Regional conflict and war are persistent problems of international politics. These problems have greatly intensified since the end of the Cold War. The wars that revolved around the Democratic Republic of Congo and the clashes between India and Pakistan are some prominent examples of this trend. At the same time, economic regionalism is proliferating, with regional integration arrangements (RIAs) such as the North American Free Trade Agreement (NAFTA) and the Common Market of the South (MERCOSUR); and deepening, as many existing RIAs, such as the European Union (EU) and the Economic Community of West African States (ECOWAS) have greatly expanded their activities. Thus, RIAs have become a prominent feature of contemporary international politics. Mitigating violent conflict is an important goal of such organizations.
Even a cursory look suggests that these institutions vary a great deal.1
Some RIAs, such as the EU, cover a wide range of issue-areas, while others, such as the European Free Trade Association (EFTA), have rather limited aims. Many RIAs institute annual meetings of top-level policymakers, while others, such as NAFTA, do not. Finally, the members of some RIAs, the Association of Southeast Asian Nations (ASEAN) for instance, largely realize their signed agreements; while others, such as the Mano River Union (MRU), sign many more agreements than they carry out. This variation raises an important question: which institutional features of these RIAs actually help to mitigate conflict? Realists largely dismiss international institutions.2 The “regime theory” of the 1980s and the more recent Kantian peace research program promoted the idea that international institutions matter, but did not look inside these institutions to draw out specific features that make the difference.3See Keohane 1984; Mansfield and Pevehouse 2000; Russett and Oneal 2001; and Russett, Oneal, and Davis 1998. For two recent exceptions, see Bearce and Omori 2005; and Boehmer, Gartzke, and Nordstrom 2004.
Koremenos, Lipson, and Snidal 2001.
Expanding on recent theoretical insights that underscore the importance of information and communication in international politics, I argue that cooperation on a wide array of economic issues, as well as regular face-to-face meetings among high-level officials, mitigate violent conflict. To test these arguments, I employ an original data set that measures variation in institutional design and implementation across twenty-five RIAs in recent decades.5
For a list of these twenty-five RIAs, see the Appendix.
Mansfield, Pevehouse, and Bearce 1999/2000 find that RIAs reduce conflict. Mansfield and Pevehouse 2000, on the other hand, find that RIAs mitigate conflict only when they produce substantial trade flows. Similarly, Russett and Oneal 2001; and Russett, Oneal, and Davis 1998 find that the density of IO membership reduces conflict. In contrast, Boehmer, Gartzke, and Nordstrom 2004; and Mansfield and Pevehouse 2000 find no such effect. Gartzke, Li, and Boehmer 2001 find that shared membership in IOs increases the likelihood of violent conflict.
This article proceeds as follows. The next section examines manners by which variation in information affects the likelihood of violent conflict and links them to specific institutional features. The second section elaborates on the empirical scope of this study and discusses issues of research design. The third section reports the results of the empirical analysis. The final section concludes.
Information, Institutional Design, and Conflict
A central claim of both neoliberal institutionalism and the rationalist theory of war is that variation in the amount and quality of information has important implications for international politics.7
The latter theory, in particular, indicates that information and communication can mitigate interstate conflict. Starting with this key insight and extending its underlying logic to institutional variation across RIAs, I contend that two institutional features—labeled the scope of economic activity and regular meetings of high-level officials—are instrumental in reducing violent conflict between member-states.8Due to space constraints, my arguments are not accompanied by empirical examples. For numerous illustrations and an in-depth case study of this theoretical framework, see Haftel 2004b.
Economic Interdependence and the Scope of Economic Activity
Using Fearon's rationalist theory of war, several recent studies argue that economic interdependence can mitigate violent conflict.9
According to these studies, higher interdependence provides “a richer set of costly signals and so more efficient signaling of unobservable resolve.”10Morrow 1999, 187.
A detailed operationalization of this concept is provided below. Gartzke, Li, and Boehmer 2001; and Nye 1971 also point to the potential pacifying effect of cooperation on economic issues that go beyond trade liberalization.
First, the logic of the signaling argument is grounded in the notion that interrupting commercial relations is costly to the states involved. That is, states are assumed to be vulnerable—rather than sensitive—to the disruption of commerce, in the sense that they may find it difficult to substitute it.12
Keohane and Nye 1977.
