Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Guns versus Climate: How Militarization Amplifies the Effect of Economic Growth on Carbon Emissions

Guns versus Climate: How Militarization Amplifies the Effect of Economic Growth on Carbon Emissions Building on cornerstone traditions in historical sociology, as well as work in environmental sociology and political-economic sociology, we theorize and investigate with moderation analysis how and why national militaries shape the effect of economic growth on carbon pollution. Militaries exert a substantial influence on the production and consumption patterns of economies, and the environmental demands required to support their evolving infrastructure. As far-reaching and distinct characteristics of contemporary militarization, we suggest that both the size and capital intensiveness of the world’s militaries enlarge the effect of economic growth on nations’ carbon emissions. In particular, we posit that each increases the extent to which the other amplifies the effect of economic growth on carbon pollution. To test our arguments, we estimate longitudinal models of emissions for 106 nations from 1990 to 2016. Across various model specifications, robustness checks, a range of sensitivity analyses, and counterfactual analysis, the findings consistently support our propositions. Beyond advancing the environment and economic growth literature in sociology, this study makes significant contributions to sociological research on climate change and the climate crisis, and it underscores the importance of considering the military in scholarship across the discipline. Keywords climate change, environmental sociology, political-economic sociology, development sociology, militarization Rich bodies of sociological theory and analy- University of British Columbia ses have significantly advanced scientific Vilnius University understanding of the human dimensions of University of Utah Boston College climate change (e.g., Davidson 2022; Dietz, Wayne State University Shwom, and Whitley 2020; Dunlap and Utah State University Brulle 2015; Klinenberg, Araos, and Koslow Drexel University 2020; Norgaard 2018). Distinct research tra- University of Wyoming ditions focus on how structural characteristics University of California-Riverside of societies, usually nation-states, generate Corresponding Author: different levels and rates of carbon dioxide Andrew K. Jorgenson, Department of Sociology, and other greenhouse gas emissions, and also University of British Columbia, AnSo 2111, how relationships between nations shape their 6303 NW Marine Drive, Vancouver, BC V6T 1Z1, unequal contributions to emissions and Canada. Email: andrew.jorgenson@ubc.ca planetary warming (e.g., Givens, Huang, and 2 American Sociological Review 00(0) Jorgenson 2019; Kelly 2020; Pellow and sociology, studying the role of militaries is a Brehm 2013; Rice 2007; Rudel, Roberts, and vital direction to pursue, given the emergence Carmin 2011). While broadening and deepen- of the war economy and defense industry ing the presence of environmental sociology (see also Giddens 1987; Hooks 1990). As in the discipline as a whole (Lockie 2022; Andreski (1968:1) argued decades ago: Mezey 2020; Scott and Johnson 2017; Smith 2017), this scholarship also contributes to The problem of the influence of military interdisciplinary climate science efforts and organization on society has, on the whole, policy considerations (Haberl et al. 2020; failed to attract the attention of social sci- IPCC 2022; Jorgenson et al. 2019; Longo entists. To be sure, much has been written et al. 2021; Rosa and Dietz 2012; Thomas about war, its alleged evil or beneficial et al. 2019). effects, its causes and the possibilities of its Drawing on various disciplinary subfields, abolition. But the only writers who appre- the most central question within this area of ciated the importance of military factors sociological inquiry concerns the relationship in shaping societies were Max Weber and between nations’ carbon emissions and their Gaetano Mosca. This persistent neglect is economic growth (Bohr and Dunlap 2018; due, I think, to the insidious utopianism Fisher and Jorgenson 2019; Stuart 2021). A which pervades sociological thinking. suite of critical perspectives, including tread- mill of production and metabolic rift theories, Building on foundational traditions in his- argue that economic growth is antithetical to torical sociology (e.g., Chase-Dunn 1998; environmental protection, given the increased Mann 2012; Mills 1956; Tilly 1990) as well energy and resource demands, as well as the as work in environmental sociology (e.g., subsequent environmental harms, associated Hooks and Smith 2004; Smith and Lengefeld with such growth (e.g., Clark and York 2005; 2020) and political-economic sociology (e.g., Foster 1999; Gould, Pellow, and Schnaiberg Boswell 1989; Scanlan and Jenkins 2001), 2008). In contrast, more optimistic perspec- we theorize and investigate with moderation tives, such as ecological modernization theory analysis how and why militarization ampli- and environmental state approaches, suggest fies the effect of economic growth on nations’ energy efficiency improvement and environ- carbon emissions. As complex social institu- mental protection measures often accompany tions, the world’s militaries exert a sizable growth (e.g., Fisher and Freudenburg 2004; influence on the production and consumption Hironaka 2014; Mol 2003). Prior research patterns of nations and their economies, and tends to support the more critical perspec- the environmental demands required to sup- tives. Analyses consistently show positive port their evolving infrastructure. associations between carbon emissions and The rise of the world’s militaries, espe- economic growth, with the magnitude of the cially after the Second World War, led to relationship varying for nations in differ- modern forms of militarization shaping ent structural and temporal contexts (e.g., carbon-intensive growth in national and Dietz 2017; Dietz and Rosa 1997; Jorgenson international economies through contracts for 2014; Jorgenson and Clark 2012; Rosa, York, research, development, production, and sup- and Dietz 2004; Thombs 2018a; Thombs and port. The continual preparation for potential Huang 2019; York 2012). conflicts and the desire to maintain national To advance sociological research on cli- security increases the scale of resource- mate change, it is necessary to gain greater consuming economic activities within the understanding of how other prominent soci- defense industry. With elaborate communi- etal characteristics shape the relationship cation technologies, larger ships, and faster between emissions and economic growth. planes and helicopters, militaries move peo- Although largely overlooked in generalist ple and equipment throughout the world more Jorgenson et al. 3 quickly. Extensive production systems and differences. Across various model specifica- supply chains within the defense industry tions for different samples, robustness checks, and other areas of the private sector operate a range of sensitivity analyses, counterfactual to meet the various needs of militaries’ infra- analysis, and for the two carbon measures, the structures, including their bases and instal- findings confirm our propositions. lations scattered around the globe, and the needs of their soldiers and support personnel. LiTERATuRE REviEw National militaries are increasingly capi- Economic Growth and Environmental tal intensive, focusing on technologies in Change weaponry, transportation, and communica- tions. In line with prior research, we use military expenditures per soldier to measure Sociological research on the human dimen- these capital-intensive features of militariza- sions of global environmental change, includ- tion (e.g., Jorgenson and Clark 2009; Kentor ing energy consumption and greenhouse gas and Jorgenson 2017; Kentor, Jorgenson, and emissions, largely focuses on the effects of Kick 2012; Looney 1990). Likewise, militar- economic growth (Caniglia et al. 2021; Dietz ies with relatively larger forces require expan- 2015; Dietz et al. 2020). On the one hand, sive built infrastructures and huge amounts optimistic perspectives argue that as societies of material goods, such as food and clothing. experience economic growth, the magnitude Consistent with other sociological inquir- of environmental harms is likely to decrease. ies, we use military participation rate, which The reductions in environmental harms are quantifies a nation’s military personnel as due to the emergence of an environmentally percent of total labor force, to capture the focused state (Dietz et al. 2015; Falkner 2021; relative size of militaries (e.g., Carlton-Ford Fisher and Freudenburg 2004; Frank 1997; 2010; Carlton-Ford et al. 2019; Kick et al. Frank, Hironaka, and Schofer 2000; Spaar- 1998; Kleykamp 2007). We suggest that both garen, Mol, and Buttel 2006) and a growing expenditures per soldier and participation culture of post-materialism, coupled with a rate measure far-reaching characteristics of strengthening commitment to sustainability militarization that enlarge the effect of eco- within civil society (Givens and Jorgenson nomic growth on nations’ carbon emissions. 2013; Inglehart and Baker 2000; Kennedy In particular, we argue that each increases the and Givens 2019; Longhofer and Schofer extent to which the other amplifies the effect 2010; Marquart-Pyatt 2012; Running 2013; of economic growth on carbon pollution. Vasi et al. 2015). Other theorized mechanisms To test our arguments, we estimate longi- include emerging technologies driving the tudinal models of emissions for 106 nations ecological modernization of production and from 1990 to 2016, with a particular focus on distribution systems (Bugden 2022; Huber the three-way interaction between economic 2010; Mol 2003; Rieger 2021), as well as growth, measured as GDP per capita, military the overall greening of organizational culture expenditures per soldier, and military partici- and practices in the private sector through the pation rate. The three-way interaction allows adoption of an ecological rationality and the us to quantify the effect of economic growth diffusion of corporate social responsibility on emissions at different levels of military (Sharkey and Bromley 2015; Vandenbergh expenditures per soldier and military partici- and Gilligan 2017; cf. Grant, Bergesen, and pation rate simultaneously. We treat carbon Jones 2002; Lim and Tsutsui 2012). dioxide emissions per capita as our primary On the other hand, critical perspectives dependent variable, as it reflects international argue that economic growth leads to increased inequities in contributions to global emissions environmental effects, including higher lev- and climate change. We also estimate mod- els of carbon emissions. Economic growth els of total emissions, which capture scale is predicated on the continual expansion of 4 American Sociological Review 00(0) markets, supported by treadmills of produc- also Adua, York, and Schuelke-Leech 2016; tion with extensive horizontal and vertical Burns, Davis, and Kick 1997; Greiner 2022; linkages as well as transportation networks Huang 2018; Huang and Jorgenson 2018; and logistics systems moving enormous vol- Hyde and Vachon 2019; Kelly, Thombs, and umes of raw materials and finished commodi- Jorgenson 2021; Mejia 2021; Rosa et al. ties throughout the world (Braswell 2022; 2004; Soener 2019; York 2012). This corpus Bunker and Ciccantell 2005; Clark, Auerbach, of research supports the general arguments of and Longo 2018; Deb 2021; Gould et al. 2008; the more critical sociological approaches, and Pellow 2007; York, Rosa, and Dietz 2003). it has gained increased recognition among the If left unchecked, these energy-intensive and climate change mitigation community (e.g., waste-generating processes disrupt socio- Haberl et al. 2020; IPCC 2022; Keyßer and ecological systems, often exceeding natural Lenzen 2021). limits while contributing to a global “carbon Other studies focus on how certain factors rift” (Clark and York 2005; see also Davidson moderate the positive relationship between and Andrews 2013; Foster 1999; Foster and carbon emissions and economic growth. For Clark 2020; Foster and Holleman 2012). example, income inequality intensifies their Although nation-states adopt environmen- association in affluent nations (McGee and tal regulations, they simultaneously prioritize Greiner 2018), whereas political inequality economic growth through the protection of does the opposite: the positive relationship private property, bailing out different sectors between emissions and growth is stronger and industries when deemed necessary, main- for nations with higher levels of political taining energy security, and promoting trade equality (Thombs 2021). The overall role of agreements, all of which place increased pres- renewable energy technology in shaping the sure on the environment (Almeida and Chase- emissions and economic growth association Dunn 2018; Buttel 2000; Elliott and Frickel is inconclusive (e.g., Davidson 2019; Thombs 2015; Gareau and Lucier 2018; Rudel 2009). 2017; York and McGee 2017), and nations The environmental benefits of technology are more embedded in global environmental civil often reduced if not entirely outpaced, given the society experience a modest decrease in the contradictory position of the state, the increas- positive relationship between emissions and ing energy and material demands of societies growth through time (Longhofer and Jor- (partly due to efficiency-driven cost reductions genson 2017; see also Fisher 2022; Grant that encourage greater production and con- and Vasi 2017; Schofer and Hironaka 2005; sumption), and the overall growth and diversifi- Shandra et al. 2004; Shorette 2012). cation of markets (Adua, Clark, and York 2021; Driscoll 2021; Grant, Jorgenson, and Longhofer Militarization and Environmental 2020; Gunderson, Stuart, and Petersen 2018; Change Malin et al. 2019; Sanderson and Hughes 2019; Shwom 2011; York and McGee 2016). To advance sociological understanding of the A substantial body of sociological research causes of climate change, we theorize and consistently finds positive associations test how a powerful yet overlooked dimen- between nations’ carbon emissions and eco- sion of human social organization influ- nomic growth. Longitudinal studies indicate ences the effect of economic growth on that the positive relationship increases in mag- the environment: militarization. War itself nitude through time for less affluent nations, destroys the environment through scorched while remaining large and relatively stable earth practices, the use of biological and for more affluent nations (Jorgenson 2014; chemical weapons, and the killing of flora Jorgenson and Clark 2012; Knight and Schor and fauna (Brauer 2009; Frey 2013; Mitchell 2014; Thombs 2018a; Thombs and Huang 2020; Sills 2014; Wilcox 2011; Zierler 2011). 2019; Vesia, Mahutga, and Buì 2021; see From the mid-1940s to the early 1960s, the Jorgenson et al. 5 atmospheric testing of atomic and nuclear and petroleum-based items as well as other weapons produced radioactive fallout that material resources ranging from steel to cot- spread great distances by wind, water, and ton (Belcher, Neimark, and Bigger 2020; living organisms, leading to increased cancer Lawrence et al. 2015; USDOD 2020). These rates among downwinders (Commoner 1971; capital-intensive and scale characteristics Rice 2015). of militarization all contribute, directly or The environmental consequences of war indirectly, to greenhouse gas emissions and in the modern era continue to evolve, as various forms of environmental degradation emerging technologies in weapons, transpor- (Belcher, Bigger, et al. 2020; Clark, Jorgen- tation, and communications systems shape the son, and Kentor 2010; Gould 2007; Jorgenson, scale and precision of destruction (Lengefeld, Clark, and Kentor 2010; Roberts, Grimes, and Hooks, and Smith 2021; Levy and Sidel 2007; Manale 2003; Smith and Lengefeld 2020). Machlis and Hanson 2008). For the more Driven by risk and cost reduction as well capital-intensive and technologically advanced as energy security concerns, and often publicly militaries, this manifests in forms of “risk- framed as climate mitigation efforts, the mili- transfer militarism” (Shaw 2002, 2005), which taries of many nations increasingly focus on shields their homeland’s citizens, minimizes enhanced fossil fuel efficiency and the growing casualties for their soldiers, and decreases use of renewable forms of energy (Bigger and loss of machinery, while inflicting damage Neimark 2017; Condliffe 2017; Light 2014; on human populations, the built environment, Samaras, Nuttall, and Bazilian 2019; USDOD nonhuman species, and the overall natural 2020). However, the pursuit of carbon effi- environment of distant locations (Hooks, ciencies and renewable energy is challenging. Lengefeld, and Smith 2021; Lengefeld and Militaries traditionally prioritize bigger and Smith 2013; Smith and Lengefeld 2020). faster weapons and transportation systems to As noted by sociologists advancing the gain strategic and competitive advantages over treadmill of destruction perspective, the envi- geopolitical rivalries. Modern fighter planes, ronmental effects of militarization are not such as the F-15 and F-16, burn 1,500 to 1,700 limited to war and weapons testing (Hooks gallons of fuel per hour, military helicopters and Smith 2004, 2005; see also Alvarez 2016; consume approximately five gallons for each Bradford and Stoner 2017; Clark and Jor- mile traveled, and non-nuclear aircraft carriers genson 2012; Lawrence et al. 2015). In the utilize close to 6,000 gallons of fuel per hour name of national security, and motivated by while in operation (Jorgenson and Clark 2016; geopolitics and risk-transfer militarism, mili- Levy and Sidel 2007; Sanders 2009). Similar to taries continually invest in and pursue new what occurs in private- and other public-sector technologies in weapons, transportation, and contexts (see Grant et al. 2020; Mazur 2013; communications systems (Alic et al. 2010; Mitchell and York 2020; Simpson, Dunlap, and Burmaoglu and Sarıtas 2017; Mann 2014). Fullerton 2019; Thombs 2018b), the energy The United States alone has over 900 domes- required for militaries’ information technology tic bases and over 800 international bases systems could also involve contradictions and in 130 countries, as well as smaller military conflicts between increased carbon efficiency, installations known as lily pads throughout the transition to sustainable energy sources, and the world (Johnson 2004; Sanders 2009; Turse overall growth as they become more capital- 2015; Vine 2015). The scale of militaries’ intensive and technologically focused (Alic evolving infrastructure, including their trans- et al. 2010; Samaras et al. 2019; Sohag et al. portation systems to move people, supplies, 2021). and weaponry by land, air, and water through- Some nations’ militaries have made efforts out the globe, and their constant research and to become more energy efficient and ecologi- development activities, involve the consump- cally sustainable. However, military opera- tion of substantial amounts of fossil fuels tions, training exercises, and related land 6 American Sociological Review 00(0) holdings are often exempt from environmen- Beckfield, and Seeleib-Kaiser 2005; Brady, tal regulations domestically and abroad (e.g., Beckfield, and Zhao 2007; Mahutga 2006; Kramer 2020; Light 2014; Lynch et al. 2017; Thombs 2018a). National militaries propel Smith 2020; Wilcox 2007). A national secu- these socio-environmental processes through rity justification for such exemptions was attempts to sustain relative international sta- articulated by the commander of a military bility (Cooley, Nexon, and Ward 2019; Hirst base in response to a community’s concern 2001). An absence of large-scale conflicts about pollution and land degradation: “we are minimizes disruptions to global production in the business of protecting the nation, not and trade networks (Chase-Dunn, Kawano, the environment” (Renner 1991:152). Schol- and Brewer 2000; Kentor, Clark, and Jorgen- ars have noted the potential for militaries as son 2023), further contributing to fossil fuel actors in climate governance (Jayaram and consumption and economic growth (Clark Brisbois 2021), and a growing number of the and Mahutga 2013; Givens 2018; Mahutga world’s militaries consider climate change a and Smith 2011; Vesia et al. 2021). “threat multiplier” to national security and According to Mills (1956:198), the impor- international stability (Burnett and Mach tance and influence of the military increased 2021; Machlis and Hanson 2008; Marzec through its “ascendancy” into the power elite 2016; USDOD 2010; see also CNA 2007; from the Second World War to the present. IPCC 2007; Klare 2019). In spite of this, The military “became enlarged and decisive nations with larger and more powerful mili- to the shape of the entire economic struc- taries are slow to ratify international climate ture,” and as a result, “the economic and the agreements (Givens 2014). For the United military have become structurally and deeply States, President Biden’s Executive Order interrelated, as the economy has become a 14057, signed in late 2021, directs the U.S. seemingly permanent war economy” (Mills government to reach 100 percent carbon-free 1956:215; see also Downey 2015). Others electricity by 2030, net-zero emissions by highlight the broader institutional intercon- 2050, and eliminate carbon pollution from nections between the military, the economy, federal buildings and vehicles, but exempts and the state as the core of the military- anything related to the U.S. military and industrial complex (e.g., Adams 1982; Hooks national security. 1990; Siebold 2001; Staples 2000). Rich sociological analyses indicate that the needs of the world’s militaries provide oppor- How Militarization Amplifies tunities for a variety of old and emerging the Effect of Economic Growth private-sector industries (e.g., Custers 2010; on Carbon Emissions Hooks 1994; Hooks and Bloomquist 1992). Throughout history, societies with larger National militaries facilitate scientific inquiry and more technologically advanced militar- and technological innovation, and they shape ies have utilized their coercive power in production in the private sector while simul- geopolitical contexts to secure and maintain taneously acting as downstream consumers, access to energy and other natural resources both domestically and internationally, given (Tilly 1990; see also Beckley 2010; Black the global market for armaments and mili- 2008; Boswell 1989; Boswell and Dixon tary equipment (Smart 2016; Soeters 2018; 1990; Chase-Dunn 1998; Jorgenson and see also Levine, Sen, and Smith 1994; Mills Clark 2009; Kentor 2000; Magdoff 1978; 1956; Schofer 2003; Thayer 1969; Turse McNeill 1982; Podobnik 2006). In the mod- 2008). Governments, especially in wealthier ern era, access to fossil fuels, often from nations, provide research funding to develop distant places, facilitates carbon-polluting and enhance military weapons systems. development for nations as they compete These systems include cutting-edge com- in regional and global economies (Brady, munication technologies and infrastructure Jorgenson et al. 7 for coordinating routine operations, strate- also benefits from the application of tech- gic maneuvers, data collection, cybersecurity, nologies, often initially designed for military and surveillance (Collins 1981; Foster and purposes, to commercial products for global McChesney 2014; Shaw 1988; Wills 2017). markets (Hooks 1990; McChesney 2013; Research and development linked to the Turse 2008). For instance, military spend- capital intensiveness and size of militaries ing spearheaded research and development for increase the overall resource demands of this personal computers and networking technolo- institution (Jorgenson et al. 2010; Kentor gies, giving rise to the internet and e-commerce et al. 2012; Kentor and Kick 2008; Schnai- (McQuaig and Brooks 2012; Newman 2002; berg 1980). Efforts to maintain a strategic Nowak 2011). advantage generate a path dependency, con- Militaries also provide a release valve for stantly elevating the standard of military pre- the economy, absorbing excess capacity tied to paredness (Thee 1990; U.S. Army 1999). For occurrences of carbon-polluting overproduc- example, risk-transfer militarism involves the tion in the private sector, which helps reduce development by private-sector military con- macroeconomic disruptions and stabilize over- tractors of high-tech air and undersea vehi- all economic growth (Cypher 2015; Griffin, cles, such as drones and “robot subs,” that can Devine, and Wallace 1982). Law enforce- launch missiles at designated targets (Cypher ment agencies and private security entities 2022; O’Rourke 2021). While initially used throughout the world are major clients for the by the most dominant militaries, as part of defense industry, purchasing armored vehicles, an ever-evolving arms race, such high-tech weapons, communications systems, and other equipment is increasingly in demand for mili- specialized equipment initially developed taries throughout the world. for nations’ military purposes (Avant 2005; Overall, the interrelated activities embed- Dunlap and Brock 2022; Krahmann 2010; ded within the military-industrial complex Kraska 2007; Singer 2008; Swed and Crosbie include contracts for research, development, 2019). Through the demand for services, fuel, manufacturing, and servicing of weapons and and other resources, the presence of mili- their delivery systems, transportation vehi- tary bases and installations affects surround- cles, information technology, cybersecurity, ing communities and regions, influencing their communications equipment, and other infra- carbon-polluting economic activities and structural needs (Baran and Sweezy 1966; related environmental effects (Alvarez 2021; Block 1980; Foster and McChesney 2014). Correa and Simpson 2022; Durant 2007; Each of the nodes and links in these produc- Hooks 1994; Vine 2015; Wilcox 2007). tion systems, supply chains, and ancillary In summary, we argue that the com- services involves the burning of fossil fuels plex and evolving arrangements among the and the consumption