The Gravity of Arms

Abstract

The aim of this paper is to investigate the determinants of international arms transfers in a gravity model framework. By distinguishing between the decision to export arms (extensive margin) and the value of the arms exported (intensive margin), while also considering its interdependence, is what differentiates this paper from previous research. A theoretically justified gravity model of trade augmented with political and security motives is estimated using a two-stage panel data approach for 104 exporting countries over the period from 1950 to 2007. In addition to the usual gravity variables related to the economic mass of the trading countries and the trade cost factors, the model is extended with political and security factors. The level of democracy in both trading partners, political differences between trading partners and voting similarity with the United States in the UN General Assembly of the countries engaged in trade are the main political factors, whereas the existence of conflicts, military pacts, and embargoes are taken as security motives. The key result indicates that both political and security motives are an important determinant of an arms trade, but their effects on the extensive margin of exports (the decision to order a transfer) differs from their effect on the intensive margin (average value of exports). Moreover, the relative importance of the factors under study has changed since 1989. In the post-cold war period, countries that are less democratic are more likely to export arms, military pacts are less relevant and embargoes play a role.

Keywords:

• Arms trade
• political factors
• margins of trade
• gravity model
• two-stage approach

JEL Codes:

• F14
• F51

Acknowledgments

The useful comments and constructive suggestions by two anonymous referees are gratefully acknowledged. The usual disclaimer applies.

Notes

1. This is commonly attributed to countries reacting against ‘military Malthusianism,’ according to which unit costs of major weapon systems rise faster than government budget revenues. As a reaction, countries might try to reduce the unit costs by increasing exports and profiting from economies of scale to counteract the increasing costs of domestic production (Brauer and Dunne 2011).

2. The tree dimensions are right, center, and left oriented governments from the World Bank Development Research Group’s Database of Political Institutions.

3. According to Brauer and Dunne (2011), the knowledge transfer of offset agreements is, if it exists at all, relatively small.

4. Investigating United States arms exports, Sislin (1994) finds successful attempts to influence the partner countries under certain conditions, especially in the first decades of the cold war.

5. The list of exporters and importers included in the estimations is not provided by Akerman and Seim (2014). The indicated number of observations included in Table 1 (page 541) is more than 600 thousand in the models with country fixed effects, shown in columns (3)–(6), but only around 300 thousand in the models without country fixed effects, columns (1)–(3), the authors do not explain this discrepancy.

6. Blanton (2005) focused on United States exports exclusively.

7. List of participants: Argentina, Australia, Austria, Belgium, Bulgaria, Canada, Croatia, the Czech Republic, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Ireland, Italy, Japan, Latvia, Lithuania, Luxembourg, Malta, Mexico, the Netherlands, New Zealand, Norway, Poland, Portugal, the Republic of Korea, Romania, Russian Federation, Slovakia, Slovenia, South Africa, Spain, Sweden, Switzerland, Turkey, Ukraine, the United Kingdom, and the United States.

8. Military pacts can be classified as defense, nonaggression, entente, and neutrality pacts.

9. For example, the HS-1992 goods category has an entry for ‘arms and ammunition and parts and accessories thereof’ (HS-93).

10. Due to confidentiality reasons, countries may not report all of their detailed trade. In data sources, such as UN Comtrade, this trade will usually be included in a category called ‘others’ and in the total trade value.

11. This fact could be linked to the September 2011 terrorist attack and the subsequent increase in the military budgets in Western countries.

12. The development and expansion of the domestic arms industry could also be related to the trade theories that identify imperfect competition and increasing economies of scale as an explanation for international trade (Helpman and Krugman 1985). In this sense, the size of the arms industry could also be positively related to lower unit costs of production and to higher exports.

13. See Table A.5 for a detailed description of all variables in the model.

14. As a robustness check, we also use the square difference of the democracy scores (as in Akerman and Seim 2014).

15. The SIPRI Arms Transfers Database provides information on the order of a transfer and the value of the transfer in separate data-sets with different timings.

16. A comparison between the arms trade and in goods other than arms is not straightforward. As described in section 3.1, arms or components of arms (e.g. ship engines) are often labeled as non-military goods or not reported for reasons of confidentiality.

17. The large time dimension in our data-set mitigates the potential bias that arises due to the incidental parameter problem in Probit panel data models with country dummy variables. The bias is of order (1/T), therefore it is lower than 2% in our sample.

18. When estimating the model with only gravity variables, only political variables and only security factors, the pseudo-R² equals, respectively, 0.30, 0.16, and 0.06.

19. See the robustness section for a closer replication of the results in Akerman and Seim (2014).

20. When estimating the model with only gravity variables, only political variables and only security factors, the pseudo-R² equals, respectively, 0.18, 0.09, and 0.09.

21. Comola’s (2012) specification is substantially different to ours; she models separated exporter and importer fixed effects and the estimation method is a Tobit, instead of method Mundlak approach. Also, the sample of countries and years is different; she considers only the 20 major MCW producers over the period 1975–2004, whereas we consider all countries that export arms over the period 1950–2007.

22. Using common religion as additional variable in the first-stage regression renders the coefficient of GDP in the exporter country in column (6) non-significant, while the rest of results remain very similar.

23. This is in accordance with Blanton (2005), in that it could be due to the United States’ support for countries in close political alignment.

24. We estimate the model with OLS (linear probability model) without (column 1) and with exporter and importer fixed effects (column 2). Columns (3) to (5) also include year dummies (time fixed effects). Akerman and Seim (2014) do not mention the inclusión of time fixed effects, which in any case should be added to the model to control for unobserved heterogeneity that is common to all country-pairs and varies over time.

25. Results are available upon request.