The Forces of Attraction: How Security Interests Shape Membership in Economic Institutions
Christina Davis and Tyler Pratt
Multilateral economic institutions govern an increasing range of international economic activity, often using rules of non-discrimination to insulate patterns of trade, aid, and investment from political influence. In “The Forces of Attraction,” we explore how states use geopolitical discrimination over membership to politicize economic cooperation under the guise of multilateralism.
To examine the politics of institutional membership, we construct a dataset of “potential memberships” in multilateral economic organizations from 1816-2014. This is a bigger sample of intergovernmental organizations (IGOs ) than previous studies that examine only the largest IGOs or regional IGOs, and is more closely matched to our theoretical test than studies that aggregate all IGOs. We start with the Correlates of War (COW) International Organizations Dataset (Pevehouse et al. 2004). We select organizations with an economic focus using the Yearbook of International Organizations’ description of the aims and subject of each organization. We combine this information with original data on the formal charters that established the organizations. This yields a sample of 231 intergovernmental organizations that govern economic exchange.
We structure our data at the level of the state-IGO-year. Our focus on the strategic selection of members motivates narrowing the sample. Using the terms in the IGO Charter documents, we exclude “universal” IGOs that do not require the approval of other members as part of the accession process. We also exclude state-IGO pairings for regional organizations where the state resides outside the regional scope of the IGO. This generates 713,333 state-IGO-year observations. In order to compare different phases of evolution in IGO membership, we include data on both IGO formation and enlargement.
The flexible state-IGO-year structure allows us to examine how relational variables specific to state-IGO pairs (e.g., alliance ties between a potential entrant and existing members of an IGO) predict membership. Rather than focusing on state decisions in isolation or as dyadic links, our data structure examines connections between a state and a group of other states within the IGO as a forum. We include both state- and IGO-level variables, as well as interactions between them. The variables we examine include:
- Lead State Allianceijt: an indicator equal to one when state i shares an alliances with the largest economic power among member states of IGO j in year t.
- Average Alliancesijt: the proportion of IGO j’s member states with which state i shares a formal alliance in year t.
- Trade with Membersijt: average (logged) volume of merchandise imports and exports between state i and member states of IGO j in year t.
- Stringent Accessionj: an indicator for IGOs that require a supermajority vote, unanimous consent, or special committee approval to admit new members.
- Regional IGOj: an indicator for IGOs with a geographic scope focused on one region
Our codebook includes a complete list of variables in the dataset.
While our article focused on the geopolitical origins of membership patterns, the data can be used to examine additional determinants of IGO membership. For example, our new data on IGO charters enables scholars to compare trends in IGO creation over time (Figure 1 below), examine alliance patterns between IGO members and non-members (Figure 2) or theorize the rationale for variation in the accession rules. The dataset will support a range of empirical approaches to address the challenge of endogenous institutions. Improvements in modeling selection into IGOs are necessary to better understand the effectiveness of the organizations on policy outcomes.
Figure 1: Number of Economic IGOs with Open vs. Stringent Accession Procedures
Figure 2: Alliance Ties with IGO Members, Member vs. Non-Member States