of other resources, all world’s militaries and the private sector of which are amplified by the size and capital shape the relationship between national car- intensiveness of nations’ militaries. In other bon emissions and economic growth. The words, the effects of economic growth on car- effect of economic growth on emissions is bon emissions are shaped by both the capital likely greater for nations with larger and more intensiveness and size of nations’ militaries, capital-intensive militaries. As measures that and each likely increases the extent to which capture these far-reaching and distinct char- the other enlarges the effect of growth on acteristics of contemporary militarization, carbon pollution. we posit that both military expenditures per Militaries minimize risk for industry, as soldier (i.e., capital intensiveness) and mili- they provide an assured market. They help tary participation rate (i.e., size) enlarge the “reduce towards zero the gap in time between effect of economic growth on nations’ carbon profitable original production and profitable emissions. In particular, we argue that mili- replacement” (Mumford 1963:93). Industry tary participation rate increases the extent to 8 American Sociological Review 00(0) which expenditures per solider amplifies the disproportionately responsible on a per person effect of growth on carbon pollution, and basis for the amount of carbon emitted into likewise, expenditures per solider increases the atmosphere from human activities (e.g., the extent to which participation rate enlarges IPCC 2013; Royal Society and U.S. National the effect of economic growth on emissions. Academy of Sciences 2020). Consistent with We test our arguments with moderation other sociological research (e.g., Jorgenson analysis and multiple longitudinal modeling and Clark 2012; Longhofer and Jorgenson techniques, across unbalanced and balanced 2017; Thombs 2018a; Vesia et al. 2021), we panel datasets of nations, for two measures of also estimate models of total carbon dioxide carbon dioxide emissions, and with counter- emissions (measured in kilotons), which we factual analysis. report in the Appendix. Total emissions are analogous with the overall scale of emissions and are centrally relevant for climate mitiga- DATA AnD METHoDs tion concerns (IPCC 2013; Royal Society and The Dataset U.S. National Academy of Sciences 2020). We maximize the use of available data. The Primary Independent Variables overall panel dataset consists of 2,563 annual observations for 106 nations (24.2 mean, 9 The primary independent variables for this minimum, and 27 maximum annual observa- study include gross domestic product (GDP) tions per nation) for 1990 to 2016. Due to per capita, military expenditures per soldier missing data for the different measures, the (MEPS), military participation rate (MPR), samples vary across the estimated models, their two-way interactions (GDP per capita depending on which independent variables × MEPS, GDP per capita × MPR, MEPS are included. The year 1990 is the earliest, × MPR), and most importantly, their three- and 2016 is the most recent year, in which way interaction: GDP per capita × MEPS × some of the primary independent variables MPR (Jaccard and Turrisi 2003). For ease of are currently available. Appendix Table A1 interpretation, we calculate and use the grand lists the number of observations for each mean-centered versions for these three vari- nation in the overall dataset. We also esti- ables in the reported models that include their mate and report models where we restrict interactions. the dataset to nations with no missing data, GDP per capita is measured in constant which consists of perfectly balanced panels 2010 U.S. dollars. Military expenditures per of 27 annual observations for 53 nations. All soldier is calculated by dividing total military analyzed data are publicly available, and the expenditures by total armed forces person- overall panel dataset is available from the nel. Military participation rate is measured as lead author upon request. armed forces personnel as a percent of total labor force. Military expenditures per soldier quantifies the capital intensiveness of nations’ Dependent Variables militaries, and military participation rate The primary dependent variable is carbon measures the relative size of nations’ militar- dioxide emissions per capita, which we ies (see Carlton-Ford et al. 2019; Jorgenson obtained from the World Bank’s online World and Clark 2009; Jorgenson at al. 2010; Kentor Development Indicators Database (World et al. 2012; Kentor and Kick 2008; Kick et al. Bank 2022). These data, measured in metric 1998; Lengefeld and Smith 2013; Smith and tons per person, include emissions from the Lengefeld 2020). For the overall dataset, they burning of fossil fuels and the manufacture of are weakly correlated at –.11 in their original cement. Per capita emissions is commonly metrics and .01 in logarithmic form. used as a measure of international inequality Total military expenditures are measured in in emissions as it quantifies how nations are constant 2018 U.S. dollars and obtained from Jorgenson et al. 9 Stockholm International Peace Research Insti- To further enhance the validity of the tute’s online Military Expenditure Database hypotheses testing, we estimate models that (SIPRI 2022). These data include expenditures also control for military expenditures as a on personnel, operations and maintenance, pro- percent of general government expenditures. curement, military research and development, This additional variable, which we obtained military infrastructure spending (including from the World Bank (2022), is moderately military bases), and military aid (in the mili- correlated with military participation rate tary expenditure of the donor country). They (.513) and weakly correlated with military exclude civil defense and current expenditures expenditures per soldier (.074). on previous military activities, demobilization, conversion, and weapon destruction. Armed Model Estimation Techniques forces personnel consist of active-duty military personnel, including paramilitary forces if the We estimate and report two-way fixed-effects training, organization, equipment, and control regression models with robust standard errors suggest they may be used to support or replace clustered by nation, correcting for unobserved regular military forces. Measures of GDP per heterogeneity that is time-invariant within capita, total armed forces personnel, and mili- nations as well as cross-sectionally invariant tary participation rate come from the World within years. We estimate the models with Bank (2022). the xtreg command in Stata software, which uses the within estimator to account for the country-level fixed effects, and the temporal Additional Independent Variables fixed effects are derived from the inclusion The reported models include a variety of of year-specific dummy variables. Consistent additional independent variables common in with the majority of sociological research on sociological research on the human drivers the anthropogenic drivers of national emis- of carbon emissions (Dietz at al. 2020; Rosa sions (see Jorgenson et al. 2019; Rosa and and Dietz 2012). Each model includes urban Dietz 2012), we transform all nonbinary vari- population as a percent of the total popu- ables into logarithmic form. This means the lation, non-dependent population (percent models estimate elasticity coefficients where of the total population age 15 to 64), and the coefficient for the independent variable services as a percent of GDP, all obtained is the estimated net percentage change in from the World Bank (2022). Prior studies the dependent variable associated with a 1 generally find that both urban population percent increase in the independent variable. and non-dependent population are positively Appendix Table A2 provides descriptive sta- associated with emissions, and services as tistics for the substantive variables included percent of GDP is negatively associated with in the study. All variable transformation infor- carbon pollution. mation and the Stata code used to estimate the Additional models include trade as percent reported models are available from the lead of GDP, also obtained from the World Bank author upon request. (2022), and level of democratization in the The baseline model we estimate for per form of the institutionalized democracy index, capita emissions is as follows: an additive 11-point scale (with higher values meaning greater levels of democracy), which CO Emissions percapita 2 it , we obtained from the Center for Systemic Peace =+ ββ GDP percapita MEPS 12 it ,, it and Societal-Systems Research (2018). Total + β MPR + β Urban Population population, which counts all residents regard- 3 iit ,, 4 it less of legal status or citizenship, is included + β Nond - ependent Population 5 it , in the models of total carbon emissions. These + β Seervices %. GDP ++ αε u + 6 it ,, it it data come from the World Bank (2022). (1) 10 American Sociological Review 00(0) The baseline model with the inclusion of the significant in Models 3 through 5. The esti- three-way interaction is as follows: mated effect of non-dependent population on per capita emissions is positive across all five models, the effect of services as percent of CO Emissions percapita 2 it , GDP is negative in all but the second and fifth =+ ββ GDP percapita MEPS 12 it ,, it models, and the effect of urban population + β MPR + β GDP percapita *MEPS 3 iit ,, 4 it it , is positive and statistically significant in the + β GDP percapita *MPR 5 it ,, it first two models. The results, particularly the + β M MEPS *MPR significant coefficient for GDP per capita × 6 it ,, it MEPS × MPR, confirm our arguments. + β GDP percapita ** MEPS MPR 7 it ,, it it , To provide a more nuanced assessment + β Urban Poppulation 8 it , and clearer interpretation of the three-way + β Nond - ependent Population 9 it , interaction, Figures 1 and 2 plot the average + β Services %GDP ++ αε u + . 10 i,,ti ti,t marginal effects, with 95 percent confidence (2) intervals (95 percent CI), of GDP per capita by MEPS and MPR. The estimates are based REsuLTs on Model 3 in Table 1, which we generate using Stata’s margins command. Although Table 1 reports five models of per capita the 95 percent confidence intervals of the carbon emissions. Model 1 is the initial estimates overlap for most of the marginal baseline, consisting of GDP per capita, mili- effects, the differences between the point esti- tary expenditures per soldier (MEPS), and mates for the marginal effects are statistically military participation rate (MPR), as well significant unless noted otherwise. as urban population, non-dependent popula- Figure 1 reports the marginal effect of tion, and services as percent of GDP. Model GDP per capita on per capita emissions at 2 introduces each of the two-way interac- the 10th, 50th, and 90th percentiles of MEPS tions for GDP per capita, MEPS, and MPR. across levels of MPR. The marginal effect of Models 3 through 5 include their three-way GDP per capita at each percentile of MEPS interaction, with Model 3 for the overall panel increases across the MPR distribution, with dataset of 106 nations and Model 4 for the the exception of the 10th percentile of MEPS perfectly balanced panel dataset reduced to ($3,315). The effects in this case are statisti- 53 nations. Model 5 is for the overall dataset, cally equivalent across the MPR distribution, and also controls for trade as percent of GDP, ranging from .278 (95 percent CI = .185 to democratization, and military expenditures as .370) at the 10th percentile of MPR to .255 percent of government expenditures. For ease (95 percent CI = .151 to .359) at the 90th of interpretation, we exclude the estimated percentile of MPR. At the 50th percentile of coefficients for these three additional controls MEPS ($20,075), the effect of GDP per capita in Table 1 (all not statistically significant), but on emissions ranges from .303 (95 percent they are provided in Appendix Table A3. CI = .222 to .384) at the 10th percentile of Model 1 indicates that per capita emissions MPR (1.28 percent) to .385 (95 percent CI = .290 is positively associated with GDP per capita to .480) at the 90th percentile of MPR (4.31 and MEPS, and the effect of MPR is not statis- percent). At the 90th percentile of MEPS tically significant. In Model 2, the estimated ($179,553), the effect of GDP per capita coefficients for GDP per capita × MEPS and ranges from .333 (95 percent CI = .248 to GDP per capita × MPR are positive, whereas .418) at the 10th percentile of MPR (1.28 per- the coefficient for MEPS × MPR is not statis- cent) to .544 (95 percent CI = .432 to .656) tically significant. The estimated coefficient at the 90th percentile of MPR (4.31 percent). for the three-way interaction, GDP per capita Figure 2 provides the marginal effect of × MEPS × MPR, is positive and statistically GDP per capita on per capita carbon emissions Jorgenson et al. 11 Table 1. Elasticity Coefficients for the Regression of Carbon Emissions per Capita, 1990 to Model 1 Model 2 Model 3 Model 4 Model 5 *** *** *** *** *** GDP per Capita .400 .333 .342 .505 .361 (.045) (.044) (.041) (.074) (.056) *** *** *** * ** Military Expenditures per Soldier (MEPS) .068 .098 .097 .063 .097 (.014) (.020) (.018) (.026) (.031) Military Participation Rate (MPR) .023 .070 .001 –.019 –.067 (.043) (.049) (.047) (.068) (.081) *** *** ** *** GDP per Capita × MEPS .030 .039 .063 .045 (.009) (.009) (.019) (.010) * * * ** GDP per Capita × MPR .065 072 .142 .126 (.026) (.030) (.060) (.046) MEPS × MPR .015 .024 .032 –.023 (.019) (.023) (.047) (.036) *** ** *** GDP per Capita × MEPS × MPR .048 .073 .053 (.012) (.026) (.016) * * Urban Population .226 .265 .213 .330 .273 (.109) (.114) (.113) (.192) (.140) *** *** *** *** *** Non-dependent Population 1.116 1.093 1.055 1.251 1.271 (.274) (.271) (.251) (.315) (.278) * * ** Services as % GDP –.123 –.105 –.111 –.320 –.152 (.058) (.055) (.053) (.114) (.084) R Overall .825 .823 .820 .786 .841 Note: For Models 1, 2, and 3, N = 2,563 for 106 nations, with 24.2 mean observations per nation. For Model 4, N = 1,431 for 53 nations, with 27 mean observations per nation. For Model 5, N = 2,079 for 100 nations, with 20.8 mean observations per nation. Model 5 also controls for trade as % GDP, democratization, and military expenditures as % government expenditures. Robust standard errors clustered by nation are in parentheses. GDP per capita, MEPS, and MPR are mean centered. All variables are in logarithmic form. All models include nation-specific fixed effects derived from the within estimator. All models include unreported year-specific intercepts. * ** *** p < .05; p < .01; p < .001 (two-tailed). at the 10th, 50th, and 90th percentiles of (4.31 percent), the effect of GDP per capita MPR across levels of MEPS. Like in Figure ranges from .255 (95 percent CI = .151 to 1, the marginal effect of GDP per capita at .359) at the 10th percentile of MEPS ($3,315) each percentile of MPR increases across the to .544 (95 percent CI = .432 to .656) at the MEPS distribution, with the exception of the 90th percentile of MEPS ($179,553). 10th percentile of MPR (1.28 percent). The statistically equivalent effects in this case Additional Models range from .277 (95 percent CI = .185 to .370) at the 10th percentile of MEPS to .333 To further broaden the testing of the three- (95 percent CI = .248 to .418) at the 90th way interaction, we estimate models for per percentile of MEPS. At the 50th percentile capita emissions and total emissions that of MPR (2.01 percent), the effect of GDP per include additional controls, which we report capita on emissions ranges from .269 (95 per- in Appendix Table A6. First, we estimate cent CI = .185 to .354) at the 10th percentile models that control for renewable energy of MEPS ($3,315) to .412 (95 percent CI = consumption. Next, we control for arms .325 to .499) at the 90th percentile of MEPS exports. Due to their unavailability for many ($179,553). At the 90th percentile of MPR nations, including arms exports greatly 12 American Sociological Review 00(0) Average Marginal Effects of GDP per capita by Military Expenditures per Soldier $3,315 per Soldier $20,075 per Soldier $179,553 per Soldier 1.28% 2.01% 4.31% 1.28% 2.01% 4.31% 1.28% 2.01% 4.31% Military Participation Rate Figure 1. Marginal Effects of GDP per Capita for Model of CO Emissions per Capita by Military Expenditures per Soldier Note: $3,315, $20,075, and $179,553 are the 10th, 50th, and 90th percentiles for the distribution of military expenditures per soldier; 1.28 percent, 2.01 percent, and 4.31 percent are the 10th, 50th, and 90th percentiles for the distribution of military participation rate. reduces the overall sample. Finally, we esti- and indicate that the estimated effect of the mate models that control for oil production. two-way interaction on both per capita emis- The oil production data are also limited to a sions and total emissions is not statistically relatively small number of nations. Across significant. These findings further validate each model of emissions, the estimated coef- our focus on military expenditures per solider ficient for GDP per capita × MEPS × MPR and military participation rate as far-reaching is positive and statistically significant. As and distinct characteristics of militarization expected, the estimated effect of renewable that shape the effect of economic growth on energy consumption is negative and statisti- nations’ carbon pollution. cally significant. The estimated effects of arms exports and oil production are not sta- Robustness Checks and Sensitivity tistically significant. Analyses To determine if other national military measures moderate the effect of economic An interaction in fixed-effects regression is growth on emissions, we estimate models usually specified by demeaning the prod- that include the two-way interaction between uct term. Giesselmann and Schmidt-Catran GDP per capita and military expenditures as (2022) show that demeaning the product percent of government expenditures. As noted between time-varying variables may not pro- in the Data and Methods section, this mili- duce a true within-unit estimate because it tary measure is moderately correlated with incorporates between-unit differences. They MPR and weakly correlated with MEPS. The propose using the double-demeaned esti- models are reported in Appendix Table A7 mator, which gives unbiased results but is Effects on Linear Prediction Jorgenson et al. 13 Average Marginal Effects of GDP per capita by Military Participation Rate 1.28% Participation Rate 2.01% Participation Rate 4.31% Participation Rate $3,315 $20,075 $179,553 $3,315 $20,075 $179,553 $3,315 $20,075 $179,553 Military Expenditures per Soldier Figure 2. Marginal Effects of GDP per Capita for Model of CO Emissions per Capita by Military Participation Rate Note: 1.28 percent, 2.01 percent, and 4.31 percent are the 10th, 50th, and 90th percentiles for the distribution of military participation rate; $3,315, $20,075, and $179,553 are the 10th, 50th, and 90th percentiles for the distribution of military expenditures per soldier. inefficient compared to the fixed-effects esti- .16, which fails to reject the null hypothesis mator. They suggest using a Hausman test that the estimates are statistically equivalent to decide which estimator is more appropri- (p-value = .69). We therefore rely on the ate. If the estimators produce statistically fixed-effects estimates of the three-way inter- identical estimates, then the standard fixed- action in the reported analyses. effects estimator should be used. As a robust- A possible limitation of using year-specific ness check, we perform the double-demeaned fixed effects is that they assume time-specific estimator and extend it using a seemingly shocks homogeneously affect each case in unrelated regression framework. This has the dataset, meaning they may not adequately two advantages over using a Hausman test. model the cross-sectional dependence, poten- First, it allows for robust standard errors, tially leading to biased and inconsistent which the Hausman test does not. Second, it results. Accordingly, we use Pesaran’s test allows us to use a simple Wald test to assess for weak cross-sectional dependence to assess whether the coefficient on the three-way whether the year-specific fixed effects elimi- interaction of interest is statistically different nate the strong cross-sectional dependence across the two models. In contrast, the Haus- from our reported models. The cross-sectional man test assesses the equality of two estima- dependence test statistics of the residuals are tors instead of individual coefficients. We not statistically significant, meaning there is perform this approach for Model 3 in Table no strong cross-sectional dependence, and 1, combining the results of the two estimators the reported two-way fixed-effects models using the suest command in Stata. The Wald are unlikely biased or inconsistent in this way test produces a chi-square test statistic of (Thombs 2022). Effects on Linear Prediction 14 American Sociological Review 00(0) Panel data are often autoregressive, mean- The CCE estimator assumes the cross-sec- ing the data tend to be correlated over time, tional dependence is due to unobserved, time- and excluding the lag of the dependent vari- varying, common factors that affect each case able from the model will result in omitted differently. It approximates the common fac- variable bias if the outcome variable is truly tors by adding cross-sectional averages to the a function of their past value (Pickup 2015). model and estimates a factor loading for each Therefore, as robustness checks, we estimate case in the analysis (Pesaran 2006). We use two-way fixed-effects dynamic models for the pooled version of the CCE because the both per capita emissions and total emissions, relatively short time span prevents us from with a focus on the coefficient for the three- estimating a time-series regression on each way interaction. individual nation (Thombs 2022). For per capita emissions, this model is as A limitation of using the pooled version follows: is that it does not account for the potential issue of slope heterogeneity. We test for this CO Emissions percapita using the instrumental-variable approach, a 2 it , two-stage procedure that works by eliminat- = λ CO Emissions percapita 12 it , −1 ing the common factors in the covariates + β GDP Pper capita + β MEPS 1 it ,, 2 it using principal component analysis in stage ++ ββ MPR GDP percapita *MEPS 34 it ,, it i,tt one, and obtains consistent estimates using + β GDP percapita *MPR defactored covariates as instruments (Norkutė 5 it ,, it et al. 2021). In stage two, the whole model is + β MEPS *MPR 6 it ,, it defactored using the residuals from stage one, + β GDP percapitaaM ** EPS MPR 7 it ,, it it , and instrumental-variable estimation is per- + β Urban Population 8 it , formed using the same instruments from the + β Nond -o ependent Population 21 9 it , first stage (Kripfganz and Sarafidis 2021). ++ βα Services %. GDP ++ u ε This estimation technique is robust to Nickell 10 it ,, it it (3) bias, and we test the effect of slope hetero- There are potential issues to consider regard- geneity on the model with the Hansen test of ing fixed-effects estimation of a dynamic overidentifying restrictions (J-statistic). model. First, estimating dynamic panel mod- Table 2 reports the estimates of the four els can produce the “Nickell bias” (Nickell robustness check models for per capita emis- 1981). The bias stems from the correlation sions. Appendix Table A8 reports the same between the lagged dependent variable and sequence of estimated models for total carbon the error term, a product of the demeaning emissions. Model 1 is for the static pooled process of fixed-effects estimation. However, common correlated effects estimator, and the this bias tends to lessen as T increases (Hsiao, remaining three are dynamic models and thus Pesaran, and Tahmiscioglu 2002; Thombs include the lagged dependent variable. Model 2022). Second, fixed-effects estimation of a 2 is for the two-way fixed-effects dynamic dynamic model with slope heterogeneity can model, Model 3 is for the dynamic pooled lead to inconsistent and misleading estimates common correlated effects estimator, and (Pesaran and Smith 1995; Thombs, Huang, Model 4 is for the two-stage instrumental- and Fitzgerald 2022). variable estimator. To address these concerns and to model The estimated coefficient for GDP per cross-sectional dependence in alterna- capita × MEPS × MPR is positive and statis- tive ways, we also estimate models with tically significant in each model, regardless of the common correlated effects (CCE) esti- estimator type. The lagged dependent variable mator (Ditzen 2018; Pesaran 2006) and the has a positive effect in each dynamic model, instrumental-variable estimation approach and the J-statistic is not statistically signifi- with common factors (Norkutė et al. 2021). cant in the two-stage instrumental-variable Jorgenson et al. 15 Table 2. Elasticity Coefficients for the Regression of Carbon Emissions per Capita, 1990 to Model 1 Model 2 Model 3 Model 4 ** *** *** *** GDP per Capita × MEPS × MPR .022 .013 .019 .016 (.009) (.004) (.005) (.005) *** *** *** *** GDP per Capita .526 .119 .084 .277 (.054) (.017) (.016) (.033) *** *** * *** Military Expenditures per Soldier (MEPS) .046 .027 .018 .026 (.013) (.006) (.008) (.007) * ** * Military Participation Rate (MPR) .006 .032 .049 .051 (.038) (.015) (.018) (.020) ** * ** GDP per Capita × MEPS .024 .007 .005 .014 (.009) (.003) (.004) (.005) ** GDP per Capita × MPR .035 .010 .018 .032 (.022) (.011) (.011) (.012) * ** ** MEPS × MPR .033 .015 .022 .026 (.014) (.008) (.007) (.010) *** *** *** CO per Capita Lagged .716 .712 .373 (.027) (.030) (.065) R Overall .983 J-statistic 7.066 Note: Model 1 is for the static pooled common correlated effects estimator. Model 2 is for the two-way fixed-effects dynamic model with clustered robust standard errors. Model 3 is for the dynamic pooled common correlated effects estimator. Westerlund, Perova, Norkute standard errors are reported for Models 1 and 3. Model 4 is for the two-stage instrumental-variable estimation with two lags used as instruments. For Model 1, N = 2,439 for 96 nations, with 26 mean observations per nation. For Model 2, N = 2,477 for 106 nations, with 23.4 mean observations per nation. For Model 3, N = 2,096 for 96 nations, with 22 mean observations per nation. For Model 4, N = 2,256 for 106 nations, with 21.3 mean observations per nation. Due to insufficient observations, the analysis drops 10 nations from Models 1 and 3. Standard errors are in parentheses. All variables are in logarithmic form. J-statistic for Model 4 is not statistically significant (H : overidentifying restrictions are valid). All models control for urban population, non-dependent population, and services as % GDP. GDP per capita, MEPS, and MPR are mean centered. * ** *** p < .05; p < .01; p < .001 (two-tailed). models. Overall, the findings of interest for Counterfactual Analysis and this study appear robust to a variety of poten- Substantive Significance tial modeling concerns. Finally, to determine if the analyses and Having demonstrated that our results are findings are sensitive to any particular nations robust to a host of modeling considerations included in the study, we re-estimate each and not sensitive to sample characteristics, we reported model where we systematically now turn to the question of substantive sig- exclude, one at a time, each of the 106 nations nificance. Here we ask how the moderating in the overall dataset. The results indicate that effect of militarization matters for observed none of the included nations are overly influ- levels of carbon emissions per capita. In par- ential: the estimated elasticity coefficients ticular, we use Model 3 of Table 1 to engage for the three-way interaction across all re- in a counterfactual history exercise under two estimated models are positive and statistically scenarios for the overall dataset. First, we significant. The estimated coefficients for the ask what average emissions per capita would other independent variables remain consistent look like if every nation in the sample had as well. military expenditures per soldier and military 16 American Sociological Review 00(0) Figure 3. Average CO Emissions per Capita under Different Militarization Scenarios Note: Estimates derived from Model 3 in Table 1 for the overall dataset. 