Tripwires and free-riders: Do forward-deployed U.S. troops reduce the willingness of host-country citizens to fight for their country?

ABSTRACT

This article investigates the relationship between U.S. overseas troops and the willingness of the citizens of host states to fight for their country. The study joins the long-running debate about burden-sharing and free-riding among U.S. allies. Unlike most previous empirical studies, we focus on non-material or intangible measures of the underlying concepts. Our dependent variable estimates the proportion of citizens expressing a willingness to fight for their country. Scores at the aggregate-national as well as the individual level are shaped by the presence of U.S. military forces, which act as a “tripwire” signaling credible security commitments. This increases opportunities of (non-material) free-riding. We present both bivariate and multivariate analyses covering the period 1981–2014 to test this supposition. Findings indicate that once U.S. troop levels reach a certain threshold (between 100 and 500 troops), citizens’ willingness to fight drops significantly. This likely reflects non-material free-riding.

KEYWORDS:

• Willingness to fight
• U.S. troops
• free-riding
• burden-sharing
• tripwire
• U.S. military bases

Acknowledgments

The empirical section of this article builds in part on an empirical study—“Democratic peace and the norms of the public”—published in Review of International Studies in 2016. The authors are deeply appreciative of the insightful comments made by two anonymous reviewers on the original version of the manuscript. The authors take full responsibility for any remaining errors.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes on contributors

Jo Jakobsen is a professor in political science at Norwegian University of Science and Technology (NTNU). He obtained his PhD at NTNU in 2007. Since then, Jakobsen has conducted research and teaching on a broad range of issues related to International Relations and international political economy. His work has been published in a number of international journals, including Review of International Studies, International Interactions, Conflict Management and Peace Science, Kyklos, Middle Eastern Studies, Business Horizons, and Chinese Journal of International Politics. Jakobsen’s current research focuses in particular on international security and geopolitical risks for international business.

Tor G. Jakobsen is a professor in political science at Norwegian University of Science and Technology, NTNU Business School. He received his PhD at NTNU in 2011. His fields of interest include political behavior, international politics, studies of conflict and peace, and quantitative methods. Jakobsen has authored and co-authored articles in a broad range of international journals, including European Sociological Review, European Societies, Work, Employment and Society, Conflict Management and Peace Science, and Sport Management Review. He has also co-authored the book Applied Statistics with Stata (Sage). Jakobsen has been a visiting scholar at University of California, Berkeley, and Cornell University.

Notes

1. When denoting security goods provided to alliance members, Olson and Zeckhauser (1966) alternate between using “public,” “collective” and “organizational.” Such goods, of course, are really “public” only within a given group or alliance (see also Olson, 1965, p. 38, footnote 58).

2. In January 2017, the United States began the deployment of an army brigade to Poland, to bolster NATO deterrence against Russia; see https://pl.usembassy.gov/abct/.

3. Data are available from: http://www.sipri.org/research/armaments/milex.

4. A handful of empirical studies have investigated the determinants of war willingness (Anderson, Getmansky, & Hirsch-Hoefler, in press; Diener & Tov, 2007; Inglehart et al., 2015; Jakobsen et al., 2016; Paez et al., 2008; Puranen, 2014; Torgler, 2003). One of these, while focusing mainly on the impact of regime type, also includes a dummy for the presence of US troops (with a 1,000-troops threshold), which, in their multilevel analysis, is shown to suppress war willingness somewhat (Jakobsen et al., 2016). In the next section, we follow up on this finding, investigating more extensively the relationship between military-security ties with the US and non-material burden-sharing and free-riding.

5. Data are available from the authors on request.

6. For further information on the WVS and its methods and sampling procedures, see http://www.worldvaluessurvey.org/wvs.jsp. The datasets are made available through the Norwegian Social Science Data Services (NSD). Neither Ronald Inglehart, WVS, or NSD are responsible for the analysis of these data, or for the analyses and conclusions made in this article.

7. For the years 1981–2003, data are organized and provided by Kane (2004) and can be found at: https://www.heritage.org/defense/report/global-us-troop-deployment-1950-2003. Data for 2004–2007 are retrieved from Vetfriends: https://www.vetfriends.com/US-deployments-overseas/. For the years 2008–2014, we extracted data from the US Defense Manpower Data Center, which regularly publishes updated data on US overseas troops: https://www.dmdc.osd.mil/appj/dwp/dwp_reports.jsp.

8. USdefpact includes NATO members, members of the Organization of American States (OAS), Japan, South Korea, the Philippines, and Australia. Broadly following the categorization used by the Correlates of War (COW) Project’s Alliances Data Set (Gibler, 2009)—which can be accessed at: http://www.correlatesofwar.org/news/alliances-data-set-v4-1-available-1—an alliance is here conceived of in rather narrow terms. What constitutes a(n) (formal) alliance is not always straightforward, though, and “the precise arrangements embodied in different alliances vary enormously” (Walt, 2009, p. 86). In the case of the US and Israel, for example, a military alliance has never been signed “because there has never been a question that the United States would provide military assistance to Israel in a crisis” (Morrow, 1991, pp. 906–7). In the sensitivity analysis we therefore add Israel, along with Taiwan and Pakistan, so as to commensurate with Michael Beckley’s (2015, p. 23) extended list of US defense pacts.

9. Data are available from: http://www.systemicpeace.org/polity/polity4.htm.

10. We used, as a base, data from the Military Recruitment Dataset (Nathan Toronto, Military Recruitment Data Set, codebook, version 2005.1), which provides information up until 2004/2005 (depending on the country) and which is available at: http://fmso.leavenworth.army.mil/documents/mildat/RecruitmentCodebook.pdf. We used Chartsbin for 2010 and 2011 (see: http://chartsbin.com/view/1887), and normally also for the 5–6 previous years. We generally drew on the CIA World Factbook for the years 2012–2014, see: https://www.cia.gov/library/publications/the-world-factbook/fields/2024.html. Any missing country-years were set to the same values as those of these three sources if they corresponded with each other. If they did not, we used alternative sources of information about the exact year of change in military recruitment policy.

11. Available at: http://data.worldbank.org/.

12. Note that the relatively low N means that we cannot employ a fixed-effect model, nor one that measures changes over time between values on the key variables.

13. MID data are available at: http://www.correlatesofwar.org/data-sets/MIDs. We gave country-years the value 1 if they had been involved in a MID in the year in question or one of the two previous years. Data on interstate wars are from Uppsala University and from the Peace Research Institute Oslo: http://ucdp.uu.se/downloads/. The variable we use includes participation as an external actor intervening in a civil war. Data on troops per capita and economic growth are from the World Bank Development Indicators. Data on military expenditures are from Stockholm International Peace Research Institute: http://www.sipri.org/research/armaments/milex.