Proposed Network Analysis of the International Arms Trade

Kirk Bansak

SNA Project Proposal

To be undertaken for credit in D217

Research Question:

Does the international arms trade reflect or affect interstate power relations?

Background:

In recent years, a number of political scientists have begun to adopt formal network analysis methods.  As Emilie Hafner-Burton, Miles Kahler, and Alexander Montgomery have argued, “Two issues regarding network power are particularly important to international relations: whether actors can increase their power by enhancing and exploiting their network positions, and the fungibility of power—whether network power can be used to supplement or offset other forms of power” (Hafner-Burton, Kahler, and Montgomery, 2009: 573).  Accordingly, these three political scientists have proposed and begun to develop an agenda for applying network analysis to international relations, focusing on states as the objects, or nodes, of analysis.

However, Hafner-Burton, Kahler, and Montgomery have observed that most existing international relations data, such as mutual membership in international institutions, correspond to loose affiliation ties rather than the “directly measured ties” that network analysts consider to be “the most valuable and least problematic data” (579).  A key exception, they note, is trade data, which is directed, measurable, and documented.  Accordingly, this project will seek to perform network analysis on states using trade data as the underlying network structure.  Furthermore, this project will focus on one specific category of trade: trade in major conventional military armaments.  Given the timeless emphasis on security and power in interstate affairs, and states’ dependence (at least partially) upon arms for security and power, the international arms trade may present fertile (and underexplored) ground for examining international relations through network analysis.

Objective:

The objective of this project will be to use network analysis to gain insights into the global arms trade and how trends in the global arms trade may reflect or even impact power relationships among states.  By considering international arms transfers as constituting a global network, in contrast to analysis that focuses on bilateral trade relationships, broader and perhaps more unique observations can be made about international relations phenomena pertaining to the influence, security, and foreign policy strategy of individual countries, regions, or coalitions.  Furthermore, any lessons learned (successes and failures) through the attempt to apply network analysis to this problem could serves as contributions to the budding IR Network Analysis agenda.  Existing data on international arms trade will be used; a survey will not be necessary.

Hypothesis:

It is hypothesized that network analysis of the global arms trade will reveal a small number of reasonably distinguishable arms trade “cliques” that are not completely isolated from one another but rather linked together by a small number of mutual arms trade partners.  (It is also hypothesized that the network analysis will reveal a range of completely unexpected results.)

Methodology:

As indicated, states will be the objects (nodes) of analysis for this project, and trade in arms will be the network connections (directed and measurable ties).  A general assumption will be made that trade in arms (of the major conventional type I will examine) is managed, regulated, and determined by the state, as is standard practice for countries with stable governance structures.  Thus, the general assumption of the state as a rational actor can be made in this context of arms trade decisions and behavior, which will aid in the IR network analysis.

Prior to performing network analysis, a number of prerequisite steps must be taken.  The first is to select the number of states and the specific states that will constitute the network, and the second is to select the trade data that will constitute the network ties.  State selection can be performed in various ways, including by highest volume of arms imports or arms exports overall, by highest military expenditure overall, to include broad regional representation, etc.  Meanwhile, trade data selection requires two decisions: selection of a data source and selection of the specific categories of armaments to be considered.  Two existing data sources seem particularly appropriate for this project: the Stockholm International Peace Research Institute (SIPRI) Arms Transfers Database and the UN Register of Conventional Arms.  Further consultation with the class and professor will be pursued in order to make final decisions on state selection and trade data selection.

Furthermore, the timeframe of the arms trade flows must be determined.  Due to the vicissitudes of arms trade deals, capturing data over an extended period of time (e.g. 10 years) makes more sense and would be more representative of arms trade relationships than would capturing data over a single year.

Next, attribute data will be added to the states.  Attributes for consideration include: region, security/alliance organization affiliations, trade/economic organization affiliations, GDP, GDP per capita, regime type, estimated defense spending overall, existence (or not) of a regional adversary, and length of time since last conventional war.

A number of centrality measures will be applied to make inferences about individual states’ roles or status in the international system.  Out degree may be a measure of political power or economic clout.  In degree may be a measure of dependence.  Eigenvector scores may denote the future (un)reliability of trade partners, based on the extent to which those trade partners have alternative suppliers or alternative buyers.  Betweenness scores may indicate the extent to which a country serves an intermediary trade role or holds macroscopic trade flows stable, given the complex intra-industry trade (i.e. various categories, components, and sub-components) that occurs in the arms trade.  Closeness may denote the aggregate quality of a state’s relations with the international community—for instance, relatively high closeness might be indicative of not existing exclusively within any single security bloc or coalition.

In addition to centrality scores, clique analysis will be performed in order to identify and make observations about alliances and trade blocs.  In particular, it would be interesting to assess to what extent arms trade blocs reflect formal alliances.  Furthermore, ego network analysis will be focused on the global geopolitical heavyweights, namely the United States and China.  In addition, analysis will be performed to assess to what extent network trends reflect physical geographical realities, such as proximity between trading partners.

Potential Problems:

As described above, the arms trade is characterized by complex intra-industry trade trends: armaments are not a homogenous good but rather a highly heterogeneous good with varying degrees of interchangeability and synergy between different categories of armament for different countries.  For instance, the SIPRI database identifies the following categories of major conventional armaments: aircraft, air defense systems, anti-submarine warfare weapons, armored vehicles, artillery, engines, missiles, satellites, sensors, ships, and other.  However, it is unclear that this project will be able to completely capture these nuances because the data available does not necessarily break down each bilateral trade flow (each directed tie) by category.

In addition, while SIPRI and the UN have produced rich databases of the global arms trade, the data compiled does not encapsulate one hundred percent of arms traded internationally due to various problems of access to information.

Preliminary References:

Brauer, Jurgen. “Arms Industries, Arms Trade, and Developing Countries.” In: Sandler, Todd, and Keith Hartley (Eds.), Handbook of Defense Economics: Defense in a Globalized World, Volume 2. Amsterdam: Elsevier, 2007.

Garcia-Alonso, Maria D.C., and Paul Levine. “Arms Trade and Arms Races: A Strategic Analysis.” In: Sandler, Todd, and Keith Hartley (Eds.), Handbook of Defense Economics: Defense in a Globalized World, Volume 2. Amsterdam: Elsevier, 2007.

Hafner-Burton, Emilie M., Miles Kahler, and Alexander H. Montgomery. “Network Analysis for International Relations.” International Organization 63 (2009), 559-592.

Hafner-Burton, Emilie M., and Alexander H. Montgomery. “Centrality in Politics: How Networks Confer Power.” Southern Illinois University Carbondale, OpenSIUC, Conference Proceedings (2010). Paper 9. http://opensiuc.lib.siu.edu/pnconfs_2010/9.