10th percentile militarization refers to average carbon emissions per capita if every nation in the sample had military expenditures per soldier and military participation rates equal to that observed at the 10th percentile of the nation- year distribution. 90th percentile militarization refers to average emissions per capita if every nation in the sample had military expenditures per soldier and military participation rates equal to that observed at the 90th percentile of the nation-year distribution. Carbon emissions reported in metric tons per capita. participation rates equal to that observed at average per capita emissions grew by .614 the 90th percentile of the nation-year distribu- metric tons over the entire period. This num- tion. Second, we ask what average emissions ber falls to .419 metric tons in a world of 10th per capita would look like if every nation in percentile militarization, and the increase the sample instead had military expenditures rises to 1.14 metric tons in a world of 90th per soldier and military participation rates percentile militarization. Holding rates of equal to that observed at the 10th percentile economic growth fixed, worldwide reduc- of the nation-year distribution. tions in the capital intensiveness and size Figure 3 reports the yearly average per of militarization could produce substantial capita carbon emissions under each of these declines in carbon emissions. scenarios, as well as the observed average per capita emissions per year. Consistent with DisCussion AnD our overall intervention, there is a wide gap ConCLusions between the observed emissions and those that would occur under militarization at the 10th This study makes significant contributions and 90th percentiles. In 1990, observed aver- to sociological work on climate change and age emissions are .705 metric tons per capita the climate crisis. Bridging multiple sub- higher than that which would occur under mili- fields, including environmental sociology, tarization at the 10th percentile, and this gap the sociology of development, global political grows to .899 metric tons per capita by 2016. economy, historical sociology, and politi- Conversely, observed emissions are .459 metric cal sociology, we argue that militarization tons per capita lower than what would occur moderates the effect of economic growth with 90th percentile militarization. This gap on nations’ carbon emissions. Many of the grows to .984 metric tons per capita by 2016. world’s national militaries are increasingly The growth rate for emissions also varies capital intensive, with a focus on the develop- considerably under these scenarios. Observed ment of longer-range weapons, transportation, Jorgenson et al. 17 and communications systems, and larger mili- for military purposes are often sold to private taries possess expansive built infrastructures security firms and law enforcement agencies, that require considerable amounts of energy and transformed into commercial products and material goods. We use military expen- for domestic and global markets. Military ditures per soldier and military participation bases and installations routinely obtain fuel rate to measure these far-reaching character- as well as various material goods and services istics of contemporary militarization. from business entities. These structural condi- The findings for the longitudinal analyses tions, institutional relationships, and underly- provide substantial support for our arguments. ing processes all contribute to how the capital Both military expenditures per solider and intensiveness and scale of militaries shape the military participation rate enlarge the posi- association between national emissions and tive effect of economic growth on national economic growth. carbon emissions. We observe these relation- Anthropogenic climate change increases ships through modeling the two-way interac- the likelihood of large-scale conflicts between tions for growth, measured as GDP per capita, and within nations (Alario, Nath, and Carlton- and each military measure. More importantly, Ford 2016; Cane et al. 2014; Giddens 2011; through modeling their three-way interac- Hsiang, Burke, and Miguel 2013; Mach et al. tion, we find that each militarization attribute 2019). This expands and intensifies military increases the extent to which the other ampli- activities for nations involved in international fies the effect of economic growth on carbon and domestic engagements (Belcher, Bigger, pollution. The effect of GDP per capita on et al. 2020; Pathak 2020; Raleigh and Urdal emissions is larger at higher levels of expen- 2007; Smith and Lengefeld 2020), further ditures per solider, and this increases across driving the material- and energy-intensive the distribution of military participation rate. production of weapons systems and muni- Likewise, the effect of GDP per capita on tions, vehicles, communications equipment, carbon pollution is larger at higher levels of and other related goods in the defense indus- military participation rate, and this increases try and private sector more broadly (Isik- across the distribution of military expendi- sal 2021; Jorgenson and Clark 2016). Thus, tures per solider. The results are robust for per as a threat multiplier, anthropogenic climate capita emissions and total emissions, various change could facilitate a greater occurrence sensitivity analyses, a range of balanced and of both domestic and international conflicts, unbalanced panel datasets, and across mul- further propelling the relationships between tiple model specifications. Their substantive national carbon emissions, economic growth, significance is further highlighted through and militarization. counterfactual analysis. Our theoretical arguments and empirical Our findings speak to the deep connec- findings highlight the value and necessity of tions between the military and the economy considering the world’s militaries in sociolog- at the core of the military-industrial complex. ical research. The work of historical sociolo- National militaries help secure access to fos- gists maps out in great detail the emergence sil fuels and other resources, and generally of nation-states from a coalescence of coer- attempt to maintain geopolitical and world- cive power with economic power. Inspired by economic stability, which enables carbon- this rich body of scholarship, we concentrate intensive economic growth. At the same time, on the environmental effects of economy and militaries spur scientific research and techno- military relationships for nations in the mod- logical advances, and they influence produc- ern era. The present study focuses on human tion in the defense industry and private sector drivers of climate change, but contemporary in general, while also serving as major con- forms of militarization likely shape the effect sumers of these items. The technologies and of economic growth on social and other envi- other goods initially developed by industry ronmental outcomes in ways similar to and 18 American Sociological Review 00(0) distinct from how it influences the associa- contemporaneous relationships observed for tion between carbon pollution and economic recent decades. While we use the best pub- growth. licly available aggregate data on national- The characteristics of national militaries level militarization characteristics, these data also likely influence how macrostructural fac- might be underestimated for some country- tors and processes besides economic growth years due to accounting practices and the affect a range of social and environmen- overall classified nature of military-related tal outcomes. Like others (e.g., Andreski information. Consequently, the reported find- 1968; Giddens 1987; Hooks 1990; Kentor ings may underestimate the direct association and Kick 2008), we suggest the military is between national-level emissions and mili- routinely overlooked by scholars across the tarization, as well as the extent to which the discipline. We hope this study will encour- capital intensiveness and size of militariza- age sociologists to consider the military and tion enlarge the effect of economic growth on militarization in future analytic frameworks carbon pollution. and empirical analyses. The ongoing growth In conclusion, this study significantly of militarism underscores the importance in advances the sociological research on climate doing so. From 1990 to 2016, global military change, and enhances sociological contribu- expenditures increased by 29 percent (1.372 tions to interdisciplinary work on planetary to 1.774 trillion constant 2018 U.S. dollars), warming and other global sustainability chal- armed forces personnel for the world grew by lenges. Militaries exert a substantial influence 15 percent (23.918 to 27.542 million person- on the production and consumption patterns nel), and global military expenditures per of economies, as well as the environmental solider increased by 12 percent (57,362 to demands required to support their evolving 64,410 constant 2018 U.S. dollars). infrastructure. Our findings indicate that two Like all research, this study has limi- major characteristics of militarization enlarge tations that can hopefully be addressed in the effect of economic growth on carbon future analyses. Although the overall sample emissions, and they increase the extent to covers the majority of the world’s popula- which the other amplifies the effect of growth tion, current data availability limits the num- on nations’ carbon pollution. By theorizing ber of nations included in the cross-national about these structural relationships and bridg- analyses. Data availability also restricts the ing multiple subfields, we push forward the temporality of the study to slightly over a foundational sociological literature concern- quarter century, from 1990 to 2016. Thus, our ing the effect of economic growth on the analyses focus on the modeling of relatively environment. Jorgenson et al. 19 APPEnDiX Table A1. Number of Annual Observations for Each Country in the Overall Dataset Country Obs. Country Obs. Country Obs. Albania 25 Gabon 13 Netherlands 27 Algeria 27 Gambia, The 23 New Zealand 27 Angola 26 Georgia 21 Nicaragua 27 Argentina 27 Germany 26 Niger 20 Armenia 24 Ghana 26 Nigeria 27 Australia 27 Greece 27 Norway 27 Austria 27 Guinea 17 Oman 27 Bahrain 27 Guinea-Bissau 18 Peru 27 Bangladesh 27 Guyana 24 Philippines 27 Belarus 25 Haiti 9 Portugal 27 Belgium 27 Honduras 20 Romania 27 Belize 25 Hungary 26 Russian Federation 25 Benin 14 India 27 Rwanda 26 Bosnia and Herzegovina 15 Iran, Islamic Rep. 27 Saudi Arabia 27 Botswana 27 Ireland 27 Serbia 11 Brazil 27 Israel 27 Sierra Leone 25 Bulgaria 27 Italy 25 Slovenia 25 Cameroon 26 Jamaica 27 South Africa 27 Canada 27 Japan 27 Spain 27 Chile 27 Jordan 27 Sudan 20 China 27 Kazakhstan 24 Sweden 27 Colombia 27 Kenya 26 Switzerland 26 Congo, Dem. Rep. 22 Korea, Rep. 27 Tajikistan 20 Congo, Rep. 13 Lebanon 27 Tanzania 26 Cote d'Ivoire 15 Lesotho 27 Timor-Leste 11 Croatia 22 Lithuania 22 Togo 18 Cuba 12 Madagascar 26 Trinidad and Tobago 18 Cyprus 25 Malaysia 27 Tunisia 27 Denmark 27 Mali 25 Turkey 27 Dominican Republic 27 Mexico 27 United Arab Emirates 18 Ecuador 27 Moldova 22 United Kingdom 27 El Salvador 27 Montenegro 11 United States 27 Ethiopia 26 Morocco 27 Uruguay 26 Fiji 27 Namibia 26 Yemen, Rep. 25 Finland 27 Nepal 27 Zambia 22 France 25 Note: N = 2,563. Obs. = number of annual observations. 20 American Sociological Review 00(0) Table A2. Descriptive Statistics Std. N Mean Deviation Carbon Emissions per Capita 2,563 1.390 .899 Total Carbon Emissions 2,563 10.187 2.221 GDP per Capita 2,563 8.590 1.525 Military Expenditures per Soldier 2,563 9.989 1.429 Military Participation Rate 2,563 .772 .468 Urban Population 2,563 3.988 .465 Non-dependent Population 2,563 4.121 .111 Services as % GDP 2,563 3.934 .238 Trade as % GDP 2,434 4.183 .502 Democratization 2,460 1.705 .877 Military Expenditures as % Government Expenditures 2,269 1.962 .615 Total Population 2,563 16.348 1.559 Renewable Energy Consumption 2,303 2.971 1.29 Arms Exports 709 18.673 2.245 Oil Production 801 3.732 1.36 Note: All variables are in logarithmic form. Table A3. Elasticity Coefficients for the Regression of Carbon Emissions per Capita, 1990 to Model 1 Model 2 Model 3 Model 4 *** *** *** *** GDP per Capita × MEPS × MPR .049 .051 .050 .053 (.012) (.013) (.015) (.016) *** *** *** *** GDP per Capita .352 .349 .360 .361 (.042) (.042) (.056) (.056) *** *** *** ** Military Expenditures per Soldier (MEPS) .088 .096 .100 .097 (.019) (.018) (.028) (.031) Military Participation Rate (MPR) –.034 –.012 –.002 –.067 (.049) (.050) (.068) (.081) *** *** *** *** GDP per Capita × MEPS .035 .040 .045 .045 (.009) (.009) (.010) (.010) * * * ** GDP per Capita × MPR .082 .074 .082 .126 (.037) (.035) (.036) (.046) MEPS × MPR .008 .022 .011 –.023 (.026) (.026) (.032) (.036) Trade as % GDP .002 .020 (.025) (.027) Democratization .010 –.002 (.011) (.014) Military Expenditures as % Government –.014 –.014 Expenditures (.035) (.039) R Overall .833 .817 .827 .841 Note: All models control for urban population, non-dependent population, and services as % GDP. Robust standard errors clustered by nation are in parentheses. All variables are in logarithmic form. All models include nation-specific fixed effects derived from the within estimator. All models include unreported year-specific intercepts. GDP per capita, MEPS, and MPR are mean centered. For Model 1, N = 2,434 for 103 nations with 23.6 mean observations per nation. For Model 2, N = 2,460 for 104 nations with 23.7 mean observations per nation. For Model 3, N = 2,269 for 105 nations with 21.6 mean observations per nation. For Model 4, N = 2,079 for 100 nations with 20.8 mean observations per nation. * ** *** p < .05; p < .01; p < .001 (two-tailed). Jorgenson et al. 21 Table A4. Elasticity Coefficients for the Regression of Total Carbon Emissions, 1990 to 2016 Model 1 Model 2 Model 3 ** * *** GDP per Capita × MEPS × MPR .063 .094 .081 (.020) (.036) (.024) *** *** *** GDP per Capita .640 .644 .535 (.095) (.102) (.096) * ** Military Expenditures per Soldier (MEPS) .084 –.018 .144 (.036) (.040) (.054) Military Participation Rate (MPR) –.001 –.058 .030 (.072) (.106) (.138) GDP per Capita × MEPS –.006 .034 .013 (.015) (.024) (.018) GDP per Capita × MPR –.009 .010 .082 (.060) (.097) (.083) MEPS × MPR .070 .086 –.032 (.041) (.065) (.063) * ** Urban Population .626 .824 .453 (.303) (.305) (.315) *** * Non-dependent Population .598 1.381 1.031 (.357) (.405) (.408) * * Services as % GDP –.261 –.343 –.157 (.102) (.154) (.126) *** *** *** Total Population 1.418 1.141 1.400 (.141) (.204) (.175) R Overall .877 .922 .888 Note: For Model 1, N = 2,563 for 106 nations with 24.2 mean observations per nation. For Model 2, N = 1,431 for 53 nations with 27 mean observations per nation. For Model 3, N = 2,079 for 100 nations with 20.8 mean observations per nation. Model 3 also controls for trade as % GDP, democratization, and military expenditures as % government expenditures. Robust standard errors clustered by nation are in parentheses. GDP per capita, MEPS, and MPR are mean centered. All variables are in logarithmic form. All models include nation-specific fixed effects derived from the within estimator. All models include unreported year-specific intercepts. * ** *** p < .05; p < .01; p < .001 (two-tailed). 22 American Sociological Review 00(0) Table A5. Pairwise Differences of Marginal Effects Pairwise Difference Figure 1 Figure 2 10th MEPS MPR 10th vs. 50th –.46 MPR 10th vs. 90th –.46 MPR 50th vs. 90th –.46 50th MEPS MPR 10th vs. 50th 2.24 MPR 10th vs. 90th 2.24 MPR 50th vs. 90th 2.24 90th MEPS *** MPR 10th vs. 50th 4.78 *** MPR 10th vs. 90th 4.78 *** MPR 50th vs. 90th 4.78 10th MPR MEPS 10th vs. 50th 1.42 MEPS 10th vs. 90th 1.42 MEPS 50th vs. 90th 1.42 50th MPR *** MEPS 10th vs. 50th 4.26 *** MEPS 10th vs. 90th 4.26 *** MEPS 50th vs. 90th 4.26 90th MPR *** MEPS 10th vs. 50th 5.58 *** MEPS 10th vs. 90th 5.58 *** MEPS 50th vs. 90th 5.58 Note: z-statistics reported. * ** *** p < .05; p < .01; p < .001 (two-tailed). 23 Table A6. Elasticity Coefficients for the Regression of Carbon Emissions per Capita and Total Carbon Emissions, 1990 to 2015 and 1990 to 2016 Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 ** ** *** ** ** ** GDP per Capita × MEPS × MPR .032 .051 .080 .069 .111 .140 (.010) (.018) (.020) (.024) (.033) (.050) *** *** *** *** *** *** GDP per Capita .335 .476 .536 .667 .431 .780 (.034) (.086) (.088) (.094) (.076) (.123) *** * ** *** Military Expenditures per Soldier (MEPS) .073 .067 .087 .088 .113 .150 (.016) (.029) (.044) (.044) (.035) (.053) Military Participation Rate (MPR) .005 .008 .124 .167 –.107 .020 (.038) (.061) (.088) (.096) (.104) (.132) *** *** ** ** GDP per Capita × MEPS .037 –.008 .067 .056 .063 .056 (.007) (.014) (.018) (.018) (.021) (.033) GDP per Capita × MPR .041 –.038 .030 –.042 .008 .095 (.024) (.054) (.053) (.076) (.042) (.051) ** *** ** *** MEPS × MPR .026 .052 .167 .212 .117 .192 (.018) (.035) (.053) (.059) (.040) (.045) *** *** *** Total Population 1.211 1.569 2.051 (.128) (.195) (.212) *** *** Renewable Energy Consumption –.211 –.310 (.023) (.042) Arms Exports .005 .004 (.004) (.005) Oil Production .058 .062 (.036) (.039) R Overall .888 .911 .454 .888 .713 .802 Note: Odd numbered models are for carbon emissions per capita. Even numbered models are for total carbon emissions. All models control for urban population, non-dependent population, and services as % GDP. Robust standard errors clustered by nation are in parentheses. GDP per capita, MEPS, and MPR are mean centered. All variables are in logarithmic form. All models include nation-specific fixed effects derived from the within estimator . All models include unreported year-specific intercepts. For Models 1 and 2, N = 2,303 for 106 nations with 21.7 mean observations per nation. For Models 3 and 4, N = 709 for 28 nations with 25.3 mean observations per nation. For Models 5 and 6, N = 801 for 32 nations with 25.0 mean observations per nation. Models 1 and 2 are for 1990 to 2015; Models 3 to 6 are for 1990 to 2016. * ** *** p < .05; p < .01; p < .001 (two-tailed). 24 American Sociological Review 00(0) Table A7. Elasticity Coefficients for the Regression of Carbon Emissions per Capita and Total Carbon Emissions, 1990 to 2016 Model 1 Model 2 *** *** GDP per Capita .396 .543 (.067) (.133) Military Expenditures as % Government Expenditures (MGE) .024 –.244 (.124) (.237) GDP per Capita × MGE –.002 .026 (.017) (.029) * * Military Expenditures per Soldier (MEPS) .055 .122 (.024) (.051) Military Participation Rate (MPR) .014 .162 (.062) (.108) *** Total Population 1.400 (.164) R Overall .833 .881 Note: Model 1 is for carbon emissions per capita. Model 2 is for total carbon emissions. Both models control for urban population, non-dependent population, and services as % GDP. All variables are in logarithmic form. Robust standard errors clustered by nation are in parentheses. Both models include nation-specific fixed effects derived from the within estimator. Both models include unreported year- specific intercepts. N = 2,269 for 105 nations with 21.6 mean observations per nation. * ** *** p < .05; p < .01; p < .001 (two-tailed). Jorgenson et al. 25 Table A8. Elasticity Coefficients for the Regression of Total Carbon Emissions, 1990 to 2016 Model 1 Model 2 Model 3 Model 4 ** * *** ** GDP per Capita × MEPS × MPR .054 .027 .029 .025 (.019) (.010) (.008) (.009) *** *** *** *** GDP per Capita .909 .294 .303 .543 (.096) (.057) (.034) (.071) * ** * Military Expenditures per Soldier (MEPS) .044 .051 .009 .024 (.021) (.017) (.012) (.010) ** Military Participation Rate (MPR) –.102 .046 .017 .077 (.056) (.039) (.024) (.029) GDP per Capita × MEPS –.005 –.008 –.008 –.005 (.017) (.008) (.007) (.007) GDP per Capita × MPR –.036 –.020 –.016 –.035 (.046) (.034) (.017) (.018) * *** MEPS × MPR .047 .039 .026 .064 (.029) (.024) (.011) (.015) *** *** *** *** Total Population 1.000 .622 .340 .845 (.214) (.127) (.079) (.128) *** *** *** Total CO Lagged .558 .630 .331 (.078) (.028) (.063) R Overall .973 J-statistic 15.125 Note: Model 1 is for the static pooled common correlated effects estimator. Model 2 is for the two-way fixed-effects dynamic model with clustered robust standard errors. Model 3 is for the dynamic pooled common correlated effects estimator. Westerlund, Perova, Norkute standard errors reported for Models 1 and 3. Model 4 is for the two-stage instrumental-variable estimation with two lags used as instruments. For Model 1, N = 2,422 for 95 nations, with 26 mean observations per nation. For Model 2, N = 2,477 for 106 nations, with 23.4 mean observations per nation. For Model 3, N = 2,096 for 96 nations, with 22 mean observations per nation. For Model 4, N = 2,256 for 106 nations, with 21.3 mean observations per nation. Due to insufficient observations, the analysis drops 11 nations from Model 1 and 10 nations from Model 3. Standard errors are in parentheses. All variables are in logarithmic form. J-statistic for Model 4 is not statistically significant (H : overidentifying restrictions are valid). All models control for urban population, non-dependent population, and services as % GDP. GDP per capita, MEPS, and MPR are mean centered. * ** *** p < .05; p < .01; p < .001 (two-tailed). Acknowledgments oRCiD iDs Prior versions of this paper were presented at the Andrew K. Jorgenson https://orcid.org/0000-0002 Center for Environmental Politics’ Duck Family Col- -5902-3704 loquium Series at the University of Washington, the Brett Clark https://orcid.org/0000-0002-3929-1322 Schiller Institute for Integrated Science and Society’s Ryan P . Thombs https://orcid.org/0000-0002-8823-6143 Climate Change Research Seminar Series, the Envi- Jeffrey Kentor https://orcid.org/0000-0001-6735 ronmental Sociology Workshop at Boston College, the -4012 Department of Sociology at the University of British Columbia, and the conference on “Navigating Uncertain Jennifer E. Givens https://orcid.org/0000-0003-3103-9795 Futures: Social Engagement and Transformative Change Xiaorui Huang https://orcid.org/0000-0002-3823-0253 in Global Socio-Ecological Systems,” co-hosted by the Daniel Auerbach https://orcid.org/0000-0002-0984-4504 Environment and Society Research Committee of the Matthew C. Mahutga https://orcid.org/0000-0003 International Sociological Association and the Section -4696-529X on Environmental Sociology of the American Sociologi- cal Association. The authors thank the participants for their helpful feedback. The authors also thank Wesley notes Longhofer, Ali Kadivar, Katrina Paxton, the editors of ASR, and the anonymous reviewers for their helpful com- 1. For recent examples, see https://www.global ments and suggestions. change.gov/content/social-science-perspectives- 26 American Sociological Review 00(0) climate-change-workshop; https://www.ipcc.ch/ lines in reporting and accounting for armed forces report/sixth-assessment-report-working-group- across nations. ii/; https://www.ipcc.ch/report/sixth-assessment- 11. To assess if the positive coefficient for the three- report-working-group-3/. way interaction holds across different measures of 2. Using JSTOR and official journal webpages, as of national carbon emissions, we estimate three mod- February 2023, we found only 37 research articles els of total emissions, which we report in Appendix with “military” or “militarization” in their titles Table A4. The models control for urban population, ever published in the two leading generalist soci- non-dependent population, services as percent of ology journals, American Sociological Review (17 GDP, and total population. Model 1 is for the overall total published) and American Journal of Sociology unbalanced panel dataset, and Model 2 is for the per- (20 total published). fectly balanced panel dataset reduced to 53 nations. 