Second, skeptics point out that international trade affects private actors and not the governments that make decisions regarding peace and war.13
Cooperation and integration in some of the issue-areas subsumed by the scope of economic activity, on the other hand, provide benefits to the governments themselves (in addition to private actors). For example, a monetary union is conducive to the stability of the currency and monetary reserves. Thus, governments that abandon a monetary union risk losing monetary assets as well as damaging the credibility of their monetary policy.14 Similarly, the improved bargaining power that RIAs sometimes provide enhances the prestige of policymakers themselves. Thus, pulling out of such regional forums is a costly action that demonstrates one's resolve.In sum, economic cooperation through international institutions provides states with a valuable informational instrument regarding their national interest and resolve. This additional information is conducive to a more peaceful bargaining process and thus less violence. We can therefore hypothesize that the wider the scope of economic activity covered by an RIA, the lower the level of intraregional violent conflict.
Diplomacy and Regular Meetings of High-Level Officials
Many RIAs institute regular face-to-face meetings of officials at the highest levels, which usually operate as the highest decision-making body. While the formal agendas of these meetings often emphasize economic cooperation, they allow senior policymakers the opportunity to discuss outstanding issues directly and openly. This face-to-face interaction, I contend, is conducive to peaceful resolution of disputes between member-states. My argument rests on three building blocks: first, that diplomacy is a useful way to exchange important information about interests and resolve; second, that face-to-face meetings amplify the pacifying effect of diplomacy; and third, that regional forums have unique characteristics—compared to bilateral meetings—that render them effective instruments of conflict reduction. I discuss these assertions in turn.
Diplomacy is a widely used form of statecraft practiced by states in their daily interaction with other states. Recent studies that employ a rationalist framework demonstrate that diplomacy can mitigate violent conflict.15
Sartori, in particular, shows that when a state interacts with several states over time and shares with them some common interests, it has a strong incentive to acquire a reputation of honesty. Thus, diplomacy can provide valuable information and prevent the escalation of intraregional disputes. States can, of course, practice diplomacy indirectly through ambassadors, letters, phone conversations, and the like. There are some additional benefits to face-to-face meetings that cannot be obtained through indirect interaction, however. Experimental research of social dilemmas demonstrates that direct communication provides the opportunity for individuals to offer and extract promises of cooperation, which in turn widens the bargaining range and improves the prospects of compromise.16Ostrom and Walker 1991, 287.
Presumably, policymakers can promote interpersonal trust and exchange reliable information through bilateral, rather than regional, diplomacy. What, then, is the value added of a regional—as opposed to bilateral—forum? Regional meetings of high-level officials have two distinct advantages. First, a regional setting allows high-level officials to negotiate outstanding issues in an informal and less demanding atmosphere, compared to bilateral negotiations or formal arbitration by extra-regional bodies. Discussion of such issues in the framework of a regional summit that deals with other issues provides the different parties with valuable flexibility. Second, a regional forum provides members that are not directly involved in a conflict with an opportunity to provide a role of third-party mediators or honest brokers. These policymakers can provide additional information to the two parties and expand the bargaining range by introducing creative solutions to outstanding problems.
Considering the arguments regarding the pacifying effect of diplomacy in general and regional face-to-face meetings in particular, we can hypothesize, then, that regular meetings of high-level officials result in a lower level of intraregional violent conflict.
Research Design
The institutional features discussed above are attributes of RIAs, which is the unit of analysis. RIAs are IOs that promote economic policy cooperation among their members.18
Mansfield and Milner 1999, 591.
See Pevehouse, Nordstrom, and Warnke 2004; Mansfield and Pevehouse 2000; Page 2000; Smith 2000; and Union of International Associations (UIA) various years.
Twenty-five RIAs that existed in recent decades correspond to the above definition.20
For more details on these RIAs, see Haftel 2004b.
