometimes illogic of the enforcement of human rights obligations by the threat of reciprocal measures or interstate action (e.g. Simma Reference Simma1985, 129–35; Reference Simma1994b, 364–65; Moravcsik Reference Moravcsik2000, 217; von Stein 2006, 24, 52). For example, there is little incentive for a state to ‘reciprocally’ reduce its own observance of human rights in response to other states’ human rights violations. It has therefore been difficult to incorporate international human rights regimes within ‘Axelrod–Keohane’ or ‘tit-for-tat’ approaches to international organization. Where human rights regimes impose a stream of costs on narrow interests to the benefit of diffuse interests, however, states with sufficiently encompassing institutions would have incentives to accept these adjustments, either unilaterally or on the basis of an attenuated form of reciprocity, even while respecting the understanding that human rights obligations are not normally enforced by interstate or reciprocity-based sanctioning mechanisms.

States with encompassing domestic political institutions and ‘self-contained regimes’ in international law

Encompassing incentives for state behavior may also be relevant to important theoretical discussions in international law. General international law authorizes state use of a wide variety of countermeasures against states responsible for an internationally wrongful act. International law is therefore traditionally characterized as ‘bilateral-minded’, and even multilateral obligations are sometimes viewed as enforced only through bilateral mechanisms (e.g. Simma Reference Simma1994b, 230, cf. Third Report on State Responsibility, Yearbook of International Law Commission, 1982, vol. 2, Part 1, 38; Goldsmith and Posner Reference Goldsmith and Posner.2005, 88). Legal scholarship is sometimes ambivalent about the role of countermeasures as international law's enforcement mechanism, viewing them as both ‘prone to abuse’ and nonetheless ‘a necessity’ (e.g. Simma Reference Simma1994a).

In 1980, however, the International Court of Justice (ICJ) decided that, where it was alleged that wrongful acts had been committed by another state's diplomatic representative, no resort could be had to any of the remedies provided for by general international law – that is countermeasures at the choice of the injured state. Rather the court argued that, in these circumstances, the sole remedy available to the hosting state was to notify the diplomat's home state that they were persona non grata and to require their recall. The ICJ's decision stated:

Prompted by in the ICJ's reasoning, Simma influentially characterized a self-contained regime as one where states would follow the rules established by a treaty regime but where the application of public international law's normal mechanisms at the disposal of an injured party – interstate countermeasures – would be more or less totally excluded (e.g. Simma Reference Simma1985; Simma and Pulkowski Reference Simma and Pulkowski.2006). On this conceptualization, the WTO regime, although a highly developed international legal system, does not constitute a self-contained regime because the use of countermeasures is organized and authorized by the WTO's dispute settlement understanding (Perez-Aznar Reference Perez-Aznar2006, 9).

While at times referring to international relations scholarship on regimes (e.g. Simma and Pulkowski Reference Simma and Pulkowski.2006, 493, footnote 37), these legal scholars do not anchor their discussion of the possibility of a self-contained regime in any simple political model of interstate relations. Perhaps, this is because dominant approaches to regime theory make interstate sanctioning central to their understandings of international organization. International organization in costly issue-areas between states with encompassing domestic political institutions could, however, explain just such a ‘self-contained’ result. A countermeasure-based multilateral regime, like the WTO, can become a multilateral self-contained regime, operating without bilateral countermeasures, in the presence of sufficiently encompassing domestic political institutions in the participating states. To be sure, this outcome is derived from self-interest in the participating states, not from any sense of ‘community interest’ (Simma Reference Simma1994b), but the result is similar to that which international law scholars seek to understand.