3. See https://www.fedcenter.gov/programs/eo14057/; Model 3 is for the larger unbalanced panel dataset https://www.whitehouse.gov/briefing-room/ and also controls for trade as percent of GDP, democ- presidential-actions/2021/12/08/executive-order- ratization, and military expenditures as percent of on-catalyzing-clean-energy-industries-and-jobs- government expenditures. The estimated coefficient through-federal-sustainability/. for GDP per capita × MEPS × MPR is positive and 4. Tilly (1990) argues that the modern nation-state statistically significant in all three models, providing emerged from a coalescence of coercive/military additional support for our key arguments. power controlled by political units (e.g., despots 12. Appendix Table A5 provides the significance tests and empires) with economic power centered in for the pairwise differences for the marginal effects city-states, beginning around 1000 CE. This fusion reported in Figures 1 and 2. of economic and coercive power was driven by 13. The renewable energy data (as percent of total final political states’ need to obtain sufficient capital to energy consumption) come from the World Bank support the rising cost of military activity, due to (2022). These data are currently unavailable beyond advances in military technology, combined with the year 2015 and are limited in coverage for some city-states’ need for protection. Overall, the mili- nations in our study. Their inclusion reduces the tary played a foundational role in the formation of overall sample to 2,303 total observations for 106 today’s interstate system. nations for 1990 to 2015. 5. Other bodies of research focus on how military 14. The arms exports data come from the Stockholm spending directly affects economic growth. The International Peace Research Institute (SIPRI) results differ across studies, partly depending on the Arms Transfers Database March 2021 Version sample, the analyzed time frame, model estimation (http://www.sipri.org/databases/armstransfers). techniques, and choice of military measures (e.g., SIPRI reports these data using a common unit, the Benoit 1973; Bullock and Firebaugh 1990; Kentor trend-indicator value, which allows for the overall and Kick 2008; Weede 1983). Whether such spend- comparison of arms trade between countries and ing directly increases or decreases economic growth over time. Given limited availability, including could lead to changes in carbon emissions growth arms exports reduces the sample to 709 observa- via mediation. This is distinct from our moderation- tions for 28 nations. oriented hypotheses, which we test with moderation 15. Oil production is measured in million metric tons, analysis. which we obtain from BP’s Statistical Review of 6. They include carbon dioxide produced during con- World Energy Database 2020 Version (https://www sumption of solid, liquid, and gas fuels and gas flar- .bp.com/en/global/corporate/energy-economics/ ing. statistical-review-of-world-energy.html). Given lim- 7. The three variables are grand mean-centered after ited availability, including oil production reduces the log transformation (see the Model Estimation Tech- sample to 801 observations for 32 nations. niques section). 16. In additional models, available from the lead author 8. Less frequently, manufacturing as percent of GDP upon request, we instead include the interaction is included as a control instead of services as per- between GDP per capita and military expenditures cent of GDP. However, manufacturing as percent of as percent of GDP. The estimated effect of the two- GDP has more missing data for nation-years than way interaction is not statistically significant, fur- does services as percent of GDP, and thus the latter ther supporting our focus on military expenditures is generally more desirable as a control for cross- per solider and military participation rate as mod- national panel analyses. erators. 9. For the equations, α refers to the nation-specific 17. The double-demeaned estimator estimates the fixed effects, u refers to the year-specific fixed within effect by demeaning each variable and then effects, and ε refers to the error term. demeaning the product of the two. i,t 10. The direct associations between national carbon 18. Seemingly unrelated regression allows for a set of emissions and the military measures are likely equations to be correlated with each other, and in underestimated, given varying practices and guide- this case, works by combining the results for the Jorgenson et al. 27 fixed-effects estimator and the double-demeaned Avant, Deborah. 2005. The Market for Force: The Con- estimator into a single model. sequences of Privatizing Security. Cambridge, UK: 19. These test statistics are available from the lead Cambridge University Press. author upon request. Baran, Paul A., and Paul M. Sweezy. 1966. Monopoly 20. For total emissions, the model also includes total Capital. New York: Monthly Review Press. population as an independent variable. Beckley, Michael. 2010. “Economic Development and 21. With this approach, the regressors are assumed to be a Military Effectiveness.” The Journal of Strategic function of unobserved common factors and an idiosyn- Studies 33(1):43–79. cratic error term that is independent of the model error Belcher, Oliver, Patrick Bigger, Benjamin Neimark, and term, and thus the defactored lags of the regressors are Cara Kennelly. 2020. “Hidden Carbon Costs of the valid instruments. We estimate models with two lags of ‘Everywhere War’: Logistics, Geopolitical Ecol- the covariates as the defactored instruments. ogy, and the Carbon Boot-Print of the US Military.” 22. If a heterogeneous model is more appropriate than Transactions of the Institute of British Geographers the pooled model, then the test of the overidentify- 45(1):65–80. ing restrictions in the pooled model will result in Belcher, Oliver, Benjamin Neimark, and Patrick Bigger. a statistically significant J-statistic (Kripfganz and 2020. “The U.S. Military Is Not Sustainable.” Science Sarafidis 2021). 367(6481):989–90. 23. These data come from SIPRI (2022) and the World Benoit, Emile. 1973. Defense and Economic Growth in Bank (2022). We provide the global measures for Developing Countries. Lexington, MA: Lexington 1990 to 2016 to correspond with our cross-national Books. analyses for the same period. Bigger, Patrick, and Benjamin Neimark. 2017. “Wea- ponizing Nature: The Geopolitical Ecology of the US Navy’s Biofuel Program.” Political Geography References 60:13–22. Black, Jeremy. 2008. War and the World: Military Power Adams, Gordon. 1982. The Iron Triangle. Piscataway, and the Fate of Continents, 1450–2000. New Haven, NJ: Transaction Publishers. CT: Yale University Press. Adua, Lazarus, Brett Clark, and Richard York. 2021. Block, Fred. 1980. “Economic Instability and Military “The Ineffectiveness of Efficiency: The Paradoxi- Strength.” Politics and Society 10(1):35–58. cal Effects of State Policy on Energy Consumption Bohr, Jeremiah, and Riley E. Dunlap. 2018. “Key Top- in the United States.” Energy Research & Social ics in Environmental Sociology, 1990–2014: Results Science 71:101806 (https://doi.org/10.1016/j.erss from a Computational Text Analysis.” Environmental .2020.101806). Sociology 4(2):181–95. Adua, Lazarus, Richard York, and Beth-Anne Schuelke- Boswell, Terry. 1989. “Colonial Empires and the Capital- Leech. 2016. “The Human Dimensions of Climate ist World-Economy: A Time Series Analysis of Colo- Change: A Micro-Level Assessment of Views from nization, 1640–1960.” American Sociological Review the Ecological Modernization, Political Economy 54(2):180–96. and Human Ecology Perspectives.” Social Science Boswell, Terry, and William Dixon. 1990. “Dependency Research 56:26–43. and Rebellion: A Cross-National Analysis.” Ameri- Alario, Margarita V., Leda Nath, and Steve Carlton-Ford. can Sociological Review 55(4):540–59. 2016. “Climatic Disruptions, Natural Resources, and Bradford, John Hamilton, and Alexander M. Stoner. Conflict: The Challenges to Governance.” Journal of 2017. “The Treadmill of Destruction in Comparative Environmental Studies and Sciences 6:251–59. Perspective: A Panel Study of Military Spending and Alic, John, Daniel Sarewitz, Charles Weiss, and William Carbon Emissions, 1960–2014.” Journal of World- Bonvillian. 2010. “A New Strategy for Energy Inno- Systems Research 23(2):298–325. vation.” Nature 466:316–17. Brady, David, Jason Beckfield, and Martin Seeleib-Kai- Almeida, Paul, and Christopher Chase-Dunn. 2018. “Glo- ser. 2005. “Economic Globalization and the Welfare balization and Social Movements.” Annual Review of State in Affluent Democracies, 1975–2001.” Ameri- Sociology 44:189–211. can Sociological Review 70(6):921–48. Alvarez, Camila H. 2016. “Militarization and Water: A Brady, David, Jason Beckfield, and Wei Zhao. 2007. Cross-National Analysis of Militarism and Fresh- “The Consequences of Economic Globalization for water Withdrawals.” Environmental Sociology Affluent Democracies.” Annual Review of Sociology 2(3):298–305. 33:313–34. Alvarez, Camila H. 2021. “Military, Race, and Urban- Braswell, Taylor. 2022. “Extended Spaces of Environmen- ization: Lessons of Environmental Injustice from tal Injustice: Hydrocarbon Pipelines in the Age of Plan- Las Vegas, Nevada.” Sociological Perspectives etary Urbanization.” Social Forces 100(3):1025–52. 64(3):325–42. Brauer, Jurgen. 2009. War and Nature: The Environmen- Andreski, Stanislav. 1968. Military Organization and tal Consequences of War in a Globalized World. Lan- Society, 2nd ed. Berkeley: University of California ham, MD: Alta Mira Press. Press. 28 American Sociological Review 00(0) Bugden, Dylan. 2022. “Technology, Decoupling, and Clark, Brett, and Andrew K. Jorgenson. 2012. “The Tread- Ecological Crisis: Examining Ecological Modern- mill of Destruction and the Environmental Impacts of ization Theory through Patent Data.” Environmental Militaries.” Sociology Compass 6(7):557–69. Sociology 8(2):228–41. Clark, Brett, Andrew K. Jorgenson, and Jeffrey Kentor. Bullock, Brad, and Glenn Firebaugh. 1990. “Guns and 2010. “Militarization and Energy Consumption: A Butter? The Effect of Militarization on Economic and Test of Treadmill of Destruction Theory in Compara- Social Development in the Third World.” Journal of tive Perspective.” International Journal of Sociology Political and Military Sociology 18(2):231–66. 40(2):23–43. Bunker, Stephen G., and Paul S. Ciccantell. 2005. Glo- Clark, Brett, and Richard York. 2005. “Carbon Metabo- balization and the Race for Resources. Baltimore, lism: Global Capitalism, Climate Change, and the MD: Johns Hopkins University Press. Biospheric Rift.” Theory and Society 34(4):391–428. Burmaoglu, Serhat, and Ozcan Sarıtas. 2017. “Changing Clark, Robert, and Matthew Mahutga. 2013. “Explaining Characteristics of Warfare and the Future of Mili- the Trade-Growth Link: Assessing Diffusion-Based tary R&D.” Technological Forecasting and Social and Structure-Based Models of Exchange.” Social Change 116:151–61. Science Research 42(2):401–17. Burnett, Mackenzie, and Katherine J. Mach. 2021. CNA. 2007. “National Security and the Threat of Climate “A ‘Precariously Unprepared’ Pentagon? Climate Change.” Alexandria, VA: CNA Corporation (https:// Security Beliefs and Decision-Making in the U.S. www.cna.org/cna_files/pdf/national%20security%20 Military.” Global Environmental Change 70:102345 and%20the%20threat%20of%20climate%20change.pdf). (https://doi.org/10.1016/j.gloenvcha.2021.102345). Collins, Randall. 1981. “Does Modern Technology Burns, Thomas J., Byron L. Davis, and Edward L. Kick. Change the Rules of Geopolitics?” Journal of Politi- 1997. “Position in the World-System and National cal & Military Sociology 9(2):163–77. Emissions of Greenhouse Gases.” Journal of World- Commoner, Barry. 1971. The Closing Circle. New York: Systems Research 3(3):432–66. Alfred A. Knopf Publishers. Buttel, Frederick H. 2000. “World Society, the Nation- Condliffe, Jamie. 2017. “The Department of Defense State, and Environmental Protection: Comment on Wants to Double Down on Renewables.” MIT Tech- Frank, Hironaka, and Schofer.” American Sociologi- nology Review (https://www.technologyreview.com/ cal Review 65(1):117–21. s/603778/the-department-of-defense-wants-to-dou Cane, Mark A., Edward Miguel, Marshall Burke, Solo- ble-down-on-renewables/). mon M. Hsiang, David B. Lobell, Kyle C. Meng, Cooley, Alexander, Daniel Nexon, and Steven Ward. and Shanker Satyanath. 2014. “Temperature and Vio- 2019. “Revising Order or Challenging the Balance lence.” Nature Climate Change 4:234–35 (https://doi of Military Power? An Alternative Typology of Revi- .org/10.1038/nclimate2171). sionist and Status-Quo States.” Review of Interna- Caniglia, Elizabeth, Andrew Jorgenson, Stephanie Malin, tional Studies 45(4):689–708. Lori Peek, David Pellow, and Xiaorui Huang, eds. Correa, Jennifer G., and Joseph M. Simpson. 2022. 2021. Handbook of Environmental Sociology. Dor- “Building Walls, Destroying Borderlands: Reper- drecht, Netherlands: Springer. toires of Militarization on the United States-Mexico Carlton-Ford, Steve. 2010. “Major Armed Conflicts, Mil- Border.” Nature and Culture 17(1):1–25. itarization, and Life Chances: A Pooled Time-Series Custers, Peter. 2010. “Military Keynesianism Today: An Analysis.” Armed Forces & Society 36(5):864–89. Innovative Discourse.” Race & Class 51(4):79–94. Carlton-Ford, Steve, Katherine Durante, David Stearns, Cypher, James. 2015. “The Origins and Evolution of Mil- and Ciera Graham. 2019. “Guns and Butter: Child itary Keynesianism in the United States.” Journal of Mortality and the Mediators of Militarization.” Post Keynesian Economics 38(3):449–76. Armed Forces & Society 45(1):177–97. Cypher, James. 2022. “The Political Economy of Systemic Center for Systemic Peace and Societal-Systems U.S. Militarism.” Monthly Review 73(11):23–37. Research. 2018. “POLITY5 Dataset Version 2018” Davidson, Debra J. 2019. “Exnovating for a Renewable (http://www.systemicpeace.org). Energy Transition.” Nature Energy 4:254–56. Chase-Dunn, Christopher. 1998. Global Formation: Davidson, Debra J. 2022. “Climate Change Sociology: Structures of the World Economy. Lanham, MD: Past Contributions and Future Research Needs.” Rowman & Littlefield Publishers. PLOS Climate 1(7):e0000055 (https://doi.org/ Chase-Dunn, Christopher, Yukio Kawano, and Benja- 10.1371/journal.pclm.0000055). min Brewer. 2000. “Trade Globalization since 1795: Davidson, Debra J., and Jeffrey Andrews. 2013. “Not All Waves of Integration in the World-System.” Ameri- About Consumption.” Science 339(6125):1286–87. can Sociological Review 65(1):77–95. Deb, Nikhel. 2021. “Slow Violence and the Gas Peedit Clark, Brett, Daniel Auerbach, and Stefano B. Longo. in Neoliberal India.” Social Problems (https://doi.org/ 2018. “The Bottom Line: Capital’s Production of 10.1093/socpro/spab058). Social Inequalities and Environmental Degrada- Dietz, Thomas. 2015. “Prolegomenon to a Structural tion.” Journal of Environmental Studies and Sciences Human Ecology of Human Well-Being.” Sociology 8:562–69. of Development 1(1):123–48. Jorgenson et al. 29 Dietz, Thomas. 2017. “Drivers of Human Stress on the Foster, John Bellamy, and Robert McChesney. 2014. Environment in the Twenty-First Century.” Annual “Surveillance Capitalism: Monopoly-Finance Capi- Review of Environment and Resources 42:189–213. tal, the Military-Industrial Complex, and the Digital Dietz, Thomas, Kenneth Frank, Cameron Whitley, Jen- Age.” Monthly Review 66(3):1–31. nifer Kelly, and Rachel Kelly. 2015. “Political Frank, David John. 1997. “Science, Nature, and the Glo- Influences on Greenhouse Gas Emissions from US balization of the Environment, 1870–1990.” Social States.” Proceedings of the National Academy of Sci- Forces 76(2):409–435. ences 112(27):8254–59. Frank, David John, Ann Hironaka, and Evan Schofer. Dietz, Thomas, and Eugene A. Rosa. 1997. “Effects 2000. “The Nation-State and the Natural Environ- of Population and Affluence on CO Emissions.” ment over the Twentieth Century.” American Socio- Proceedings of the National Academy of Sciences logical Review 65(1):96–116. 94(1):175–79. Frey, R. Scott. 2013. “Agent Orange and America at War Dietz, Thomas, Rachael Shwom, and Cameron Whitley. in Vietnam and Southeast Asia.” Human Ecology 2020. “Climate Change and Society.” Annual Review Review 20(1):1–10. of Sociology 46:135–58. Gareau, Brian, and Cristina Lucier. 2018. “Neoliberal Ditzen, Jan. 2018. “Estimating Dynamic Common- Restructuring of the World Polity: The Weakening Correlated Effects in Stata.” The Stata Journal of the Montreal Protocol and Basel Convention in 18(3):585–617. Historical Perspective.” Environmental Sociology Downey, Liam. 2015. Inequality, Democracy, and the 4(3):325–42. Environment. New York: NYU Press. Giddens, Anthony. 1987. The Nation-State and Violence. Driscoll, Daniel. 2021. “Drivers of Carbon Price Volume Two of a Contemporary Critique of Histori- Adoption in Wealthy Democracies: International cal Materialism. Berkeley: University of California or Domestic Forces?” Socius 7 (https://doi.org/ Press. 10.1177/2378023121992252). Giddens, Anthony. 2011. The Politics of Climate Change. Dunlap, Alexander, and Andrea Brock, eds. 2022. Enforc- Cambridge, UK: Polity Press. ing Ecocide: Power, Policing, and Planetary Militari- Giesselmann, Marco, and Alexander Schmidt-Catran. zation. Dordrecht, Netherlands: Springer. 2022. “Interactions in Fixed Effects Regression Mod- Dunlap, Riley E., and Robert J. Brulle, eds. 2015. Cli- els.” Sociological Methods & Research 51(3):1100– mate Change and Society: Sociological Perspectives. 27. New York: Oxford University Press. Givens, Jennifer E. 2014. “Global Climate Change Nego- Durant, Robert. 2007. The Greening of the US Military: tiations, the Treadmill of Destruction, and World Soci- Environmental Policy, National Security, and Orga- ety.” International Journal of Sociology 44(2):7–36. nizational Change. Washington, DC: Georgetown Givens, Jennifer E. 2018. “Ecologically Unequal University Press. Exchange and the Carbon Intensity of Well-Being, Elliott, James R., and Scott Frickel. 2015. “Urbaniza- 1990–2011.” Environmental Sociology 4(3):311–24. tion as Socioenvironmental Succession: The Case of Givens, Jennifer E., Xiaorui Huang, and Andrew K. Jor- Hazardous Industrial Site Accumulation.” American genson. 2019. “Ecologically Unequal Exchange: A Journal of Sociology 120(6):1736–77. Theory of Global Environmental Injustice.” Sociol- Falkner, Robert. 2021. Environmentalism and Global ogy Compass 13(5):e12693. International Society. Cambridge, UK: Cambridge Givens, Jennifer E., and Andrew K. Jorgenson. 2013. University Press. “Individual Environmental Concern in the World Pol- Fisher, Dana R. 2022. “AnthroShift in a Warming ity: A Multilevel Analysis.” Social Science Research World.” Climate Action 1:9 (https://doi.org/10.1007/ 42(2):418–31. s44168-022-00011-8). Gould, Kenneth A. 2007. “The Ecological Costs of Mili- Fisher, Dana R., and William R. Freudenburg. 2004. tarization.” Peace Review 19(4):331–34. “Post Industrialization and Environmental Quality: Gould, Kenneth A., David N. Pellow, and Allan Schnai- An Empirical Analysis of the Environmental State.” berg. 2008. The Treadmill of Production: Injustice Social Forces 83(1):157–88. and Unsustainability in the Global Economy. Boul- Fisher, Dana R., and Andrew K. Jorgenson. 2019. “End- der, CO: Paradigm Publishers. ing the Stalemate: Toward a Theory of Anthro-Shift.” Grant, Don, Albert Bergesen, and Andrew Jones. 2002. Sociological Theory 37(4):342–62. “Organizational Size and Pollution: The Case of the Foster, John Bellamy. 1999. “Marx’s Theory of Metabolic U.S. Chemical Industry.” American Sociological Rift.” American Journal of Sociology 105(2):366–405. Review 67(3):389–407. Foster, John Bellamy, and Brett Clark. 2020. The Rob- Grant, Don, Andrew Jorgenson, and Wesley Longhofer. bery of Nature. New York: Monthly Review Press. 2020. Super Polluters: Tackling the World’s Largest Foster, John Bellamy, and Hannah Holleman. 2012. Sites of Climate-Disrupting Emissions. New York: “Weber and the Environment: Classical Foundations Columbia University Press. for a Postexemptionalist Sociology.” American Jour- Grant, Don, and Ion Bogdan Vasi. 2017. “Civil Society in nal of Sociology 117(6):1625–73. an Age of Environmental Accountability: How Local 30 American Sociological Review 00(0) Environmental Nongovernmental Organizations Hsiao, Cheng, M. Hashem Pesaran, and A. Kamil Tah- Reduce U.S. Power Plants’ Carbon Dioxide Emis- miscioglu. 2002. “Maximum Likelihood Estimation sions.” Sociological Forum 32(1):94–115. of Fixed Effects Dynamic Panel Data Models Cov- Greiner, Patrick T. 2022. “Colonial Contexts and the Fea- ering Short Time Periods.” Journal of Econometrics sibility of Mitigation through Transition: A Study of 109(1):107–50. the Impact of Historical Processes on the Emissions Huang, Xiaorui. 2018. “Ecologically Unequal Exchange, Dynamics of Nation-States.” Global Environmental Recessions, and Climate Change: A Longitudinal Change 77:102609 (https://doi.org/10.1016/j.gloen Study.” Social Science Research 73:1–12. vcha.2022.102609). Huang, Xiaorui, and Andrew Jorgenson. 2018. “The Griffin, Larry, Joel Devine, and Michael Wallace. 1982. Asymmetrical Effects of Economic Development on “Monopoly Capital, Organized Labor, and Military Consumption-Based and Production-Based Carbon Expenditures in the United States, 1949–1976.” Amer- Dioxide Emissions, 1990–2014.” Socius 4 (http://doi ican Journal of Sociology 88(Supplement):S113– .org/10.1177/2378023118773626). S153. Huber, Joseph. 2010. “Upstreaming Environmental Gunderson, Ryan, Diana Stuart, and Brian Petersen. Action.” Pp. 334–55 in The Ecological Modernisa- 2018. “Ideological Obstacles to Effective Climate tion Reader, edited by A. Mol, G. Spaargaren, and D. Policy: The Greening of Markets, Technology, and Sonnenfeld. London, UK: Routledge. Growth.” Capital & Class 42(1):133–60. Hyde, Allen, and Todd Vachon. 2019. “Running With Haberl, Helmut, Dominik Wiedenhofer, Doris Virág, or Against the Treadmill? Labor Unions, Institu- Gerald Kalt, Barbara Plank, Paul Brockway, Tomer tional Contexts, and Greenhouse Gas Emissions in a Fishman, et al. 2020. “A Systematic Review of the Comparative Perspective.” Environmental Sociology Evidence on Decoupling of GDP, Resource Use and 5(3):269–82. GHG Emissions, Part II: Synthesizing the Insights.” Inglehart, Ronald, and Wayne Baker. 2000. “Modern- Environmental Research Letters 15:065003 (http:// ization, Cultural Change, and the Persistence of doi.org/10.1088/1748-9326/ab842a). Traditional Values.” American Sociological Review Hironaka, Ann. 2014. Greening the Globe: World Society 65(1):19–51. and Environmental Change. New York: Cambridge IPCC (Intergovernmental Panel on Climate Change). University Press. 2007. Climate Change 2007: Synthesis Report. Hirst, Paul. 2001. War and Power in the Twenty-First Geneva, Switzerland: IPCC. Century: The State, Military Power and the Interna- IPCC (Intergovernmental Panel on Climate Change). tional System. Cambridge, UK: Polity Press. 2013. Climate Change 2013: The Physical Science Hooks, Gregory. 1990. “The Rise of the Pentagon and US Basis. Cambridge, UK: Cambridge University Press. State Building: The Defense Program as Industrial Pol- IPCC (Intergovernmental Panel on Climate Change). icy.” American Journal of Sociology 96(2):358–404. 2022. Climate Change 2022: Mitigation of Climate Hooks, Gregory. 1994. “Regional Processes in the Hege- Change. Contribution of Working Group III to the monic Nation: Political, Economic, and Military Sixth Assessment Report of the Intergovernmental Influences on the Use of Geographic Space.” Ameri- Panel on Climate Change. Cambridge, UK: Cam- can Sociological Review 59(5):746–72. bridge University Press. Hooks, Gregory, and Leonard Bloomquist. 1992. “The Isiksal, Aliya Zhakanova. 2021. “Testing the Effect of Legacy of World War II for Regional Growth and Sustainable Energy and Military Expenses on Envi- Decline: The Cumulative Effects of Wartime Invest- ronmental Degradation: Evidence from the States ments on US Manufacturing, 1947–1972.” Social with the Highest Military Expenses.” Environmental Forces 71(2):303–337. Science and Pollution Research 28:20487–89. Hooks, Gregory, Michael R. Lengefeld, and Chad Smith. Jaccard, James, and Robert Turrisi. 2003. Interaction 2021. “Recasting the Treadmills of Production and Effects in Multiple Regression, 2nd ed. Thousand Destruction: New Theoretical Directions.” Sociology Oaks, CA: Sage. of Development 7(1):52–76. Jayaram, Dhanasree, and Marie Claire Brisbois. 2021. Hooks, Gregory, and Chad Smith. 2004. “The Tread- “Aiding or Undermining? The Military as an Emer- mill of Destruction: National Sacrifice Areas and gent Actor in Global Climate Governance.” Earth Sys- Native Americans.” American Sociological Review tem Governance 9:100107 (https://doi.org/10.1016/ 69(4):558–75. j.esg.2021.100107). Hooks, Gregory, and Chad Smith. 2005. “Treadmills of Johnson, Chalmers. 2004. The Sorrows of Empire. New Production and Destruction: Threats to the Environ- York: Metropolitan Books. ment Posed by Militarism.” Organization & Environ- Jorgenson, Andrew K. 2014. “Economic Development ment 18(1):19–37. and the Carbon Intensity of Human Well-Being.” Hsiang, Solomon M., Marchall Burke, and Edward Nature Climate Change 4:186–89. Miguel. 2013. “Quantifying the Influence of Climate Jorgenson, Andrew K., and Brett Clark. 2009. “The on Human Conflict.” Science 341(6151):1235367 Economy, Military, and Ecologically Unequal Rela- (https://doi.org/10.1126/science.1235367). tionships in Comparative Perspective: A Panel Study Jorgenson et al. 31 of the Ecological Footprints of Nations, 1975–2000.” Kick, Edward L., Byron L. Davis, David M. Kiefer, and Social Problems 56(4):621–46. Thomas J. Burns. 1998. “A Cross-National Analy- Jorgenson, Andrew K., and Brett Clark. 2012. “Are the sis of Militarization and Well-Being Relationships Economy and the Environment Decoupling? A Com- in Developing Countries.” Social Science Research parative International Study, 1960–2005.” American 27(4):351–70. Journal of Sociology 118(1):1–44. Klare, Michael. 2019. All Hell Breaking Loose: The Pen- Jorgenson, Andrew K., and Brett Clark. 2016. “The Tem- tagon’s Perspective on Climate Change. New York: poral Stability and Developmental Differences in the Metropolitan Books. Environmental Impacts of Militarism: The Tread- Kleykamp, Meredith. 2007. “Military Service as a mill of Destruction and Consumption-Based Carbon Labor Market Outcome.” Race, Gender & Class Emissions.” Sustainability Science 11:505–514. 14(3/4):65–76. Jorgenson, Andrew K., Brett Clark, and Jeffrey Kentor. Klinenberg, Eric, Malcolm Araos, and Liz Koslow. 2020. 