Due to the nature of the main explanatory variables, I conduct the empirical analysis at the regional level of analysis, with a region being defined by organizational membership. This setup expands on other quantitative analyses that link IOs and conflict, which commonly employ a dyadic setup.22
Although the dyadic level of analysis has provided numerous insights into the sources of violent conflict, some types of interactions are not adequately captured by a dyadic setup and are more amenable to a regional one. Scope of economic activity and meetings of high-level officials are, of course, organizational traits of the RIA and not of any particular dyad. The dependent variable, violent conflict, is also not necessarily a dyadic phenomenon. Recent studies indicate that security relations, regime types, and the level of economic interdependence also tend to cluster geographically, which in turn suggests that regional dynamics are at work.23The definition of particular regions according to RIA membership emanates from my attempt to evaluate the effect of these organizations and their specific institutional features on conflict in the region. While there is a lack of consensus on how to define a region and on the identification of specific regions, identifying regions by organizational ties is one conventional way to proceed.24
In addition, as pointed out above, my sample covers most regions in the world, which in turn reduces the risk of selection bias.Estimation Technique and Endogeneity
The dependent variable is a count of intraregional violent disputes and is characterized by a Poisson distribution. Therefore, an event count model is used for estimation. In addition, the significance of the goodness-of-fit parameters in the statistical models below indicates that a negative binomial regression model (NBRM) is appropriate. This model assumes a Poisson distribution but allows a conditional variance that is greater than the conditional mean. Finally, the data is arranged in a panel setup. I employ a random-effects count model to account for cross-sectional unobservable contextual heterogeneity.25
Panel data refers to data sets that are cross-sectionally dominated. My data set contains twenty-five sections (RIAs) and up to four time points. The command xtnbreg in Stata 8 is used.
I address endogeneity concerns, raised most prominently by realists, in several ways. First, the variables that capture the institutional features—as well as the rest of the independent variables—are lagged behind the dependent variable. Second, I control for the level of similarity of interests among member-states. If indeed such prior affinity determines the level of conflict, the effect of international institutions (to the extent that there is one) should disappear. Finally, I reverse the causal arrow to examine the effect of interstate conflict on the institutional variables. A negative effect of conflict on international cooperation through institutions should increase the risk of endogeneity and vice versa.26
I discuss the setup of this test in more detail below. Taken together, these three methods provide a good indication of the severity of the endogeneity problem.Dependent Variable
The dependent variable is intraregional violent conflict. As for other variables, I aggregate the number of disputes over a five-year period. Because violent disputes are rather rare, such aggregation provides meaningful variation on this variable. To minimize the risk of endogeneity, the five-year periods succeed the year in which the main independent variables are measured. For each RIA, I count all of the intraregional dyadic disputes in all years in which they took place. I do not count disputes between members and nonmembers. In infrequent cases where a dispute involves several participants that are both members and nonmembers, I count only dyads in which both participants are members. To operationalize this variable, labeled MID, I use the conventional Militarized Interstate Disputes (MIDs) data set.27
Independent Variables
scope of economic activity. I define this variable as the economic issue-areas covered by the RIA. To operationalize it, I collected information on twenty-seven indicators in nine categories that capture the range of economic activities that RIAs commonly address. A summary of these categories and indicators is reported in Table A1 in the Appendix.28
Haftel 2004b further elaborates on these categories and indicators.
Balassa 1961.
Page 2000.
The remaining three categories refer to several issue-areas that are excluded from Balassa's conventional typology. Nonetheless, these are important issues that correspond to key goals of past and present RIAs. Because there is little systematic analysis of these issue-areas in the context of regional integration, classification of these issues is not straightforward. I used one previous study31
Ibid.
Each and every indicator is measured along two dimensions. The first—which I label institutional design—refers to approved agreements that specify the mandate and institutional structure of any RIA. This is a necessary and important dimension of any institution. Each indicator can obtain a value of 1, if present, and 0 if absent. Thus, the designed scope of economic activity can range from 0 to 27. In order to code this dimension, I examined actual agreements signed by member-states, the Yearbook of International Organizations, and several studies that survey and compare a large number of RIAs.32
See IMF 1994; OECD 1993; UNCTAD 1996; Page 2000; and UIA various years.
The second dimension—which I label implementation—involves the actual steps that member-states take to realize the agreements they have reached. It captures the notion that international cooperation through institutions involves not only agreements and rules but also behavior in accordance with them. As such, it is a key component of international institutions.33
Martin 2000.
The collection of comparable data on implementation is more difficult than on institutional design and is rarely done. As a first cut into this institutional dimension, I created an ordinal scale: 0, 0.5, and 1. If the RIA did not implement the agreement regarding a specific indicator, or if implementation is low, the RIA scores 0. If implementation is complete, or nearly complete, the RIA scores 1. If implementation is partial, the RIA scores 0.5. For example, many RIAs reduce barriers to trade and factors of movement in a gradual process. If they have completed only some of these steps, I code the degree of implementation as partial. To determine the degree of implementation, I used a variety of independent secondary sources, such as in-depth case studies that evaluate the RIAs under scrutiny, newspaper articles, and other reports. In addition, to reduce the risk of measurement errors, I compared several such sources before I determined the value of each indicator.