2010. “Militarization and the Environment: A Panel “Sociology and the Climate Crisis.” Annual Review Study of Carbon Dioxide Emissions and the Ecologi- of Sociology 46:649–69. cal Footprints of Nations, 1970–2000.” Global Envi- Knight, Kyle, and Juliet Schor. 2014. “Economic Growth ronmental Politics 10(1):7–29. and Climate Change: A Cross-National Analysis of Jorgenson, Andrew K., Shirley Fiske, Klaus Hubacek, Jia Territorial and Consumption-Based Carbon Emis- Li, Tom McGovern, Torben Rick, Juliet B. Schor, Wil- sions in High-Income Countries.” Sustainability liam Solecki, Richard York, and Ariela Zycherman. 6(6):3722–31. 2019. “Social Science Perspectives on Drivers of and Krahmann, Elke. 2010. States, Citizens and the Privati- Responses to Global Climate Change.” WIREs Climate sation of Security. New York: Cambridge University Change 10(1):e554 (https://doi.org/10.1002/wcc.554). Press. Kelly, Orla. 2020. “The Silver Bullet? Assessing the Kramer, Ronald. 2020. Carbon Criminals, Climate Role of Education for Sustainability.” Social Forces Crimes. New Brunswick, NJ: Rutgers University 99(1):178–204. Press. Kelly, Orla, Ryan P. Thombs, and Andrew K. Jorgenson. Kraska, Peter. 2007. “Militarization and Policing – Its 2021. “The Unsustainable State: Greenhouse Gas Relevance to 21st Century Police.” Policing: A Jour- Emissions, Inequality, and Human Well-Being in the nal of Policy and Practice 1(4):501–513. United States, 1913 to 2017.” Socius 7 (http://doi.org/ Kripfganz, Sebastian, and Vasilis Sarafidis. 2021. 10.1177/23780231211020536). “Instrumental Variable Estimation of Large Panel Kennedy, Emily Huddart, and Jennifer E. Givens. 2019. Data Models with Common Factors.” The Stata Jour- “Eco-habitus or Eco-powerlessness? Examining nal 21(3):659–86. Environmental Concern across Social Class.” Socio- Lawrence, Michael, Holly Stemberger, Aaron Zolderdo, logical Perspectives 62(5):646–67. Daniel Struthers, and Steven Cooke. 2015. “The Kentor, Jeffrey. 2000. Capital and Coercion: The Eco- Effects of Modern War and Military Activities on nomic and Military Processes That Have Shaped the Biodiversity and the Environment.” Environmental World Economy 1800–1990. New York: Garland Pub- Reviews 23(4):443–60. lishing. Lengefeld, Michael R., Gregory Hooks, and Chad Smith. Kentor, Jeffrey, Rob Clark, and Andrew K. Jorgenson. 2021. “War and the Environment.” In Handbook of 2023. “The Hidden Cost of Global Economic Inte- Environmental Sociology, edited by B. Caniglia, gration: How Foreign Investment Drives Military A. Jorgenson, S. Malin, L. Peek, D. Pellow, and X. Expenditures.” World Development 161:106105 Huang. Dordrecht, Netherlands: Springer. (https://doi.org/10.1016/j.worlddev.2022.106105). Lengefeld, Michael R., and Chad Smith. 2013. “Nuclear Kentor, Jeffrey, and Andrew K. Jorgenson. 2017. “Mili- Shadows: Weighing the Environmental Effects tary Expenditures and Health: A Cross-National of Militarism, Capitalism, and Modernization in Study, 1975–2000.” International Journal of Sociol- a Global Context, 2001–2007.” Human Ecology ogy and Social Policy 37(13/14):755–72. Review 20(1):11–25. Kentor, Jeffrey, Andrew K. Jorgenson, and Edward Kick. Levine, Paul, Somnath Sen, and Ron Smith. 1994. “A 2012. “The ‘New’ Military and Income Inequality: A Model of the International Arms Market.” Defence Cross National Analysis.” Social Science Research and Peace Economics 5:1–18. 41(3):514–26. Levy, Barry, and Victor Sidel. 2007. War and Public Kentor, Jeffrey, and Edward Kick. 2008. “Bringing the Health, 2nd ed. New York: Oxford University Press. Military Back In: Military Expenditures and Eco- Light, Sarah. 2014. “The Military-Environmental Com- nomic Growth 1990–2003.” Journal of World Sys- plex.” Boston College Law Review 55(3):879–946. tems Research 14(2):142–72. Lim, Alwyn, and Kiyoteru Tsutsui. 2012. “Globaliza- Keyßer, Lorenz, and Manfred Lenzen. 2021. “1.5°C tion and Commitment in Corporate Social Respon- Degrowth Scenarios Suggest the Need for New Miti- sibility: Cross-National Analyses of Institutional and gation Pathways.” Nature Communications 12:2676 Political-Economy Effects.” American Sociological (https://doi.org/10.1038/s41467-021-22884-9). Review 77(1):69–98. 32 American Sociological Review 00(0) Lockie, Stewart. 2022. “Mainstreaming Climate Change McChesney, Robert. 2013. Digital Disconnect. New Sociology.” Environmental Sociology 8(1):1–6. York: New Press. Longhofer, Wesley, and Andrew K. Jorgenson. 2017. McGee, Julius Alexander, and Patrick Trent Greiner. “Decoupling Reconsidered: Does World Society 2018. “Can Reducing Income Inequality Decouple Integration Influence the Relationship between the Economic Growth from CO Emissions?” Socius 4 Environment and Economic Development?” Social (http://doi.org/10.1177/2378023118772716). Science Research 65:17–29. McNeill, William H. 1982. The Pursuit of Power. Chi- Longhofer, Wesley, and Evan Schofer. 2010. “National cago: The University of Chicago Press. and Global Origins of Environmental Association.” McQuaig, Linda, and Neil Brooks. 2012. Billionaire’s American Sociological Review 75(4):505–533. Ball. Boston: Beacon Press. Longo, Stefano, Ellinor Isgren, Brett Clark, Andrew K. Mejia, Steven. 2021. “The Climate Crisis and Export Jorgenson, Anne Jerneck, Lennart Olsson, Orla Kelly, Intensity: A Comparative International Study of David Harnesk, and Richard York. 2021. “Sociology Greenhouse Gas Emissions in the Global South, 1990– for Sustainability Science.” Discover Sustainability 2014.” International Journal of Sociology 51(1):1–22. 2:47 (http://doi.org/10.1007/s43621-021-00056-5). Mezey, Nancy J. 2020. “Presidential Address: Start Looney, Robert. 1990. “Military Expenditures, Military Spreading the News: Illuminating the Effects of Cli- Participation Rates, and Human Capital Development mate Change as a Social Problem.” Social Problems in the Arab World.” Current Research on Peace and 67(4):605–615. Violence 13(1):39–53. Mills, C. Wright. 1956. The Power Elite. New York: Lynch, Michael, Michael Long, Paul Stretesky, and Kim- Oxford University Press. berly Barrett. 2017. Green Criminology: Crime, Jus- Mitchell, Jon. 2020. Poisoning the Pacific. Lanham, MD: tice, and the Environment. Berkeley: University of Rowman & Littlefield. California Press. Mitchell, Ronald B., and Richard York. 2020. “Reduc- Mach, Katharine, Caroline Kraan, W. Neil Adger, ing the Web’s Carbon Footprint: Does Improved Halvard Buhaug, Marshall Burke, James Fearon, Electrical Efficiency Reduce Webserver Electricity Christopher Field, et al. 2019. “Climate as a Risk Fac- Use?” Energy Research & Social Science 65:101474 tor for Armed Conflict.” Nature 571:193–97. (https://doi.org/10.1016/j.erss.2020.101474). Machlis, Gary, and Thor Hanson. 2008. “Warfare Ecol- Mol, Arthur P. J. 2003. Globalization and Environmental ogy.” BioScience 58(8):729–36. Reform: The Ecological Modernization of the Global Magdoff, Harry. 1978. Imperialism. New York: Monthly Economy. Cambridge, MA: MIT Press. Review Press. Mumford, Lewis. 1963. Technics and Civilization. New Mahutga, Matthew C. 2006. “The Persistence of York: Harcourt, Brace & World, Incorporated. Structural Inequality? A Network Analysis of Newman, Nathan. 2002. Net Loss: Internet Prophets, Pri- International Trade, 1965–2000.” Social Forces vate Profits, and the Costs to Community. University 84(4):1863–89. Park: Pennsylvania State University Press. Mahutga, Matthew C., and David A. Smith. 2011. “Glo- Nickell, Stephen. 1981. “Biases in Dynamic Models with balization, the Structure of the World Economy and Fixed Effects.” Econometrica 49(6):1417–26. Economic Development.” Social Science Research Norgaard, Kari Marie. 2018. “The Sociological Imagina- 40(1):257–72. tion in a Time of Climate Change.” Global and Plan- Malin, Stephanie, Tara Opsal, Tara O’Connor Shelley, etary Change 163:171–76. and Peter Hall. 2019. “The Right to Resist or a Case Norkutė, Milda, Vasilis Sarafidis, Takashi Yamagata, and of Injustice? Meta-Power in the Oil and Gas Fields.” Guowei Cui. 2021. “Instrumental Variable Estimation Social Forces 97(4):1811–38. of Dynamic Linear Panel Data Models with Defac- Mann, Michael. 2012. The Sources of Social Power, Vol. tored Regressors and a Multifactor Error Structure.” 3, Global Empires and Revolution, 1890–1945. New Journal of Econometrics 220(2):416–46. York: Cambridge University Press. Nowak, Peter. 2011. Sex, Bombs, and Burgers: How War, Mann, Michael. 2014. The Sources of Social Power, Vol. Pornography, and Fast Food Have Shaped Modern 4, Globalizations, 1945–2011. New York: Cambridge Technology. Guilford, CT: Lyons Press. University Press. O’Rourke, Ronald. 2021. Navy Large Unmanned Surface Marquart-Pyatt, Sandra. 2012. “Contextual Influences and Undersea Vehicles. Washington, DC: Congres- on Environmental Concerns Cross-Nationally: A sional Research Service. Multilevel Investigation.” Social Science Research Pathak, Swapna. 2020. “Ecological Footprints of War: An 41(5):1085–99. Exploratory Assessment of the Long-Term Impact of Marzec, Robert. 2016. Militarizing the Environment: Violent Conflicts on National Biocapacity from Climate Change and the Security State. Minneapolis: 1962–2009.” Journal of Environmental Studies and University of Minnesota Press. Sciences 10:380–93. Mazur, Allan. 2013. Energy and Electricity in Industrial Pellow, David N. 2007. Resisting Global Toxics: Trans- Nations: The Sociology and Technology of Energy. national Movements for Environmental Justice. Cam- Oxford, UK: Earthscan. bridge, MA: MIT Press. Jorgenson et al. 33 Pellow, David N., and Hollie Nyseth Brehm. 2013. “An Samaras, Constantine, William J. Nuttall, and Morgan Environmental Sociology for the Twenty-First Cen- Bazilian. 2019. “Energy and the Military: Conver- tury.” Annual Review of Sociology 39:229–50. gence of Security, Economic, and Environmen- Pesaran, M. Hashem. 2006. “Estimation and Inference in tal Decision-Making.” Energy Strategy Reviews Large Heterogeneous Panels with a Multifactor Error 26:100409 (https://doi.org/10.1016/j.esr.2019.100 Structure.” Econometrica 74(4):967–1012. 409). Pesaran, M. Hashem, and Ron Smith. 1995. “Estimating Sanders, Barry. 2009. The Green Zone: The Environmen- Long-Run Relationships from Dynamic Heteroge- tal Costs of Militarism. Oakland, CA: AK Press. neous Panels.” Journal of Econometrics 68(1):79– Sanderson, Matthew, and Vivian Hughes. 2019. “Race to 113. the Bottom (of the Well): Groundwater in an Agri- Pickup, Mark. 2015. Introduction to Time Series Analy- cultural Production Treadmill.” Social Problems sis. Thousand Oaks, CA: Sage. 66(3):392–410. Podobnik, Bruce. 2006. Global Energy Shifts: Foster- Scanlan, Stephen, and Craig Jenkins. 2001. “Military ing Sustainability in a Turbulent Age. Philadelphia: Power and Food Security: A Cross-National Analysis Temple University Press. of Less-Developed Countries, 1970–1990.” Interna- Raleigh, Clionadh, and Henrik Urdal. 2007. “Climate tional Studies Quarterly 45(1):159–87. Change, Environmental Degradation and Armed Schnaiberg, Allan. 1980. The Environment. New York: Conflict.” Political Geography 26(6):674–94. Oxford University Press. Renner, Michael. 1991. “Assessing the Military’s War on Schofer, Evan. 2003. “The Global Institutionalization of the Environment.” Pp. 132–52 in State of the World, Geological Science, 1800 to 1990.” American Socio- edited by L. Starke. New York: W.W. Norton & Com- logical Review 68(5):730–59. pany. Schofer, Evan, and Ann Hironaka. 2005. “The Effects Rice, James. 2007. “Ecological Unequal Exchange: of World Society on Environmental Protection Out- International Trade and Uneven Utilization of Envi- comes.” Social Forces 84(1):25–47. ronmental Space in the World System.” Social Forces Scott, Lauren, and Erik Johnson. 2017. “From Fringe to 85(3):1369–92. Core? The Integration of Environmental Sociology.” Rice, James. 2015. “Downwind of the Atomic State: Environmental Sociology 3(1):17–29. US Continental Atmospheric Testing, Radioactive Shandra, John M., Bruce London, Owen P. Whooley, Fallout, and Organizational Deviance, 1951–1962.” and John B. Williamson. 2004. “International Non- Social Science History 39(4):647–76. governmental Organizations and Carbon Dioxide Rieger, Annika. 2021. “Does ICT Result in Dematerial- Emissions in the Developing World: A Quantita- ization? The Case of Europe, 2005–2017.” Environ- tive, Cross-National Analysis.” Sociological Inquiry mental Sociology 7(1):64–75. 74(4):520–45. Roberts, J. Timmons, Peter E. Grimes, and Jodie L. Sharkey, Amanda J., and Patricia Bromley. 2015. “Can Manale. 2003. “Social Roots of Global Environ- Ratings Have Indirect Effects? Evidence from the mental Change: A World-Systems Analysis of Car- Organizational Response to Peers’ Environmental bon Dioxide Emissions.” Journal of World-Systems Ratings.” American Sociological Review 80(1):63– Research 9(2):277–315. 91. Rosa, Eugene A., and Thomas Dietz. 2012. “Human Shaw, Martin. 1988. Dialectics of War. London, UK: Drivers of National Greenhouse-Gas Emissions.” Pluto Press. Nature Climate Change 2:581–86. Shaw, Martin. 2002. “Risk-Transfer Militarism: Small Rosa, Eugene A., Richard York, and Thomas Dietz. 2004. Massacres and the Historic Legitimacy of War.” “Tracking the Anthropogenic Drivers of Ecological International Relations 16(3):343–60. Impacts.” Ambio 33(8):509–512. Shaw, Martin. 2005. The New Western Way of War: Risk- Royal Society and U.S. National Academy of Sciences. Transfer War and Its Crisis in Iraq. Cambridge, UK: 2020. Climate Change: Evidence and Causes. Wash- Polity Press. ington, DC: National Academics Press. Shorette, Kristen. 2012. “Outcomes of Global Environ- Rudel, Thomas K. 2009. “How Do People Transform mentalism: Longitudinal and Cross-National Trends Landscapes? A Sociological Perspective on Suburban in Chemical Fertilizer and Pesticide Use.” Social Sprawl and Tropical Deforestation.” American Jour- Forces 91(1):299–325. nal of Sociology 115(1):129–54. Shwom, Rachael. 2011. “A Middle Range Theorization Rudel, Thomas K., J. Timmons Roberts, and JoAnn Car- of Energy Politics: The Struggle for Energy Efficient min. 2011. “Political Economy of the Environment.” Appliances.” Environmental Politics 20(5):705–26. Annual Review of Sociology 37:221–38. Siebold, Guy. 2001. “Core Issues and Theory in Military Running, Katrina. 2013. “World Citizenship and Con- Sociology.” Journal of Political and Military Sociol- cern for Global Warming: Building the Case for a ogy 29(1):140–59. Strong International Civil Society.” Social Forces Sills, Peter. 2014. Toxic War: The Story of Agent Orange. 92(1):377–99. Nashville, TN: Vanderbilt University Press. 34 American Sociological Review 00(0) Simpson, Joseph M., Riley E. Dunlap, and Andrew S. tial Vulnerability to Climate Change: A Social Sci- Fullerton. 2019. “The Treadmill of Information: ence Review.” WIREs Climate Change 10(2):e565 Development of the Information Society and Car- (https://doi.org/10.1002/wcc.565). bon Dioxide Emissions.” Sociology of Development Thombs, Ryan P. 2017. “The Paradoxical Relationship 5(4):381–409. between Renewable Energy and Economic Growth: Singer, Peter. 2008. Corporate Warriors: The Rise of A Cross-National Panel Study, 1990–2013.” Journal the Privatized Military Industry. Ithaca, NY: Cornell of World-Systems Research 23(2):540–64. University Press. Thombs, Ryan P. 2018a. “The Transnational Tilt of the SIPRI (Stockholm International Peace Research Insti- Treadmill and the Role of Trade Openness on Car- tute). 2022. “Military Expenditure Database” (https:// bon Emissions: A Comparative International Study, www.sipri.org/databases/milex). 1965–2010.” Sociological Forum 33(2):422–42. Smart, Barry. 2016. “Military-Industrial Complexities, Thombs, Ryan P. 2018b. “Has the Relationship between University Research and Neoliberal Economy.” Jour- Non-Fossil Fuel Energy Sources and CO Emissions nal of Sociology 52(3):455–81. Changed Over Time? A Cross-National Study, 2000– Smith, Chad, and Michael R. Lengefeld. 2020. “The 2013.” Climatic Change 148(4):481–90. Environmental Consequences of Asymmetric War: Thombs, Ryan P. 2021. “In-and-Beyond State Power: A Panel Study of Militarism and Carbon Emissions, How Political Equality Moderates the Economic 2000–2010.” Armed Forces & Society 46(2):214–37. Growth-CO Emissions Relationship, 1990–2014.” Smith, David A. 2017. “Globalizing Social Problems: An Sociological Quarterly 62(3):528–47. Agenda for the Twenty-First Century.” Social Prob- Thombs, Ryan P. 2022. “A Guide to Analyzing Large lems 64(1):1–13. N, Large T Panel Data.” Socius 8 (http://doi.org/ Smith, Kurt. 2020. “Environmental Protection, the Mili- 10.1177/23780231221117645). tary, and Preserving the Balance: Why It Matters, in Thombs, Ryan P., and Xiaorui Huang. 2019. “Uneven War and Peace.” Seattle Journal of Technology, Envi- Decoupling: The Economic Growth–CO Emissions ronmental & Innovation Law 11(1): Article 5. Relationship in the Global North, 1870 to 2014.” Soener, Matthew. 2019. “Profiting in a Warming World: Sociology of Development 5(4):410–27. Investigating the Link between Greenhouse Gas Thombs, Ryan P., Xiaorui Huang, and Jared B. Fitzgerald. Emissions and Capitalist Profitability in OECD 2022. “What Goes Up Might Not Come Down: Mod- States.” Sociological Forum 34(4):974–98. eling Directional Asymmetry with Large-N, Large-T Soeters, Joseph. 2018. Sociology and Military Studies: Data.” Sociological Methodology 52(1):1–29. Classical and Current Foundations. New York: Rout- Tilly, Charles. 1990. Coercion, Capital, and European ledge Press. States, AD 990–1992. Oxford, UK: Blackwell. Sohag, Kazi, Shaiara Husain, Shawkat Hammoudeh, and Turse, Nick. 2008. The Complex: How the Military Normah Omar. 2021. “Innovation, Militarization, and Invades our Everyday Lives. New York: Henry Holt Renewable Energy and Green Growth.” Environ- and Company. mental Science and Pollution Research 28:36004–17 Turse, Nick. 2015. Tomorrow’s Battlefield: US Proxy (https://doi.org/10.1007/s11356-021-13326-6). Wars and Secret Ops in Africa. Chicago: Haymarket Spaargaren, Gert, Arthur P. J. Mol, and Frederick H. But- Books. tel. 2006. Governing Environmental Flows: Global U.S. Army. 1999. Staging, Onward Movement, and Inte- Challenges to Social Theory. Cambridge, MA: MIT gration Field Manuel. Washington, DC: Headquar- Press. ters, Department of Army. Staples, Steven. 2000. “The Relationship between Global- USDOD (United States Department of Defense). 2010. ization and Militarism.” Social Justice 27(4):18–22. Quadrennial Defense Review Report. Arlington, VA: Stuart, Diana. 2021. What Is Environmental Sociology? United States Department of Defense. Cambridge, UK: Polity Press. USDOD (United States Department of Defense). 2020. Swed, Ori, and Thomas Crosbie, eds. 2019. The Sociol- Department of Defense Annual Energy Management ogy of Privatized Security. London, UK: Palgrave and Resilience Report (AEMRR) Fiscal Year 2019. Macmillan. Washington, DC: Office of the Assistant Secretary of Thayer, George. 1969. The War Business: The Interna- Defense for Sustainment. tional Trade in Armaments. New York: Simon and Vandenbergh, Michael, and Jonathan Gilligan. 2017. Schuster. Beyond Politics: The Private Governance Response Thee, Marek. 1990. “Science-Based Military Technol- to Climate Change. Cambridge, UK: Cambridge Uni- ogy as a Driving Force behind the Arms Race.” Pp. versity Press. 105–20 in Arms Races: Technological and Political Vasi, Ion Bogdan, Edward Walker, John Johnson, and Dynamics, edited by N. Petter Gleditsch and O. Njøl- Hui Fen Tan. 2015. “No Fracking Way! Documen- stad. London, UK: Sage. tary Film, Discursive Opportunity, and Local Oppo- Thomas, Kimberley, R. Dean Hardy, Heather Lazrus, sition against Hydraulic Fracturing in the United Michael Mendez, Ben Orlove, Isabel Rivera-Collazo, States, 2010 to 2013.” American Sociological Review Timmons Roberts, et al. 2019. “Explaining Differen- 80(5):934–59. Jorgenson et al. 35 Vesia, Danielle J., Matthew C. Mahutga, and Bonnie been published in outlets such as the American Journal of Khánh Hà Buì. 2021. “Flattening the Curve? The Sociology, Conservation Biology, Global Environmental Structure of the Natural Resource Exchange Network Change, Environmental Research Letters, Theory and and CO Emissions.” Social Networks (http://doi.org/ Society, and Social Problems. His most recent book, with 10.1016/j.socnet.2021.07.004). John Bellamy Foster, is The Robbery of Nature. Vine, David. 2015. Base Nation. New York: Metropolitan Books. Ryan P. Thombs is a PhD candidate in the Sociology Weede, Erich. 1983. “Military Participation Ratios, Department at Boston College. His research examines Human Capital Formation, and Economic Growth: the drivers of global environmental change, the drivers A Cross-National Analysis.” Journal of Political and of health disparities and population health outcomes, Military Sociology 11(1):11–19. and quantitative methodology, particularly related to Wilcox, Fred. 2011. Scorched Earth: Legacies of Chemi- panel data and time-series modeling. His work appears cal Warfare in Vietnam. New York: Seven Stories in journals such as Sociological Methodology, Journal Press. of Health and Social Behavior, Sociological Forum, Wilcox, William. 2007. The Modern Military and the The Sociological Quarterly, Sociology of Development, Environment: The Laws of Peace and War. Lanham, Global Environmental Change, Environmental Research MD: Government Institutes. Letters, Science of the Total Environment, Climatic Wills, Jocelyn. 2017. Tug of War: Surveillance Capital- Change, and Energy Research & Social Science. ism, Military Contracting, and the Rise of the Secu- rity State. Montreal, CA: McGill-Queen’s University Jeffrey Kentor is a Professor of Sociology at Wayne Press. State University and founding co-editor of Sociology of World Bank. 2022. “World Development Indicators” Development. Kentor’s research focuses on long-term, (https://databank.worldbank.org/source/world-devel macro-level social change, from a global political- opment-indicators). economy perspective. His work considers the fundamen- York, Richard. 2012. “Asymmetric Effects of Economic tal economic and coercive forces that shape relationships Growth and Decline on CO Emissions.” Nature Cli- among countries, and how these dynamics impact eco- mate Change 2:762–64. nomic development, inequality, health, and the environ- York, Richard, and Julius A. McGee. 2016. “Understand- ment. Kentor’s research is published in the American ing the Jevons Paradox.” Environmental Sociology Journal of Sociology, American Sociological Review, 2(1):77–87. Social Forces, Social Problems, World Development, York, Richard, and Julius A. McGee. 2017. “Does Energy Research & Social Science, Global Environmen- Renewable Energy Development Decouple Eco- tal Politics, and elsewhere. nomic Growth from CO Emissions?” Socius 3 (http://doi.org/10.1177/2378023116689098). York, Richard, Eugene A. Rosa, and Thomas Dietz. Jennifer E. Givens is an Associate Professor of Soci- 2003. “Footprints on the Earth: The Environmental ology at Utah State University. She is interested in Consequences of Modernity.” American Sociological relationships between the environment, development, Review 68(2):279–300. well-being, and various inequalities from the local to Zierler, David. 2011. The Invention of Ecocide. Athens: the global, especially in the context of climate change. University of Georgia Press. Jennifer has published in journals such as Environmental Sociology, Sociology of Development, Social Science Research, Society & Natural Resources, Social Indica- Andrew K. Jorgenson is a Professor of Sociology at tors Research, and Environmental Research Letters, University of British Columbia (UBC) and a Research and she has received funding for her interdisciplinary Fellow at Vilnius University. Prior to UBC, he was research from the U.S. National Science Foundation. faculty at Boston College. Much of his research focuses on the societal dimensions of the climate crisis. His work appears in such venues as American Journal Xiaorui Huang is an Assistant Professor of Sociology of Sociology, Nature Climate Change, Social Forces, at Drexel University. His research examines the climate Environmental Research Letters, Social Problems, and change and human well-being implications of economic Science of the Total Environment. He is a recipient of the development, international trade, renewable energy, Fred Buttel Distinguished Contribution Award from the and income inequality. He has also conducted work American Sociological Association’s Section on Envi- on quantitative methods and agricultural communities’ ronmental Sociology, and co-author of the recent book, engagement with natural resource and climate change Super Polluters. issues. His published research appears in such venues as Social Science Research, Climatic Change, Sociological Brett Clark is a Professor of Sociology at the University Forum, Sociological Methodology, Energy Research & of Utah. His research focuses primarily on the human Social Science, Ecological Economics, Socius, and Sci- dimensions of environmental change. His scholarship has ence of the Total Environment. 36 American Sociological Review 00(0) Hassan El Tinay is a PhD student in the Department of Matthew C. Mahutga is a political economist at the Sociology at Boston College. His research focuses on the University of California-Riverside. His research utilizes nexus between political economy and political ecology, theories of global and comparative political economy to including topics such as the human drivers of climate study inequality and stratification, economic develop- change and ecologically unequal exchange. ment, environmental degradation, economic organiza- tion, labor, labor markets, and political attitudes and preferences. Beyond that, he enjoys talking to people Daniel Auerbach is an Assistant Professor of Sociology about matters that are important to them, mentoring hard- at the University of Wyoming. His research examines working and ambitious graduate students, and promoting the political-economic and militaristic drivers of envi- the upward mobility of underrepresented students. ronmental change. Daniel has published in journals such as International Critical Thought, Environmental Sociol- ogy, Climatic Change, and Organization & Environment. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png American Sociological Review SAGE

Guns versus Climate: How Militarization Amplifies the Effect of Economic Growth on Carbon Emissions

Loading next page...
 
/lp/sage/guns-versus-climate-how-militarization-amplifies-the-effect-of-20JtMZP1CA

References (270)

Publisher
SAGE
Copyright
© The Author(s) 2023
ISSN
0003-1224
eISSN
1939-8271
DOI
10.1177/00031224231169790
Publisher site
See Article on Publisher Site

Abstract

Building on cornerstone traditions in historical sociology, as well as work in environmental sociology and political-economic sociology, we theorize and investigate with moderation analysis how and why national militaries shape the effect of economic growth on carbon pollution. Militaries exert a substantial influence on the production and consumption patterns of economies, and the environmental demands required to support their evolving infrastructure. As far-reaching and distinct characteristics of contemporary militarization, we suggest that both the size and capital intensiveness of the world’s militaries enlarge the effect of economic growth on nations’ carbon emissions. In particular, we posit that each increases the extent to which the other amplifies the effect of economic growth on carbon pollution. To test our arguments, we estimate longitudinal models of emissions for 106 nations from 1990 to 2016. Across various model specifications, robustness checks, a range of sensitivity analyses, and counterfactual analysis, the findings consistently support our propositions. Beyond advancing the environment and economic growth literature in sociology, this study makes significant contributions to sociological research on climate change and the climate crisis, and it underscores the importance of considering the military in scholarship across the discipline. Keywords climate change, environmental sociology, political-economic sociology, development sociology, militarization Rich bodies of sociological theory and analy- University of British Columbia ses have significantly advanced scientific Vilnius University understanding of the human dimensions of University of Utah Boston College climate change (e.g., Davidson 2022; Dietz, Wayne State University Shwom, and Whitley 2020; Dunlap and Utah State University Brulle 2015; Klinenberg, Araos, and Koslow Drexel University 2020; Norgaard 2018). Distinct research tra- University of Wyoming ditions focus on how structural characteristics University of California-Riverside of societies, usually nation-states, generate Corresponding Author: different levels and rates of carbon dioxide Andrew K. Jorgenson, Department of Sociology, and other greenhouse gas emissions, and also University of British Columbia, AnSo 2111, how relationships between nations shape their 6303 NW Marine Drive, Vancouver, BC V6T 1Z1, unequal contributions to emissions and Canada. Email: andrew.jorgenson@ubc.ca planetary warming (e.g., Givens, Huang, and 2 American Sociological Review 00(0) Jorgenson 2019; Kelly 2020; Pellow and sociology, studying the role of militaries is a Brehm 2013; Rice 2007; Rudel, Roberts, and vital direction to pursue, given the emergence Carmin 2011). While broadening and deepen- of the war economy and defense industry ing the presence of environmental sociology (see also Giddens 1987; Hooks 1990). As in the discipline as a whole (Lockie 2022; Andreski (1968:1) argued decades ago: Mezey 2020; Scott and Johnson 2017; Smith 2017), this scholarship also contributes to The problem of the influence of military interdisciplinary climate science efforts and organization on society has, on the whole, policy considerations (Haberl et al. 2020; failed to attract the attention of social sci- IPCC 2022; Jorgenson et al. 2019; Longo entists. To be sure, much has been written et al. 2021; Rosa and Dietz 2012; Thomas about war, its alleged evil or beneficial et al. 2019). effects, its causes and the possibilities of its Drawing on various disciplinary subfields, abolition. But the only writers who appre- the most central question within this area of ciated the importance of military factors sociological inquiry concerns the relationship in shaping societies were Max Weber and between nations’ carbon emissions and their Gaetano Mosca. This persistent neglect is economic growth (Bohr and Dunlap 2018; due, I think, to the insidious utopianism Fisher and Jorgenson 2019; Stuart 2021). A which pervades sociological thinking. suite of critical perspectives, including tread- mill of production and metabolic rift theories, Building on foundational traditions in his- argue that economic growth is antithetical to torical sociology (e.g., Chase-Dunn 1998; environmental protection, given the increased Mann 2012; Mills 1956; Tilly 1990) as well energy and resource demands, as well as the as work in environmental sociology (e.g., subsequent environmental harms, associated Hooks and Smith 2004; Smith and Lengefeld with such growth (e.g., Clark and York 2005; 2020) and political-economic sociology (e.g., Foster 1999; Gould, Pellow, and Schnaiberg Boswell 1989; Scanlan and Jenkins 2001), 2008). In contrast, more optimistic perspec- we theorize and investigate with moderation tives, such as ecological modernization theory analysis how and why militarization ampli- and environmental state approaches, suggest fies the effect of economic growth on nations’ energy efficiency improvement and environ- carbon emissions. As complex social institu- mental protection measures often accompany tions, the world’s militaries exert a sizable growth (e.g., Fisher and Freudenburg 2004; influence on the production and consumption Hironaka 2014; Mol 2003). Prior research patterns of nations and their economies, and tends to support the more critical perspec- the environmental demands required to sup- tives. Analyses consistently show positive port their evolving infrastructure. associations between carbon emissions and The rise of the world’s militaries, espe- economic growth, with the magnitude of the cially after the Second World War, led to relationship varying for nations in differ- modern forms of militarization shaping ent structural and temporal contexts (e.g., carbon-intensive growth in national and Dietz 2017; Dietz and Rosa 1997; Jorgenson international economies through contracts for 2014; Jorgenson and Clark 2012; Rosa, York, research, development, production, and sup- and Dietz 2004; Thombs 2018a; Thombs and port. The continual preparation for potential Huang 2019; York 2012). conflicts and the desire to maintain national To advance sociological research on cli- security increases the scale of resource- mate change, it is necessary to gain greater consuming economic activities within the understanding of how other prominent soci- defense industry. With elaborate communi- etal characteristics shape the relationship cation technologies, larger ships, and faster between emissions and economic growth. planes and helicopters, militaries move peo- Although largely overlooked in generalist ple and equipment throughout the world more Jorgenson et al. 3 quickly. Extensive production systems and differences. Across various model specifica- supply chains within the defense industry tions for different samples, robustness checks, and other areas of the private sector operate a range of sensitivity analyses, counterfactual to meet the various needs of militaries’ infra- analysis, and for the two carbon measures, the structures, including their bases and instal- findings confirm our propositions. lations scattered around the globe, and the needs of their soldiers and support personnel. LiTERATuRE REviEw National militaries are increasingly capi- Economic Growth and Environmental tal intensive, focusing on technologies in Change weaponry, transportation, and communica- tions. In line with prior research, we use military expenditures per soldier to measure Sociological research on the human dimen- these capital-intensive features of militariza- sions of global environmental change, includ- tion (e.g., Jorgenson and Clark 2009; Kentor ing energy consumption and greenhouse gas and Jorgenson 2017; Kentor, Jorgenson, and emissions, largely focuses on the effects of Kick 2012; Looney 1990). Likewise, militar- economic growth (Caniglia et al. 2021; Dietz ies with relatively larger forces require expan- 2015; Dietz et al. 2020). On the one hand, sive built infrastructures and huge amounts optimistic perspectives argue that as societies of material goods, such as food and clothing. experience economic growth, the magnitude Consistent with other sociological inquir- of environmental harms is likely to decrease. ies, we use military participation rate, which The reductions in environmental harms are quantifies a nation’s military personnel as due to the emergence of an environmentally percent of total labor force, to capture the focused state (Dietz et al. 2015; Falkner 2021; relative size of militaries (e.g., Carlton-Ford Fisher and Freudenburg 2004; Frank 1997; 2010; Carlton-Ford et al. 2019; Kick et al. Frank, Hironaka, and Schofer 2000; Spaar- 1998; Kleykamp 2007). We suggest that both garen, Mol, and Buttel 2006) and a growing expenditures per soldier and participation culture of post-materialism, coupled with a rate measure far-reaching characteristics of strengthening commitment to sustainability militarization that enlarge the effect of eco- within civil society (Givens and Jorgenson nomic growth on nations’ carbon emissions. 2013; Inglehart and Baker 2000; Kennedy In particular, we argue that each increases the and Givens 2019; Longhofer and Schofer extent to which the other amplifies the effect 2010; Marquart-Pyatt 2012; Running 2013; of economic growth on carbon pollution. Vasi et al. 2015). Other theorized mechanisms To test our arguments, we estimate longi- include emerging technologies driving the tudinal models of emissions for 106 nations ecological modernization of production and from 1990 to 2016, with a particular focus on distribution systems (Bugden 2022; Huber the three-way interaction between economic 2010; Mol 2003; Rieger 2021), as well as growth, measured as GDP per capita, military the overall greening of organizational culture expenditures per soldier, and military partici- and practices in the private sector through the pation rate. The three-way interaction allows adoption of an ecological rationality and the us to quantify the effect of economic growth diffusion of corporate social responsibility on emissions at different levels of military (Sharkey and Bromley 2015; Vandenbergh expenditures per soldier and military partici- and Gilligan 2017; cf. Grant, Bergesen, and pation rate simultaneously. We treat carbon Jones 2002; Lim and Tsutsui 2012). dioxide emissions per capita as our primary On the other hand, critical perspectives dependent variable, as it reflects international argue that economic growth leads to increased inequities in contributions to global emissions environmental effects, including higher lev- and climate change. We also estimate mod- els of carbon emissions. Economic growth els of total emissions, which capture scale is predicated on the continual expansion of 4 American Sociological Review 00(0) markets, supported by treadmills of produc- also Adua, York, and Schuelke-Leech 2016; tion with extensive horizontal and vertical Burns, Davis, and Kick 1997; Greiner 2022; linkages as well as transportation networks Huang 2018; Huang and Jorgenson 2018; and logistics systems moving enormous vol- Hyde and Vachon 2019; Kelly, Thombs, and umes of raw materials and finished commodi- Jorgenson 2021; Mejia 2021; Rosa et al. ties throughout the world (Braswell 2022; 2004; Soener 2019; York 2012). This corpus Bunker and Ciccantell 2005; Clark, Auerbach, of research supports the general arguments of and Longo 2018; Deb 2021; Gould et al. 2008; the more critical sociological approaches, and Pellow 2007; York, Rosa, and Dietz 2003). it has gained increased recognition among the If left unchecked, these energy-intensive and climate change mitigation community (e.g., waste-generating processes disrupt socio- Haberl et al. 2020; IPCC 2022; Keyßer and ecological systems, often exceeding natural Lenzen 2021). limits while contributing to a global “carbon Other studies focus on how certain factors rift” (Clark and York 2005; see also Davidson moderate the positive relationship between and Andrews 2013; Foster 1999; Foster and carbon emissions and economic growth. For Clark 2020; Foster and Holleman 2012). example, income inequality intensifies their Although nation-states adopt environmen- association in affluent nations (McGee and tal regulations, they simultaneously prioritize Greiner 2018), whereas political inequality economic growth through the protection of does the opposite: the positive relationship private property, bailing out different sectors between emissions and growth is stronger and industries when deemed necessary, main- for nations with higher levels of political taining energy security, and promoting trade equality (Thombs 2021). The overall role of agreements, all of which place increased pres- renewable energy technology in shaping the sure on the environment (Almeida and Chase- emissions and economic growth association Dunn 2018; Buttel 2000; Elliott and Frickel is inconclusive (e.g., Davidson 2019; Thombs 2015; Gareau and Lucier 2018; Rudel 2009). 2017; York and McGee 2017), and nations The environmental benefits of technology are more embedded in global environmental civil often reduced if not entirely outpaced, given the society experience a modest decrease in the contradictory position of the state, the increas- positive relationship between emissions and ing energy and material demands of societies growth through time (Longhofer and Jor- (partly due to efficiency-driven cost reductions genson 2017; see also Fisher 2022; Grant that encourage greater production and con- and Vasi 2017; Schofer and Hironaka 2005; sumption), and the overall growth and diversifi- Shandra et al. 2004; Shorette 2012). cation of markets (Adua, Clark, and York 2021; Driscoll 2021; Grant, Jorgenson, and Longhofer Militarization and Environmental 2020; Gunderson, Stuart, and Petersen 2018; Change Malin et al. 2019; Sanderson and Hughes 2019; Shwom 2011; York and McGee 2016). To advance sociological understanding of the A substantial body of sociological research causes of climate change, we theorize and consistently finds positive associations test how a powerful yet overlooked dimen- between nations’ carbon emissions and eco- sion of human social organization influ- nomic growth. Longitudinal studies indicate ences the effect of economic growth on that the positive relationship increases in mag- the environment: militarization. War itself nitude through time for less affluent nations, destroys the environment through scorched while remaining large and relatively stable earth practices, the use of biological and for more affluent nations (Jorgenson 2014; chemical weapons, and the killing of flora Jorgenson and Clark 2012; Knight and Schor and fauna (Brauer 2009; Frey 2013; Mitchell 2014; Thombs 2018a; Thombs and Huang 2020; Sills 2014; Wilcox 2011; Zierler 2011). 2019; Vesia, Mahutga, and Buì 2021; see From the mid-1940s to the early 1960s, the Jorgenson et al. 5 atmospheric testing of atomic and nuclear and petroleum-based items as well as other weapons produced radioactive fallout that material resources ranging from steel to cot- spread great distances by wind, water, and ton (Belcher, Neimark, and Bigger 2020; living organisms, leading to increased cancer Lawrence et al. 2015; USDOD 2020). These rates among downwinders (Commoner 1971; capital-intensive and scale characteristics Rice 2015). of militarization all contribute, directly or The environmental consequences of war indirectly, to greenhouse gas emissions and in the modern era continue to evolve, as various forms of environmental degradation emerging technologies in weapons, transpor- (Belcher, Bigger, et al. 2020; Clark, Jorgen- tation, and communications systems shape the son, and Kentor 2010; Gould 2007; Jorgenson, scale and precision of destruction (Lengefeld, Clark, and Kentor 2010; Roberts, Grimes, and Hooks, and Smith 2021; Levy and Sidel 2007; Manale 2003; Smith and Lengefeld 2020). Machlis and Hanson 2008). For the more Driven by risk and cost reduction as well capital-intensive and technologically advanced as energy security concerns, and often publicly militaries, this manifests in forms of “risk- framed as climate mitigation efforts, the mili- transfer militarism” (Shaw 2002, 2005), which taries of many nations increasingly focus on shields their homeland’s citizens, minimizes enhanced fossil fuel efficiency and the growing casualties for their soldiers, and decreases use of renewable forms of energy (Bigger and loss of machinery, while inflicting damage Neimark 2017; Condliffe 2017; Light 2014; on human populations, the built environment, Samaras, Nuttall, and Bazilian 2019; USDOD nonhuman species, and the overall natural 2020). However, the pursuit of carbon effi- environment of distant locations (Hooks, ciencies and renewable energy is challenging. Lengefeld, and Smith 2021; Lengefeld and Militaries traditionally prioritize bigger and Smith 2013; Smith and Lengefeld 2020). faster weapons and transportation systems to As noted by sociologists advancing the gain strategic and competitive advantages over treadmill of destruction perspective, the envi- geopolitical rivalries. Modern fighter planes, ronmental effects of militarization are not such as the F-15 and F-16, burn 1,500 to 1,700 limited to war and weapons testing (Hooks gallons of fuel per hour, military helicopters and Smith 2004, 2005; see also Alvarez 2016; consume approximately five gallons for each Bradford and Stoner 2017; Clark and Jor- mile traveled, and non-nuclear aircraft carriers genson 2012; Lawrence et al. 2015). In the utilize close to 6,000 gallons of fuel per hour name of national security, and motivated by while in operation (Jorgenson and Clark 2016; geopolitics and risk-transfer militarism, mili- Levy and Sidel 2007; Sanders 2009). Similar to taries continually invest in and pursue new what occurs in private- and other public-sector technologies in weapons, transportation, and contexts (see Grant et al. 2020; Mazur 2013; communications systems (Alic et al. 2010; Mitchell and York 2020; Simpson, Dunlap, and Burmaoglu and Sarıtas 2017; Mann 2014). Fullerton 2019; Thombs 2018b), the energy The United States alone has over 900 domes- required for militaries’ information technology tic bases and over 800 international bases systems could also involve contradictions and in 130 countries, as well as smaller military conflicts between increased carbon efficiency, installations known as lily pads throughout the transition to sustainable energy sources, and the world (Johnson 2004; Sanders 2009; Turse overall growth as they become more capital- 2015; Vine 2015). The scale of militaries’ intensive and technologically focused (Alic evolving infrastructure, including their trans- et al. 2010; Samaras et al. 2019; Sohag et al. portation systems to move people, supplies, 2021). and weaponry by land, air, and water through- Some nations’ militaries have made efforts out the globe, and their constant research and to become more energy efficient and ecologi- development activities, involve the consump- cally sustainable. However, military opera- tion of substantial amounts of fossil fuels tions, training exercises, and related land 6 American Sociological Review 00(0) holdings are often exempt from environmen- Beckfield, and Seeleib-Kaiser 2005; Brady, tal regulations domestically and abroad (e.g., Beckfield, and Zhao 2007; Mahutga 2006; Kramer 2020; Light 2014; Lynch et al. 2017; Thombs 2018a). National militaries propel Smith 2020; Wilcox 2007). A national secu- these socio-environmental processes through rity justification for such exemptions was attempts to sustain relative international sta- articulated by the commander of a military bility (Cooley, Nexon, and Ward 2019; Hirst base in response to a community’s concern 2001). An absence of large-scale conflicts about pollution and land degradation: “we are minimizes disruptions to global production in the business of protecting the nation, not and trade networks (Chase-Dunn, Kawano, the environment” (Renner 1991:152). Schol- and Brewer 2000; Kentor, Clark, and Jorgen- ars have noted the potential for militaries as son 2023), further contributing to fossil fuel actors in climate governance (Jayaram and consumption and economic growth (Clark Brisbois 2021), and a growing number of the and Mahutga 2013; Givens 2018; Mahutga world’s militaries consider climate change a and Smith 2011; Vesia et al. 2021). “threat multiplier” to national security and According to Mills (1956:198), the impor- international stability (Burnett and Mach tance and influence of the military increased 2021; Machlis and Hanson 2008; Marzec through its “ascendancy” into the power elite 2016; USDOD 2010; see also CNA 2007; from the Second World War to the present. IPCC 2007; Klare 2019). In spite of this, The military “became enlarged and decisive nations with larger and more powerful mili- to the shape of the entire economic struc- taries are slow to ratify international climate ture,” and as a result, “the economic and the agreements (Givens 2014). For the United military have become structurally and deeply States, President Biden’s Executive Order interrelated, as the economy has become a 14057, signed in late 2021, directs the U.S. seemingly permanent war economy” (Mills government to reach 100 percent carbon-free 1956:215; see also Downey 2015). Others electricity by 2030, net-zero emissions by highlight the broader institutional intercon- 2050, and eliminate carbon pollution from nections between the military, the economy, federal buildings and vehicles, but exempts and the state as the core of the military- anything related to the U.S. military and industrial complex (e.g., Adams 1982; Hooks national security. 1990; Siebold 2001; Staples 2000). Rich sociological analyses indicate that the needs of the world’s militaries provide oppor- How Militarization Amplifies tunities for a variety of old and emerging the Effect of Economic Growth private-sector industries (e.g., Custers 2010; on Carbon Emissions Hooks 1994; Hooks and Bloomquist 1992). Throughout history, societies with larger National militaries facilitate scientific inquiry and more technologically advanced militar- and technological innovation, and they shape ies have utilized their coercive power in production in the private sector while simul- geopolitical contexts to secure and maintain taneously acting as downstream consumers, access to energy and other natural resources both domestically and internationally, given (Tilly 1990; see also Beckley 2010; Black the global market for armaments and mili- 2008; Boswell 1989; Boswell and Dixon tary equipment (Smart 2016; Soeters 2018; 1990; Chase-Dunn 1998; Jorgenson and see also Levine, Sen, and Smith 1994; Mills Clark 2009; Kentor 2000; Magdoff 1978; 1956; Schofer 2003; Thayer 1969; Turse McNeill 1982; Podobnik 2006). In the mod- 2008). Governments, especially in wealthier ern era, access to fossil fuels, often from nations, provide research funding to develop distant places, facilitates carbon-polluting and enhance military weapons systems. development for nations as they compete These systems include cutting-edge com- in regional and global economies (Brady, munication technologies and infrastructure Jorgenson et al. 7 for coordinating routine operations, strate- also benefits from the application of tech- gic maneuvers, data collection, cybersecurity, nologies, often initially designed for military and surveillance (Collins 1981; Foster and purposes, to commercial products for global McChesney 2014; Shaw 1988; Wills 2017). markets (Hooks 1990; McChesney 2013; Research and development linked to the Turse 2008). For instance, military spend- capital intensiveness and size of militaries ing spearheaded research and development for increase the overall resource demands of this personal computers and networking technolo- institution (Jorgenson et al. 2010; Kentor gies, giving rise to the internet and e-commerce et al. 2012; Kentor and Kick 2008; Schnai- (McQuaig and Brooks 2012; Newman 2002; berg 1980). Efforts to maintain a strategic Nowak 2011). advantage generate a path dependency, con- Militaries also provide a release valve for stantly elevating the standard of military pre- the economy, absorbing excess capacity tied to paredness (Thee 1990; U.S. Army 1999). For occurrences of carbon-polluting overproduc- example, risk-transfer militarism involves the tion in the private sector, which helps reduce development by private-sector military con- macroeconomic disruptions and stabilize over- tractors of high-tech air and undersea vehi- all economic growth (Cypher 2015; Griffin, cles, such as drones and “robot subs,” that can Devine, and Wallace 1982). Law enforce- launch missiles at designated targets (Cypher ment agencies and private security entities 2022; O’Rourke 2021). While initially used throughout the world are major clients for the by the most dominant militaries, as part of defense industry, purchasing armored vehicles, an ever-evolving arms race, such high-tech weapons, communications systems, and other equipment is increasingly in demand for mili- specialized equipment initially developed taries throughout the world. for nations’ military purposes (Avant 2005; Overall, the interrelated activities embed- Dunlap and Brock 2022; Krahmann 2010; ded within the military-industrial complex Kraska 2007; Singer 2008; Swed and Crosbie include contracts for research, development, 2019). Through the demand for services, fuel, manufacturing, and servicing of weapons and and other resources, the presence of mili- their delivery systems, transportation vehi- tary bases and installations affects surround- cles, information technology, cybersecurity, ing communities and regions, influencing their communications equipment, and other infra- carbon-polluting economic activities and structural needs (Baran and Sweezy 1966; related environmental effects (Alvarez 2021; Block 1980; Foster and McChesney 2014). Correa and Simpson 2022; Durant 2007; Each of the nodes and links in these produc- Hooks 1994; Vine 2015; Wilcox 2007). tion systems, supply chains, and ancillary In summary, we argue that the com- services involves the burning of fossil fuels plex and evolving arrangements among the and the consumption of other resources, all world’s militaries and the private sector of which are amplified by the size and capital shape the relationship between national car- intensiveness of nations’ militaries. In other bon emissions and economic growth. The words, the effects of economic growth on car- effect of economic growth on emissions is bon emissions are shaped by both the capital likely greater for nations with