To produce the variable that captures the scope of economic activity—which I label economic scope—I combine institutional design and implementation by multiplying the values on each dimension. Member-states may implement all the planned activities; thus economic scope ranges from 0 to 27. If the agreements are not fully implemented, the value on this variable will be lower than the one on institutional design alone. To see that, imagine an RIA that has an agreement on twenty indicators, that is, institutional design equals 20. Of these, ten indicators are fully implemented, four are partly implemented, and six remain only on paper. The value on economic scope is calculated as follows: (10 × 1) + (4 × 0.5) + (6 × 0) = 12. Figure 1 presents the average level of the implemented economic scope for the RIAs included in the sample.
The scope of economic activity for twenty-five RIAs: Average 1982–97
high-level officials. Because the decision-making process is essential to the sustainability of the organization, all RIAs provide for regular meetings of a decision-making body, typically every six months or year. The level of the decision-making body varies across RIAs, however. In some RIAs, such as ASEAN and MERCOSUR, decision-making authority lies in the hands of heads of state or foreign ministers. In others, such as the Central American Common Market (CACM) before 1990 and the Bangkok Agreement, lower-level ministers or ambassadors bear this responsibility. I define high-level officials as heads of state and foreign ministers. These policymakers usually have a great deal of power with respect to issues of national security. Thus, RIAs in which high-level officials are designed to meet regularly to make decisions score 1, and 0 otherwise.
Here, again, I go beyond institutional design and examine implementation. This time, implementation refers to the actual meeting of this body. Delays or cancellations of scheduled meetings for technical or political reasons are not uncommon. The Arab Maghreb Union (AMU) Summit, for example, was convened only six times in nine years instead of eighteen times, as planned. Similarly, the annual South Asian Association for Regional Cooperation (SAARC) Summit met only twice from 1990 to 1997. Insofar as high-level officials do not meet in practice, the pacifying effects of this forum are unlikely to materialize. I thus assign 1 point to RIAs in which high-level officials meet in practice and 0 otherwise. As with the variable that pertains to economic scope, I multiply the score on institutional design with the score on implementation. Thus, high-level officials is a dichotomous variable: only RIAs that institute regular meetings of high-level officials who actually meet score a point. Seventy-five percent of the observations (sixty-eight out of ninety) score 1 on this variable and the rest score 0.
Alternative Explanations
This section considers several conventional explanations for intraregional conflict that are not pertinent to international institutions.34
I examined the pacifying effect of other, lower-level institutions as well. These are regional bureaucracy and economic dispute settlement mechanisms. The effect of these institutions was statistically insignificant and not reported in this article.
Economic interdependence, usually conceptualized in terms of international trade, is a widely cited explanation for peace. Indeed, it is argued that RIAs may mitigate conflict partly by reducing barriers to regional trade.35
Mansfield and Pevehouse 2000.
Proponents of the “democratic peace” research program contend that democracies are less likely to fight each other relative to other combinations of regime types. Several studies apply this argument on the regional level.37
To measure regional democraticness, I employ the Polity IV definitions and data.38 With this data set I first calculate the regional average of the composite Polity variable. I then distinguish between regions that inhabit mature democracies and all other regions. Thus, democracy is a dichotomous variable that scores 1 if the average level of democracy is greater than 6, and 0 otherwise.Theories of power transition hold that preponderance of power is associated with more stability and less violence. Several recent studies employ this argument at the regional level and find empirical support for this hypothesis.39
I account for this possibility with the regional concentration ratio.40 This measure takes into account both the relative power of all members, measured in GDP, and the number of members. The value of concentration increases as asymmetry grows and is bounded between 0 and 1. GDP data from the Penn World Tables is used to calculate this variable.41Heston, Summers, and Aten 2002.
The internationalization of civil wars in many parts of the world attests to the potential effect of domestic violence on neighboring states. From a theoretical standpoint, the logic of diversionary war points to the possibility that governments will turn to international conflict in order to ameliorate domestic unrest or take advantage of states that experience high levels of domestic strife. To measure the level of regional domestic conflict, I count the number of such armed conflicts as reported in the Uppsala Conflict Data Program Armed Conflict data set.42
Gleditsch et al. 2002. Internationalized internal disputes are excluded to minimize the possibility of overlap with the dependent variable.
Regional political dynamics may be affected by external forces, such as great power intervention. I control for the effect of the broader international system on the level of regional conflict with Lemke's estimate of the ability of great powers to project their power in different regions.43
Lemke 2002.
It is reasonable to expect that the number of states in a region will be associated with the number of disputes. A greater number of states may result in additional opportunities for interaction and friction, and thus more conflict. members is a count of the states that are members of an RIA. In addition, it is widely accepted that geographical proximity provides more opportunities for interaction, and in turn for conflict.44
See, for example, Gleditsch 2002.
Stinnett et al. 2002.
Finally, realists and other skeptics contend that cooperation through institutions is endogenous to the level of amity in the region. It is thus important to take into account the level of interest similarity among RIA members. I employ an S score to capture the intraregional similarity of interests.46
Signorino and Ritter 1999.
Gartzke and Jo 2002.
See Bueno de Mesquita 1981; and Signorino and Ritter 1999. I use the globally weighted measure of alliance portfolio. Data is obtained from the EUGene software; see Bennett and Stam 2000.
Results
Table 1 reports the estimated effect of the main independent variables on the number of intraregional disputes. Table 2 provides substantive interpretation of the significant variables. Because a negative binomial regression is nonlinear, the substantive interpretation of these results requires one to exponentiate the coefficients. The numbers in Table 2, then, reflect the expected value of the incidence of the dependent variable conditioned by the values of certain independent variables.
Random-effects negative binomial estimates of the effect of economic scope and regular meetings of high-level officials on intraregional MIDs, 1982–2001
Substantive interpretation of the effects of significant independent variables on the expected count of MID, 1982–2001
The results are consistent across different model specifications. They indicate that both a wider scope of economic activity and regular meetings of high-level officials have a significant negative effect on the expected number of MIDs. This is true whether one runs each variable in a separate model (Models 1,2, 4 and 5) or the two variables together (Models 3 and 6). These results provide strong empirical support for my hypotheses. The pacifying effect of the scope of economic activity and regular meetings of high-level officials is not only statistically significant, but also substantively meaningful. For the scope of economic activity, moving from the first to the third quartile reduces the expected count of disputes by about 48 percent. Similarly, RIAs in which meetings of heads of state or foreign ministers take place regularly experience about one-half the number of conflicts, relative to RIAs in which such meetings do not take place.
The performance of the control variables is consistent across the various model specifications. Intraregional commercial interdependence does not have a significant effect on the number of militarized disputes in the region. This result suggests that intraregional trade—in and of itself—may capture the sensitivity rather than the vulnerability dimension of economic interdependence, and that societal groups that benefit from international trade may have only limited impact on their leaders' assessments of the costs and benefits of interstate conflict.49
My analysis indicates that democratic regions are not more peaceful than their nondemocratic counterparts. The estimates of democracy flip signs and never approach a conventional level of statistical significance. This result points to the possibility that—outside the advanced industrialized countries—the effect of democracy on conflict is indeterminate. The vulnerability of a region to great power intervention does not seem to affect the level of violent conflict as well.Regional concentration of power has a negative and significant effect on the level of regional conflict across most models. This finding suggests that regions with a hegemon enjoy low levels of conflict, while regions in which power is more evenly distributed are more violence-prone. These findings are consistent with recent theoretical and empirical studies on this question.50
The estimate of civil war is positive and highly significant across all models. This variable has a sizeable substantive effect as well. This result points to the potential international repercussions of domestic conflicts, which may spill over to produce interstate violence in particular. Finally, as expected, as the number of borders between member-states increases, the number of MIDs increases as well. Thus, more opportunities for interaction produce disagreements and conflict. In contrast, the sheer number of RIA members does not have a bearing on the expected number of militarized disputes.Finally, controlling for intraregional affinity, the significant pacifying effect of the scope of economic activity and regular meetings of high-level officials remain intact. The estimates of similarity of interests—measured either by alliance portfolios or by voting in the UN—are not statistically significant. These results indicate that the similarity of interests is not a good predictor of violent conflict. Moreover, they mitigate concerns of realists that the effect of institutional features reported above is endogenous to prior relationships between states.
Reversing the Causal Arrow
Skeptics may argue that the level of conflict affects the willingness of states to cooperate through international institutions and the willingness of their leaders to meet regularly. It is thus important to ensure that the results are not driven by the reversed causal direction. Above, I have lagged the independent variables to address this issue. Here I take an additional step to tackle this issue by examining the effect of violent conflict on the scope of economic activity and regular meetings of high-level officials.51
For a similar approach to this issue, see Fortna 2004; Mansfield and Pevehouse 2000; and Russett and Oneal 2001.