larger and more intensiveness and size of nations’ militaries, capital-intensive militaries. As measures that and each likely increases the extent to which capture these far-reaching and distinct char- the other enlarges the effect of growth on acteristics of contemporary militarization, carbon pollution. we posit that both military expenditures per Militaries minimize risk for industry, as soldier (i.e., capital intensiveness) and mili- they provide an assured market. They help tary participation rate (i.e., size) enlarge the “reduce towards zero the gap in time between effect of economic growth on nations’ carbon profitable original production and profitable emissions. In particular, we argue that mili- replacement” (Mumford 1963:93). Industry tary participation rate increases the extent to 8 American Sociological Review 00(0) which expenditures per solider amplifies the disproportionately responsible on a per person effect of growth on carbon pollution, and basis for the amount of carbon emitted into likewise, expenditures per solider increases the atmosphere from human activities (e.g., the extent to which participation rate enlarges IPCC 2013; Royal Society and U.S. National the effect of economic growth on emissions. Academy of Sciences 2020). Consistent with We test our arguments with moderation other sociological research (e.g., Jorgenson analysis and multiple longitudinal modeling and Clark 2012; Longhofer and Jorgenson techniques, across unbalanced and balanced 2017; Thombs 2018a; Vesia et al. 2021), we panel datasets of nations, for two measures of also estimate models of total carbon dioxide carbon dioxide emissions, and with counter- emissions (measured in kilotons), which we factual analysis. report in the Appendix. Total emissions are analogous with the overall scale of emissions and are centrally relevant for climate mitiga- DATA AnD METHoDs tion concerns (IPCC 2013; Royal Society and The Dataset U.S. National Academy of Sciences 2020). We maximize the use of available data. The Primary Independent Variables overall panel dataset consists of 2,563 annual observations for 106 nations (24.2 mean, 9 The primary independent variables for this minimum, and 27 maximum annual observa- study include gross domestic product (GDP) tions per nation) for 1990 to 2016. Due to per capita, military expenditures per soldier missing data for the different measures, the (MEPS), military participation rate (MPR), samples vary across the estimated models, their two-way interactions (GDP per capita depending on which independent variables × MEPS, GDP per capita × MPR, MEPS are included. The year 1990 is the earliest, × MPR), and most importantly, their three- and 2016 is the most recent year, in which way interaction: GDP per capita × MEPS × some of the primary independent variables MPR (Jaccard and Turrisi 2003). For ease of are currently available. Appendix Table A1 interpretation, we calculate and use the grand lists the number of observations for each mean-centered versions for these three vari- nation in the overall dataset. We also esti- ables in the reported models that include their mate and report models where we restrict interactions. the dataset to nations with no missing data, GDP per capita is measured in constant which consists of perfectly balanced panels 2010 U.S. dollars. Military expenditures per of 27 annual observations for 53 nations. All soldier is calculated by dividing total military analyzed data are publicly available, and the expenditures by total armed forces person- overall panel dataset is available from the nel. Military participation rate is measured as lead author upon request. armed forces personnel as a percent of total labor force. Military expenditures per soldier quantifies the capital intensiveness of nations’ Dependent Variables militaries, and military participation rate The primary dependent variable is carbon measures the relative size of nations’ militar- dioxide emissions per capita, which we ies (see Carlton-Ford et al. 2019; Jorgenson obtained from the World Bank’s online World and Clark 2009; Jorgenson at al. 2010; Kentor Development Indicators Database (World et al. 2012; Kentor and Kick 2008; Kick et al. Bank 2022). These data, measured in metric 1998; Lengefeld and Smith 2013; Smith and tons per person, include emissions from the Lengefeld 2020). For the overall dataset, they burning of fossil fuels and the manufacture of are weakly correlated at –.11 in their original cement. Per capita emissions is commonly metrics and .01 in logarithmic form. used as a measure of international inequality Total military expenditures are measured in in emissions as it quantifies how nations are constant 2018 U.S. dollars and obtained from Jorgenson et al. 9 Stockholm International Peace Research Insti- To further enhance the validity of the tute’s online Military Expenditure Database hypotheses testing, we estimate models that (SIPRI 2022). These data include expenditures also control for military expenditures as a on personnel, operations and maintenance, pro- percent of general government expenditures. curement, military research and development, This additional variable, which we obtained military infrastructure spending (including from the World Bank (2022), is moderately military bases), and military aid (in the mili- correlated with military participation rate tary expenditure of the donor country). They (.513) and weakly correlated with military exclude civil defense and current expenditures expenditures per soldier (.074). on previous military activities, demobilization, conversion, and weapon destruction. Armed Model Estimation Techniques forces personnel consist of active-duty military personnel, including paramilitary forces if the We estimate and report two-way fixed-effects training, organization, equipment, and control regression models with robust standard errors suggest they may be used to support or replace clustered by nation, correcting for unobserved regular military forces. Measures of GDP per heterogeneity that is time-invariant within capita, total armed forces personnel, and mili- nations as well as cross-sectionally invariant tary participation rate come from the World within years. We estimate the models with Bank (2022). the xtreg command in Stata software, which uses the within estimator to account for the country-level fixed effects, and the temporal Additional Independent Variables fixed effects are derived from the inclusion The reported models include a variety of of year-specific dummy variables. Consistent additional independent variables common in with the majority of sociological research on sociological research on the human drivers the anthropogenic drivers of national emis- of carbon emissions (Dietz at al. 2020; Rosa sions (see Jorgenson et al. 2019; Rosa and and Dietz 2012). Each model includes urban Dietz 2012), we transform all nonbinary vari- population as a percent of the total popu- ables into logarithmic form. This means the lation, non-dependent population (percent models estimate elasticity coefficients where of the total population age 15 to 64), and the coefficient for the independent variable services as a percent of GDP, all obtained is the estimated net percentage change in from the World Bank (2022). Prior studies the dependent variable associated with a 1 generally find that both urban population percent increase in the independent variable. and non-dependent population are positively Appendix Table A2 provides descriptive sta- associated with emissions, and services as tistics for the substantive variables included percent of GDP is negatively associated with in the study. All variable transformation infor- carbon pollution. mation and the Stata code used to estimate the Additional models include trade as percent reported models are available from the lead of GDP, also obtained from the World Bank author upon request. (2022), and level of democratization in the The baseline model we estimate for per form of the institutionalized democracy index, capita emissions is as follows: an additive 11-point scale (with higher values meaning greater levels of democracy), which CO Emissions percapita 2 it , we obtained from the Center for Systemic Peace =+ ββ GDP percapita MEPS 12 it ,, it and Societal-Systems Research (2018). Total + β MPR + β Urban Population population, which counts all residents regard- 3 iit ,, 4 it less of legal status or citizenship, is included + β Nond - ependent Population 5 it , in the models of total carbon emissions. These + β Seervices %. GDP ++ αε u + 6 it ,, it it data come from the World Bank (2022). (1) 10 American Sociological Review 00(0) The baseline model with the inclusion of the significant in Models 3 through 5. The esti- three-way interaction is as follows: mated effect of non-dependent population on per capita emissions is positive across all five models, the effect of services as percent of CO Emissions percapita 2 it , GDP is negative in all but the second and fifth =+ ββ GDP percapita MEPS 12 it ,, it models, and the effect of urban population + β MPR + β GDP percapita *MEPS 3 iit ,, 4 it it , is positive and statistically significant in the + β GDP percapita *MPR 5 it ,, it first two models. The results, particularly the + β M MEPS *MPR significant coefficient for GDP per capita × 6 it ,, it MEPS × MPR, confirm our arguments. + β GDP percapita ** MEPS MPR 7 it ,, it it , To provide a more nuanced assessment + β Urban Poppulation 8 it , and clearer interpretation of the three-way + β Nond - ependent Population 9 it , interaction, Figures 1 and 2 plot the average + β Services %GDP ++ αε u + . 10 i,,ti ti,t marginal effects, with 95 percent confidence (2) intervals (95 percent CI), of GDP per capita by MEPS and MPR. The estimates are based REsuLTs on Model 3 in Table 1, which we generate using Stata’s margins command. Although Table 1 reports five models of per capita the 95 percent confidence intervals of the carbon emissions. Model 1 is the initial estimates overlap for most of the marginal baseline, consisting of GDP per capita, mili- effects, the differences between the point esti- tary expenditures per soldier (MEPS), and mates for the marginal effects are statistically military participation rate (MPR), as well significant unless noted otherwise. as urban population, non-dependent popula- Figure 1 reports the marginal effect of tion, and services as percent of GDP. Model GDP per capita on per capita emissions at 2 introduces each of the two-way interac- the 10th, 50th, and 90th percentiles of MEPS tions for GDP per capita, MEPS, and MPR. across levels of MPR. The marginal effect of Models 3 through 5 include their three-way GDP per capita at each percentile of MEPS interaction, with Model 3 for the overall panel increases across the MPR distribution, with dataset of 106 nations and Model 4 for the the exception of the 10th percentile of MEPS perfectly balanced panel dataset reduced to ($3,315). The effects in this case are statisti- 53 nations. Model 5 is for the overall dataset, cally equivalent across the MPR distribution, and also controls for trade as percent of GDP, ranging from .278 (95 percent CI = .185 to democratization, and military expenditures as .370) at the 10th percentile of MPR to .255 percent of government expenditures. For ease (95 percent CI = .151 to .359) at the 90th of interpretation, we exclude the estimated percentile of MPR. At the 50th percentile of coefficients for these three additional controls MEPS ($20,075), the effect of GDP per capita in Table 1 (all not statistically significant), but on emissions ranges from .303 (95 percent they are provided in Appendix Table A3. CI = .222 to .384) at the 10th percentile of Model 1 indicates that per capita emissions MPR (1.28 percent) to .385 (95 percent CI = .290 is positively associated with GDP per capita to .480) at the 90th percentile of MPR (4.31 and MEPS, and the effect of MPR is not statis- percent). At the 90th percentile of MEPS tically significant. In Model 2, the estimated ($179,553), the effect of GDP per capita coefficients for GDP per capita × MEPS and ranges from .333 (95 percent CI = .248 to GDP per capita × MPR are positive, whereas .418) at the 10th percentile of MPR (1.28 per- the coefficient for MEPS × MPR is not statis- cent) to .544 (95 percent CI = .432 to .656) tically significant. The estimated coefficient at the 90th percentile of MPR (4.31 percent). for the three-way interaction, GDP per capita Figure 2 provides the marginal effect of × MEPS × MPR, is positive and statistically GDP per capita on per capita carbon emissions Jorgenson et al. 11 Table 1. Elasticity Coefficients for the Regression of Carbon Emissions per Capita, 1990 to Model 1 Model 2 Model 3 Model 4 Model 5 *** *** *** *** *** GDP per Capita .400 .333 .342 .505 .361 (.045) (.044) (.041) (.074) (.056) *** *** *** * ** Military Expenditures per Soldier (MEPS) .068 .098 .097 .063 .097 (.014) (.020) (.018) (.026) (.031) Military Participation Rate (MPR) .023 .070 .001 –.019 –.067 (.043) (.049) (.047) (.068) (.081) *** *** ** *** GDP per Capita × MEPS .030 .039 .063 .045 (.009) (.009) (.019) (.010) * * * ** GDP per Capita × MPR .065 072 .142 .126 (.026) (.030) (.060) (.046) MEPS × MPR .015 .024 .032 –.023 (.019) (.023) (.047) (.036) *** ** *** GDP per Capita × MEPS × MPR .048 .073 .053 (.012) (.026) (.016) * * Urban Population .226 .265 .213 .330 .273 (.109) (.114) (.113) (.192) (.140) *** *** *** *** *** Non-dependent Population 1.116 1.093 1.055 1.251 1.271 (.274) (.271) (.251) (.315) (.278) * * ** Services as % GDP –.123 –.105 –.111 –.320 –.152 (.058) (.055) (.053) (.114) (.084) R Overall .825 .823 .820 .786 .841 Note: For Models 1, 2, and 3, N = 2,563 for 106 nations, with 24.2 mean observations per nation. For Model 4, N = 1,431 for 53 nations, with 27 mean observations per nation. For Model 5, N = 2,079 for 100 nations, with 20.8 mean observations per nation. Model 5 also controls for trade as % GDP, democratization, and military expenditures as % government expenditures. Robust standard errors clustered by nation are in parentheses. GDP per capita, MEPS, and MPR are mean centered. All variables are in logarithmic form. All models include nation-specific fixed effects derived from the within estimator. All models include unreported year-specific intercepts. * ** *** p < .05; p < .01; p < .001 (two-tailed). at the 10th, 50th, and 90th percentiles of (4.31 percent), the effect of GDP per capita MPR across levels of MEPS. Like in Figure ranges from .255 (95 percent CI = .151 to 1, the marginal effect of GDP per capita at .359) at the 10th percentile of MEPS ($3,315) each percentile of MPR increases across the to .544 (95 percent CI = .432 to .656) at the MEPS distribution, with the exception of the 90th percentile of MEPS ($179,553). 10th percentile of MPR (1.28 percent). The statistically equivalent effects in this case Additional Models range from .277 (95 percent CI = .185 to .370) at the 10th percentile of MEPS to .333 To further broaden the testing of the three- (95 percent CI = .248 to .418) at the 90th way interaction, we estimate models for per percentile of MEPS. At the 50th percentile capita emissions and total emissions that of MPR (2.01 percent), the effect of GDP per include additional controls, which we report capita on emissions ranges from .269 (95 per- in Appendix Table A6. First, we estimate cent CI = .185 to .354) at the 10th percentile models that control for renewable energy of MEPS ($3,315) to .412 (95 percent CI = consumption. Next, we control for arms .325 to .499) at the 90th percentile of MEPS exports. Due to their unavailability for many ($179,553). At the 90th percentile of MPR nations, including arms exports greatly 12 American Sociological Review 00(0) Average Marginal Effects of GDP per capita by Military Expenditures per Soldier $3,315 per Soldier $20,075 per Soldier $179,553 per Soldier 1.28% 2.01% 4.31% 1.28% 2.01% 4.31% 1.28% 2.01% 4.31% Military Participation Rate Figure 1. Marginal Effects of GDP per Capita for Model of CO Emissions per Capita by Military Expenditures per Soldier Note: $3,315, $20,075, and $179,553 are the 10th, 50th, and 90th percentiles for the distribution of military expenditures per soldier; 1.28 percent, 2.01 percent, and 4.31 percent are the 10th, 50th, and 90th percentiles for the distribution of military participation rate. reduces the overall sample. Finally, we esti- and indicate that the estimated effect of the mate models that control for oil production. two-way interaction on both per capita emis- The oil production data are also limited to a sions and total emissions is not statistically relatively small number of nations. Across significant. These findings further validate each model of emissions, the estimated coef- our focus on military expenditures per solider ficient for GDP per capita × MEPS × MPR and military participation rate as far-reaching is positive and statistically significant. As and distinct characteristics of militarization expected, the estimated effect of renewable that shape the effect of economic growth on energy consumption is negative and statisti- nations’ carbon pollution. cally significant. The estimated effects of arms exports and oil production are not sta- Robustness Checks and Sensitivity tistically significant. Analyses To determine if other national military measures moderate the effect of economic An interaction in fixed-effects regression is growth on emissions, we estimate models usually specified by demeaning the prod- that include the two-way interaction between uct term. Giesselmann and Schmidt-Catran GDP per capita and military expenditures as (2022) show that demeaning the product percent of government expenditures. As noted between time-varying variables may not pro- in the Data and Methods section, this mili- duce a true within-unit estimate because it tary measure is moderately correlated with incorporates between-unit differences. They MPR and weakly correlated with MEPS. The propose using the double-demeaned esti- models are reported in Appendix Table A7 mator, which gives unbiased results but is Effects on Linear Prediction Jorgenson et al. 13 Average Marginal Effects of GDP per capita by Military Participation Rate 1.28% Participation Rate 2.01% Participation Rate 4.31% Participation Rate $3,315 $20,075 $179,553 $3,315 $20,075 $179,553 $3,315 $20,075 $179,553 Military Expenditures per Soldier Figure 2. Marginal Effects of GDP per Capita for Model of CO Emissions per Capita by Military Participation Rate Note: 1.28 percent, 2.01 percent, and 4.31 percent are the 10th, 50th, and 90th percentiles for the distribution of military participation rate; $3,315, $20,075, and $179,553 are the 10th, 50th, and 90th percentiles for the distribution of military expenditures per soldier. inefficient compared to the fixed-effects esti- .16, which fails to reject the null hypothesis mator. They suggest using a Hausman test that the estimates are statistically equivalent to decide which estimator is more appropri- (p-value = .69). We therefore rely on the ate. If the estimators produce statistically fixed-effects estimates of the three-way inter- identical estimates, then the standard fixed- action in the reported analyses. effects estimator should be used. As a robust- A possible limitation of using year-specific ness check, we perform the double-demeaned fixed effects is that they assume time-specific estimator and extend it using a seemingly shocks homogeneously affect each case in unrelated regression framework. This has the dataset, meaning they may not adequately two advantages over using a Hausman test. model the cross-sectional dependence, poten- First, it allows for robust standard errors, tially leading to biased and inconsistent which the Hausman test does not. Second, it results. Accordingly, we use Pesaran’s test allows us to use a simple Wald test to assess for weak cross-sectional dependence to assess whether the coefficient on the three-way whether the year-specific fixed effects elimi- interaction of interest is statistically different nate the strong cross-sectional dependence across the two models. In contrast, the Haus- from our reported models. The cross-sectional man test assesses the equality of two estima- dependence test statistics of the residuals are tors instead of individual coefficients. We not statistically significant, meaning there is perform this approach for Model 3 in Table no strong cross-sectional dependence, and 1, combining the results of the two estimators the reported two-way fixed-effects models using the suest command in Stata. The Wald are unlikely biased or inconsistent in this way test produces a chi-square test statistic of (Thombs 2022). Effects on Linear Prediction 14 American Sociological Review 00(0) Panel data are often autoregressive, mean- The CCE estimator assumes the cross-sec- ing the data tend to be correlated over time, tional dependence is due to unobserved, time- and excluding the lag of the dependent vari- varying, common factors that affect each case able from the model will result in omitted differently. It approximates the common fac- variable bias if the outcome variable is truly tors by adding cross-sectional averages to the a function of their past value (Pickup 2015). model and estimates a factor loading for each Therefore, as robustness checks, we estimate case in the analysis (Pesaran 2006). We use two-way fixed-effects dynamic models for the pooled version of the CCE because the both per capita emissions and total emissions, relatively short time span prevents us from with a focus on the coefficient for the three- estimating a time-series regression on each way interaction. individual nation (Thombs 2022). For per capita emissions, this model is as A limitation of using the pooled version follows: is that it does not account for the potential issue of slope heterogeneity. We test for this CO Emissions percapita using the instrumental-variable approach, a 2 it , two-stage procedure that works by eliminat- = λ CO Emissions percapita 12 it , −1 ing the common factors in the covariates + β GDP Pper capita + β MEPS 1 it ,, 2 it using principal component analysis in stage ++ ββ MPR GDP percapita *MEPS 34 it ,, it i,tt one, and obtains consistent estimates using + β GDP percapita *MPR defactored covariates as instruments (Norkutė 5 it ,, it et al. 2021). In stage two, the whole model is + β MEPS *MPR 6 it ,, it defactored using the residuals from stage one, + β GDP percapitaaM ** EPS MPR 7 it ,, it it , and instrumental-variable estimation is per- + β Urban Population 8 it , formed using the same instruments from the + β Nond -o ependent Population 21 9 it , first stage (Kripfganz and Sarafidis 2021). ++ βα Services %. GDP ++ u ε This estimation technique is robust to Nickell 10 it ,, it it (3) bias, and we test the effect of slope hetero- There are potential issues to consider regard- geneity on the model with the Hansen test of ing fixed-effects estimation of a dynamic overidentifying restrictions (J-statistic). model. First, estimating dynamic panel mod- Table 2 reports the estimates of the four els can produce the “Nickell bias” (Nickell robustness check models for per capita emis- 1981). The bias stems from the correlation sions. Appendix Table A8 reports the same between the lagged dependent variable and sequence of estimated models for total carbon the error term, a product of the demeaning emissions. Model 1 is for the static pooled process of fixed-effects estimation. However, common correlated effects estimator, and the this bias tends to lessen as T increases (Hsiao, remaining three are dynamic models and thus Pesaran, and Tahmiscioglu 2002; Thombs include the lagged dependent variable. Model 2022). Second, fixed-effects estimation of a 2 is for the two-way fixed-effects dynamic dynamic model with slope heterogeneity can model, Model 3 is for the dynamic pooled lead to inconsistent and misleading estimates common correlated effects estimator, and (Pesaran and Smith 1995; Thombs, Huang, Model 4 is for the two-stage instrumental- and Fitzgerald 2022). variable estimator. To address these concerns and to model The estimated coefficient for GDP per cross-sectional dependence in alterna- capita × MEPS × MPR is positive and statis- tive ways, we also estimate models with tically significant in each model, regardless of the common correlated effects (CCE) esti- estimator type. The lagged dependent variable mator (Ditzen 2018; Pesaran 2006) and the has a positive effect in each dynamic model, instrumental-variable estimation approach and the J-statistic is not statistically signifi- with common factors (Norkutė et al. 2021). cant in the two-stage instrumental-variable Jorgenson et al. 15 Table 2. Elasticity Coefficients for the Regression of Carbon Emissions per Capita, 1990 to Model 1 Model 2 Model 3 Model 4 ** *** *** *** GDP per Capita × MEPS × MPR .022 .013 .019 .016 (.009) (.004) (.005) (.005) *** *** *** *** GDP per Capita .526 .119 .084 .277 (.054) (.017) (.016) (.033) *** *** * *** Military Expenditures per Soldier (MEPS) .046 .027 .018 .026 (.013) (.006) (.008) (.007) * ** * Military Participation Rate (MPR) .006 .032 .049 .051 (.038) (.015) (.018) (.020) ** * ** GDP per Capita × MEPS .024 .007 .005 .014 (.009) (.003) (.004) (.005) ** GDP per Capita × MPR .035 .010 .018 .032 (.022) (.011) (.011) (.012) * ** ** MEPS × MPR .033 .015 .022 .026 (.014) (.008) (.007) (.010) *** *** *** CO per Capita Lagged .716 .712 .373 (.027) (.030) (.065) R Overall .983 J-statistic 7.066 Note: Model 1 is for the static pooled common correlated effects estimator. Model 2 is for the two-way fixed-effects dynamic model with clustered robust standard errors. Model 3 is for the dynamic pooled common correlated effects estimator. Westerlund, Perova, Norkute standard errors are reported for Models 1 and 3. Model 4 is for the two-stage instrumental-variable estimation with two lags used as instruments. For Model 1, N = 2,439 for 96 nations, with 26 mean observations per nation. For Model 2, N = 2,477 for 106 nations, with 23.4 mean observations per nation. For Model 3, N = 2,096 for 96 nations, with 22 mean observations per nation. For Model 4, N = 2,256 for 106 nations, with 21.3 mean observations per nation. Due to insufficient observations, the analysis drops 10 nations from Models 1 and 3. Standard errors are in parentheses. All variables are in logarithmic form. J-statistic for Model 4 is not statistically significant (H : overidentifying restrictions are valid). All models control for urban population, non-dependent population, and services as % GDP. GDP per capita, MEPS, and MPR are mean centered. * ** *** p < .05; p < .01; p < .001 (two-tailed). models. Overall, the findings of interest for Counterfactual Analysis and this study appear robust to a variety of poten- Substantive Significance tial modeling concerns. Finally, to determine if the analyses and Having demonstrated that our results are findings are sensitive to any particular nations robust to a host of modeling considerations included in the study, we re-estimate each and not sensitive to sample characteristics, we reported model where we systematically now turn to the question of substantive sig- exclude, one at a time, each of the 106 nations nificance. Here we ask how the moderating in the overall dataset. The results indicate that effect of militarization matters for observed none of the included nations are overly influ- levels of carbon emissions per capita. In par- ential: the estimated elasticity coefficients ticular, we use Model 3 of Table 1 to engage for the three-way interaction across all re- in a counterfactual history exercise under two estimated models are positive and statistically scenarios for the overall dataset. First, we significant. The estimated coefficients for the ask what average emissions per capita would other independent variables remain consistent look like if every nation in the sample had as well. military expenditures per soldier and military 16 American Sociological Review 00(0) Figure 3. Average CO Emissions per Capita under Different Militarization Scenarios Note: Estimates derived from Model 3 in Table 1 for the overall dataset. 