In order to test the effect of militarized disputes on international institutions, one needs to have a more general model of the determinants of international cooperation. Research in international relations offers several conventional explanations for international cooperation in general and for international institutions in particular. Most of these explanations examine the same factors that explain conflict but turn them on their head. It is commonly believed that higher commercial interdependence, higher levels of democraticness, greater concentration of power, and greater affinity are conducive to cooperation through international institutions. In contrast, a higher level of domestic violence and a greater number of members are expected to undermine the prospects of international institutions.52
For a detailed discussion of, and citations for, the determinant of international cooperation through institutions, see Haftel 2004b; and Haftel and Thompson 2006.
The dependent variables are economic scope and high-level officials. The two variables have different distributions that call for different estimation techniques. economic scope is a continuous variable and is thus estimated with a random-effects ordinary least squares (OLS) regression technique. high-level officials is a dichotomous variable and is estimated with a random-effects probit regression technique. Table 3 reports the results for the dependent variables, each with two alternative specifications for similarity of interests. The empirical results do not support the notion that the prospects of international institutions are endogenous to initial levels of amity among RIA members. Looking first at the determinants of the scope of economic activity, the coefficients of mids members are negative but not statistically significant. Looking next at the determinants of regular meetings of high-level officials, the coefficients of mids members flip signs and are, again, not significant. These results indicate that the levels of the two institutional features that are at the center of this study are not determined by the level of prior intraregional conflict and peace. Thus, my initial findings—that these institutional features have an independent pacifying effect on relationships among RIA members—are further substantiated.
The determinants of economic scope and high-level officials, 1982–97
Conclusion
RIAs exhibit a great deal of variation in their institutional design and the implementation of agreed-on institutional features. The implications of such variation for questions of war and peace are not well understood, however. Extant research treats RIAs—as well as other IOs—as homogenous and lumps together a variety of potential causal mechanisms that may (or may not) link these institutions to conflict. This study advances the understanding of the effect of international institutions on violent conflict by providing one of the first systematic empirical analyses pertaining to the effect of institutional variation on violent conflict.
My findings indicate that the two institutional features that are the focus of this study—scope of economic activity and regular meetings of high-level officials—are indeed conducive to a peaceful resolution of disputes. These two institutional features, it seems, go a long way to reduce uncertainty regarding states' interests, motivations, and resolve. As such, they are effective instruments of interstate bargaining in times of conflict and disagreement. These results indicate that the realist outright dismissal of the potential pacifying effect of international institutions on violent conflict is unfounded. That is, my analysis shows that RIAs are not mere reflections of preexisting power relationships, and that they can promote regional peace and stability.
My study also provides a reminder that RIAs are neither created equal nor equally successful in meeting their stated objectives,53
Mansfield and Milner 1999, 615.
From a practical perspective, when policymakers sit at the negotiation table and design their regional organizations, they ought to expand their economic scope and consider a variety of issue-areas that go beyond trade. From this viewpoint, the recent wave of regionalism—which broadens the agendas of many RIAs to include a variety of economic issues—appears to be a step in the right direction. RIAs should also involve regular meetings among top-level policymakers, which appear to promote a peaceful resolution of political tensions. Finally, implementation of signed agreements is vital. Considering the oftentimes wide gap between institutional design and implementation, if policymakers hope to promote intraregional peace through international institutions, they should better match rhetoric and practice.
Appendix: List of Twenty-Five RIAs
Arab Maghreb Union (AMU), Andean Community (ANCOM), Association of Southeast Asian Nations (ASEAN), Bangkok Agreement (BA), Central American Common Market (CACM), Caribbean Community (CARICOM), Economic Community of the Countries of the Great Lakes (CEPGL), Common Market for Eastern and Southern Africa (COMESA), Economic Community of Central African States (ECCAS), Economic Cooperation Organization (ECO), Economic Community of West African States (ECOWAS), European Free Trade Association (EFTA), European Union (EU), Gulf Cooperation Council (GCC), Indian Ocean Commission (IOC), Latin American Integration Association (LAIA), Common Market of the South (MERCOSUR), Mano River Union (MRU), North American Free Trade Agreement (NAFTA), Organization of Eastern Caribbean States (OECS), South Asian Association for Regional Cooperation (SAARC), South African Customs Union (SACU), Southern African Development Community (SADC), Central African Customs and Economic Union (UDEAC), and West African Economic and Monetary Union (WAEMU).
Categories and indicators of designed scope of economic activity