10th percentile militarization refers to average carbon emissions per capita if every nation in the sample had military expenditures per soldier and military participation rates equal to that observed at the 10th percentile of the nation- year distribution. 90th percentile militarization refers to average emissions per capita if every nation in the sample had military expenditures per soldier and military participation rates equal to that observed at the 90th percentile of the nation-year distribution. Carbon emissions reported in metric tons per capita. participation rates equal to that observed at average per capita emissions grew by .614 the 90th percentile of the nation-year distribu- metric tons over the entire period. This num- tion. Second, we ask what average emissions ber falls to .419 metric tons in a world of 10th per capita would look like if every nation in percentile militarization, and the increase the sample instead had military expenditures rises to 1.14 metric tons in a world of 90th per soldier and military participation rates percentile militarization. Holding rates of equal to that observed at the 10th percentile economic growth fixed, worldwide reduc- of the nation-year distribution. tions in the capital intensiveness and size Figure 3 reports the yearly average per of militarization could produce substantial capita carbon emissions under each of these declines in carbon emissions. scenarios, as well as the observed average per capita emissions per year. Consistent with DisCussion AnD our overall intervention, there is a wide gap ConCLusions between the observed emissions and those that would occur under militarization at the 10th This study makes significant contributions and 90th percentiles. In 1990, observed aver- to sociological work on climate change and age emissions are .705 metric tons per capita the climate crisis. Bridging multiple sub- higher than that which would occur under mili- fields, including environmental sociology, tarization at the 10th percentile, and this gap the sociology of development, global political grows to .899 metric tons per capita by 2016. economy, historical sociology, and politi- Conversely, observed emissions are .459 metric cal sociology, we argue that militarization tons per capita lower than what would occur moderates the effect of economic growth with 90th percentile militarization. This gap on nations’ carbon emissions. Many of the grows to .984 metric tons per capita by 2016. world’s national militaries are increasingly The growth rate for emissions also varies capital intensive, with a focus on the develop- considerably under these scenarios. Observed ment of longer-range weapons, transportation, Jorgenson et al. 17 and communications systems, and larger mili- for military purposes are often sold to private taries possess expansive built infrastructures security firms and law enforcement agencies, that require considerable amounts of energy and transformed into commercial products and material goods. We use military expen- for domestic and global markets. Military ditures per soldier and military participation bases and installations routinely obtain fuel rate to measure these far-reaching character- as well as various material goods and services istics of contemporary militarization. from business entities. These structural condi- The findings for the longitudinal analyses tions, institutional relationships, and underly- provide substantial support for our arguments. ing processes all contribute to how the capital Both military expenditures per solider and intensiveness and scale of militaries shape the military participation rate enlarge the posi- association between national emissions and tive effect of economic growth on national economic growth. carbon emissions. We observe these relation- Anthropogenic climate change increases ships through modeling the two-way interac- the likelihood of large-scale conflicts between tions for growth, measured as GDP per capita, and within nations (Alario, Nath, and Carlton- and each military measure. More importantly, Ford 2016; Cane et al. 2014; Giddens 2011; through modeling their three-way interac- Hsiang, Burke, and Miguel 2013; Mach et al. tion, we find that each militarization attribute 2019). This expands and intensifies military increases the extent to which the other ampli- activities for nations involved in international fies the effect of economic growth on carbon and domestic engagements (Belcher, Bigger, pollution. The effect of GDP per capita on et al. 2020; Pathak 2020; Raleigh and Urdal emissions is larger at higher levels of expen- 2007; Smith and Lengefeld 2020), further ditures per solider, and this increases across driving the material- and energy-intensive the distribution of military participation rate. production of weapons systems and muni- Likewise, the effect of GDP per capita on tions, vehicles, communications equipment, carbon pollution is larger at higher levels of and other related goods in the defense indus- military participation rate, and this increases try and private sector more broadly (Isik- across the distribution of military expendi- sal 2021; Jorgenson and Clark 2016). Thus, tures per solider. The results are robust for per as a threat multiplier, anthropogenic climate capita emissions and total emissions, various change could facilitate a greater occurrence sensitivity analyses, a range of balanced and of both domestic and international conflicts, unbalanced panel datasets, and across mul- further propelling the relationships between tiple model specifications. Their substantive national carbon emissions, economic growth, significance is further highlighted through and militarization. counterfactual analysis. Our theoretical arguments and empirical Our findings speak to the deep connec- findings highlight the value and necessity of tions between the military and the economy considering the world’s militaries in sociolog- at the core of the military-industrial complex. ical research. The work of historical sociolo- National militaries help secure access to fos- gists maps out in great detail the emergence sil fuels and other resources, and generally of nation-states from a coalescence of coer- attempt to maintain geopolitical and world- cive power with economic power. Inspired by economic stability, which enables carbon- this rich body of scholarship, we concentrate intensive economic growth. At the same time, on the environmental effects of economy and militaries spur scientific research and techno- military relationships for nations in the mod- logical advances, and they influence produc- ern era. The present study focuses on human tion in the defense industry and private sector drivers of climate change, but contemporary in general, while also serving as major con- forms of militarization likely shape the effect sumers of these items. The technologies and of economic growth on social and other envi- other goods initially developed by industry ronmental outcomes in ways similar to and 18 American Sociological Review 00(0) distinct from how it influences the associa- contemporaneous relationships observed for tion between carbon pollution and economic recent decades. While we use the best pub- growth. licly available aggregate data on national- The characteristics of national militaries level militarization characteristics, these data also likely influence how macrostructural fac- might be underestimated for some country- tors and processes besides economic growth years due to accounting practices and the affect a range of social and environmen- overall classified nature of military-related tal outcomes. Like others (e.g., Andreski information. Consequently, the reported find- 1968; Giddens 1987; Hooks 1990; Kentor ings may underestimate the direct association and Kick 2008), we suggest the military is between national-level emissions and mili- routinely overlooked by scholars across the tarization, as well as the extent to which the discipline. We hope this study will encour- capital intensiveness and size of militariza- age sociologists to consider the military and tion enlarge the effect of economic growth on militarization in future analytic frameworks carbon pollution. and empirical analyses. The ongoing growth In conclusion, this study significantly of militarism underscores the importance in advances the sociological research on climate doing so. From 1990 to 2016, global military change, and enhances sociological contribu- expenditures increased by 29 percent (1.372 tions to interdisciplinary work on planetary to 1.774 trillion constant 2018 U.S. dollars), warming and other global sustainability chal- armed forces personnel for the world grew by lenges. Militaries exert a substantial influence 15 percent (23.918 to 27.542 million person- on the production and consumption patterns nel), and global military expenditures per of economies, as well as the environmental solider increased by 12 percent (57,362 to demands required to support their evolving 64,410 constant 2018 U.S. dollars). infrastructure. Our findings indicate that two Like all research, this study has limi- major characteristics of militarization enlarge tations that can hopefully be addressed in the effect of economic growth on carbon future analyses. Although the overall sample emissions, and they increase the extent to covers the majority of the world’s popula- which the other amplifies the effect of growth tion, current data availability limits the num- on nations’ carbon pollution. By theorizing ber of nations included in the cross-national about these structural relationships and bridg- analyses. Data availability also restricts the ing multiple subfields, we push forward the temporality of the study to slightly over a foundational sociological literature concern- quarter century, from 1990 to 2016. Thus, our ing the effect of economic growth on the analyses focus on the modeling of relatively environment. Jorgenson et al. 19 APPEnDiX Table A1. Number of Annual Observations for Each Country in the Overall Dataset Country Obs. Country Obs. Country Obs. Albania 25 Gabon 13 Netherlands 27 Algeria 27 Gambia, The 23 New Zealand 27 Angola 26 Georgia 21 Nicaragua 27 Argentina 27 Germany 26 Niger 20 Armenia 24 Ghana 26 Nigeria 27 Australia 27 Greece 27 Norway 27 Austria 27 Guinea 17 Oman 27 Bahrain 27 Guinea-Bissau 18 Peru 27 Bangladesh 27 Guyana 24 Philippines 27 Belarus 25 Haiti 9 Portugal 27 Belgium 27 Honduras 20 Romania 27 Belize 25 Hungary 26 Russian Federation 25 Benin 14 India 27 Rwanda 26 Bosnia and Herzegovina 15 Iran, Islamic Rep. 27 Saudi Arabia 27 Botswana 27 Ireland 27 Serbia 11 Brazil 27 Israel 27 Sierra Leone 25 Bulgaria 27 Italy 25 Slovenia 25 Cameroon 26 Jamaica 27 South Africa 27 Canada 27 Japan 27 Spain 27 Chile 27 Jordan 27 Sudan 20 China 27 Kazakhstan 24 Sweden 27 Colombia 27 Kenya 26 Switzerland 26 Congo, Dem. Rep. 22 Korea, Rep. 27 Tajikistan 20 Congo, Rep. 13 Lebanon 27 Tanzania 26 Cote d'Ivoire 15 Lesotho 27 Timor-Leste 11 Croatia 22 Lithuania 22 Togo 18 Cuba 12 Madagascar 26 Trinidad and Tobago 18 Cyprus 25 Malaysia 27 Tunisia 27 Denmark 27 Mali 25 Turkey 27 Dominican Republic 27 Mexico 27 United Arab Emirates 18 Ecuador 27 Moldova 22 United Kingdom 27 El Salvador 27 Montenegro 11 United States 27 Ethiopia 26 Morocco 27 Uruguay 26 Fiji 27 Namibia 26 Yemen, Rep. 25 Finland 27 Nepal 27 Zambia 22 France 25 Note: N = 2,563. Obs. = number of annual observations. 20 American Sociological Review 00(0) Table A2. Descriptive Statistics Std. N Mean Deviation Carbon Emissions per Capita 2,563 1.390 .899 Total Carbon Emissions 2,563 10.187 2.221 GDP per Capita 2,563 8.590 1.525 Military Expenditures per Soldier 2,563 9.989 1.429 Military Participation Rate 2,563 .772 .468 Urban Population 2,563 3.988 .465 Non-dependent Population 2,563 4.121 .111 Services as % GDP 2,563 3.934 .238 Trade as % GDP 2,434 4.183 .502 Democratization 2,460 1.705 .877 Military Expenditures as % Government Expenditures 2,269 1.962 .615 Total Population 2,563 16.348 1.559 Renewable Energy Consumption 2,303 2.971 1.29 Arms Exports 709 18.673 2.245 Oil Production 801 3.732 1.36 Note: All variables are in logarithmic form. Table A3. Elasticity Coefficients for the Regression of Carbon Emissions per Capita, 1990 to Model 1 Model 2 Model 3 Model 4 *** *** *** *** GDP per Capita × MEPS × MPR .049 .051 .050 .053 (.012) (.013) (.015) (.016) *** *** *** *** GDP per Capita .352 .349 .360 .361 (.042) (.042) (.056) (.056) *** *** *** ** Military Expenditures per Soldier (MEPS) .088 .096 .100 .097 (.019) (.018) (.028) (.031) Military Participation Rate (MPR) –.034 –.012 –.002 –.067 (.049) (.050) (.068) (.081) *** *** *** *** GDP per Capita × MEPS .035 .040 .045 .045 (.009) (.009) (.010) (.010) * * * ** GDP per Capita × MPR .082 .074 .082 .126 (.037) (.035) (.036) (.046) MEPS × MPR .008 .022 .011 –.023 (.026) (.026) (.032) (.036) Trade as % GDP .002 .020 (.025) (.027) Democratization .010 –.002 (.011) (.014) Military Expenditures as % Government –.014 –.014 Expenditures (.035) (.039) R Overall .833 .817 .827 .841 Note: All models control for urban population, non-dependent population, and services as % GDP. Robust standard errors clustered by nation are in parentheses. All variables are in logarithmic form. All models include nation-specific fixed effects derived from the within estimator. All models include unreported year-specific intercepts. GDP per capita, MEPS, and MPR are mean centered. For Model 1, N = 2,434 for 103 nations with 23.6 mean observations per nation. For Model 2, N = 2,460 for 104 nations with 23.7 mean observations per nation. For Model 3, N = 2,269 for 105 nations with 21.6 mean observations per nation. For Model 4, N = 2,079 for 100 nations with 20.8 mean observations per nation. * ** *** p < .05; p < .01; p < .001 (two-tailed). Jorgenson et al. 21 Table A4. Elasticity Coefficients for the Regression of Total Carbon Emissions, 1990 to 2016 Model 1 Model 2 Model 3 ** * *** GDP per Capita × MEPS × MPR .063 .094 .081 (.020) (.036) (.024) *** *** *** GDP per Capita .640 .644 .535 (.095) (.102) (.096) * ** Military Expenditures per Soldier (MEPS) .084 –.018 .144 (.036) (.040) (.054) Military Participation Rate (MPR) –.001 –.058 .030 (.072) (.106) (.138) GDP per Capita × MEPS –.006 .034 .013 (.015) (.024) (.018) GDP per Capita × MPR –.009 .010 .082 (.060) (.097) (.083) MEPS × MPR .070 .086 –.032 (.041) (.065) (.063) * ** Urban Population .626 .824 .453 (.303) (.305) (.315) *** * Non-dependent Population .598 1.381 1.031 (.357) (.405) (.408) * * Services as % GDP –.261 –.343 –.157 (.102) (.154) (.126) *** *** *** Total Population 1.418 1.141 1.400 (.141) (.204) (.175) R Overall .877 .922 .888 Note: For Model 1, N = 2,563 for 106 nations with 24.2 mean observations per nation. For Model 2, N = 1,431 for 53 nations with 27 mean observations per nation. For Model 3, N = 2,079 for 100 nations with 20.8 mean observations per nation. Model 3 also controls for trade as % GDP, democratization, and military expenditures as % government expenditures. Robust standard errors clustered by nation are in parentheses. GDP per capita, MEPS, and MPR are mean centered. All variables are in logarithmic form. All models include nation-specific fixed effects derived from the within estimator. All models include unreported year-specific intercepts. * ** *** p < .05; p < .01; p < .001 (two-tailed). 22 American Sociological Review 00(0) Table A5. Pairwise Differences of Marginal Effects Pairwise Difference Figure 1 Figure 2 10th MEPS MPR 10th vs. 50th –.46 MPR 10th vs. 90th –.46 MPR 50th vs. 90th –.46 50th MEPS MPR 10th vs. 50th 2.24 MPR 10th vs. 90th 2.24 MPR 50th vs. 90th 2.24 90th MEPS *** MPR 10th vs. 50th 4.78 *** MPR 10th vs. 90th 4.78 *** MPR 50th vs. 90th 4.78 10th MPR MEPS 10th vs. 50th 1.42 MEPS 10th vs. 90th 1.42 MEPS 50th vs. 90th 1.42 50th MPR *** MEPS 10th vs. 50th 4.26 *** MEPS 10th vs. 90th 4.26 *** MEPS 50th vs. 90th 4.26 90th MPR *** MEPS 10th vs. 50th 5.58 *** MEPS 10th vs. 90th 5.58 *** MEPS 50th vs. 90th 5.58 Note: z-statistics reported. * ** *** p < .05; p < .01; p < .001 (two-tailed). 23 Table A6. Elasticity Coefficients for the Regression of Carbon Emissions per Capita and Total Carbon Emissions, 1990 to 2015 and 1990 to 2016 Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 ** ** *** ** ** ** GDP per Capita × MEPS × MPR .032 .051 .080 .069 .111 .140 (.010) (.018) (.020) (.024) (.033) (.050) *** *** *** *** *** *** GDP per Capita .335 .476 .536 .667 .431 .780 (.034) (.086) (.088) (.094) (.076) (.123) *** * ** *** Military Expenditures per Soldier (MEPS) .073 .067 .087 .088 .113 .150 (.016) (.029) (.044) (.044) (.035) (.053) Military Participation Rate (MPR) .005 .008 .124 .167 –.107 .020 (.038) (.061) (.088) (.096) (.104) (.132) *** *** ** ** GDP per Capita × MEPS .037 –.008 .067 .056 .063 .056 (.007) (.014) (.018) (.018) (.021) (.033) GDP per Capita × MPR .041 –.038 .030 –.042 .008 .095 (.024) (.054) (.053) (.076) (.042) (.051) ** *** ** *** MEPS × MPR .026 .052 .167 .212 .117 .192 (.018) (.035) (.053) (.059) (.040) (.045) *** *** *** Total Population 1.211 1.569 2.051 (.128) (.195) (.212) *** *** Renewable Energy Consumption –.211 –.310 (.023) (.042) Arms Exports .005 .004 (.004) (.005) Oil Production .058 .062 (.036) (.039) R Overall .888 .911 .454 .888 .713 .802 Note: Odd numbered models are for carbon emissions per capita. Even numbered models are for total carbon emissions. All models control for urban population, non-dependent population, and services as % GDP. Robust standard errors clustered by nation are in parentheses. GDP per capita, MEPS, and MPR are mean centered. All variables are in logarithmic form. All models include nation-specific fixed effects derived from the within estimator . All models include unreported year-specific intercepts. For Models 1 and 2, N = 2,303 for 106 nations with 21.7 mean observations per nation. For Models 3 and 4, N = 709 for 28 nations with 25.3 mean observations per nation. For Models 5 and 6, N = 801 for 32 nations with 25.0 mean observations per nation. Models 1 and 2 are for 1990 to 2015; Models 3 to 6 are for 1990 to 2016. * ** *** p < .05; p < .01; p < .001 (two-tailed). 24 American Sociological Review 00(0) Table A7. Elasticity Coefficients for the Regression of Carbon Emissions per Capita and Total Carbon Emissions, 1990 to 2016 Model 1 Model 2 *** *** GDP per Capita .396 .543 (.067) (.133) Military Expenditures as % Government Expenditures (MGE) .024 –.244 (.124) (.237) GDP per Capita × MGE –.002 .026 (.017) (.029) * * Military Expenditures per Soldier (MEPS) .055 .122 (.024) (.051) Military Participation Rate (MPR) .014 .162 (.062) (.108) *** Total Population 1.400 (.164) R Overall .833 .881 Note: Model 1 is for carbon emissions per capita. Model 2 is for total carbon emissions. Both models control for urban population, non-dependent population, and services as % GDP. All variables are in logarithmic form. Robust standard errors clustered by nation are in parentheses. Both models include nation-specific fixed effects derived from the within estimator. Both models include unreported year- specific intercepts. N = 2,269 for 105 nations with 21.6 mean observations per nation. * ** *** p < .05; p < .01; p < .001 (two-tailed). Jorgenson et al. 25 Table A8. Elasticity Coefficients for the Regression of Total Carbon Emissions, 1990 to 2016 Model 1 Model 2 Model 3 Model 4 ** * *** ** GDP per Capita × MEPS × MPR .054 .027 .029 .025 (.019) (.010) (.008) (.009) *** *** *** *** GDP per Capita .909 .294 .303 .543 (.096) (.057) (.034) (.071) * ** * Military Expenditures per Soldier (MEPS) .044 .051 .009 .024 (.021) (.017) (.012) (.010) ** Military Participation Rate (MPR) –.102 .046 .017 .077 (.056) (.039) (.024) (.029) GDP per Capita × MEPS –.005 –.008 –.008 –.005 (.017) (.008) (.007) (.007) GDP per Capita × MPR –.036 –.020 –.016 –.035 (.046) (.034) (.017) (.018) * *** MEPS × MPR .047 .039 .026 .064 (.029) (.024) (.011) (.015) *** *** *** *** Total Population 1.000 .622 .340 .845 (.214) (.127) (.079) (.128) *** *** *** Total CO Lagged .558 .630 .331 (.078) (.028) (.063) R Overall .973 J-statistic 15.125 Note: Model 1 is for the static pooled common correlated effects estimator. Model 2 is for the two-way fixed-effects dynamic model with clustered robust standard errors. Model 3 is for the dynamic pooled common correlated effects estimator. Westerlund, Perova, Norkute standard errors reported for Models 1 and 3. Model 4 is for the two-stage instrumental-variable estimation with two lags used as instruments. For Model 1, N = 2,422 for 95 nations, with 26 mean observations per nation. For Model 2, N = 2,477 for 106 nations, with 23.4 mean observations per nation. For Model 3, N = 2,096 for 96 nations, with 22 mean observations per nation. For Model 4, N = 2,256 for 106 nations, with 21.3 mean observations per nation. Due to insufficient observations, the analysis drops 11 nations from Model 1 and 10 nations from Model 3. Standard errors are in parentheses. All variables are in logarithmic form. J-statistic for Model 4 is not statistically significant (H : overidentifying restrictions are valid). All models control for urban population, non-dependent population, and services as % GDP. GDP per capita, MEPS, and MPR are mean centered. * ** *** p < .05; p < .01; p < .001 (two-tailed). Acknowledgments oRCiD iDs Prior versions of this paper were presented at the Andrew K. Jorgenson https://orcid.org/0000-0002 Center for Environmental Politics’ Duck Family Col- -5902-3704 loquium Series at the University of Washington, the Brett Clark https://orcid.org/0000-0002-3929-1322 Schiller Institute for Integrated Science and Society’s Ryan P . Thombs https://orcid.org/0000-0002-8823-6143 Climate Change Research Seminar Series, the Envi- Jeffrey Kentor https://orcid.org/0000-0001-6735 ronmental Sociology Workshop at Boston College, the -4012 Department of Sociology at the University of British Columbia, and the conference on “Navigating Uncertain Jennifer E. Givens https://orcid.org/0000-0003-3103-9795 Futures: Social Engagement and Transformative Change Xiaorui Huang https://orcid.org/0000-0002-3823-0253 in Global Socio-Ecological Systems,” co-hosted by the Daniel Auerbach https://orcid.org/0000-0002-0984-4504 Environment and Society Research Committee of the Matthew C. Mahutga https://orcid.org/0000-0003 International Sociological Association and the Section -4696-529X on Environmental Sociology of the American Sociologi- cal Association. The authors thank the participants for their helpful feedback. The authors also thank Wesley notes Longhofer, Ali Kadivar, Katrina Paxton, the editors of ASR, and the anonymous reviewers for their helpful com- 1. For recent examples, see https://www.global ments and suggestions. change.gov/content/social-science-perspectives- 26 American Sociological Review 00(0) climate-change-workshop; https://www.ipcc.ch/ lines in reporting and accounting for armed forces report/sixth-assessment-report-working-group- across nations. ii/; https://www.ipcc.ch/report/sixth-assessment- 11. To assess if the positive coefficient for the three- report-working-group-3/. way interaction holds across different measures of 2. Using JSTOR and official journal webpages, as of national carbon emissions, we estimate three mod- February 2023, we found only 37 research articles els of total emissions, which we report in Appendix with “military” or “militarization” in their titles Table A4. The models control for urban population, ever published in the two leading generalist soci- non-dependent population, services as percent of ology journals, American Sociological Review (17 GDP, and total population. Model 1 is for the overall total published) and American Journal of Sociology unbalanced panel dataset, and Model 2 is for the per- (20 total published). fectly balanced panel dataset reduced to 53 nations. 3. See https://www.fedcenter.gov/programs/eo14057/; Model 3 is for the larger unbalanced panel dataset https://www.whitehouse.gov/briefing-room/ and also controls for trade as percent of GDP, democ- presidential-actions/2021/12/08/executive-order- ratization, and military expenditures as percent of on-catalyzing-clean-energy-industries-and-jobs- government expenditures. The estimated coefficient through-federal-sustainability/. for GDP per capita × MEPS × MPR is positive and 4. Tilly (1990) argues that the modern nation-state statistically significant in all three models, providing emerged from a coalescence of coercive/military additional support for our key arguments. power controlled by political units (e.g., despots 12. Appendix Table A5 provides the significance tests and empires) with economic power centered in for the pairwise differences for the marginal effects city-states, beginning around 1000 CE. This fusion reported in Figures 1 and 2. of economic and coercive power was driven by 13. The renewable energy data (as percent of total final political states’ need to obtain sufficient capital to energy consumption) come from the World Bank support the rising cost of military activity, due to (2022). These data are currently unavailable beyond advances in military technology, combined with the year 2015 and are limited in coverage for some city-states’ need for protection. Overall, the mili- nations in our study. Their inclusion reduces the tary played a foundational role in the formation of overall sample to 2,303 total observations for 106 today’s interstate system. nations for 1990 to 2015. 5. Other bodies of research focus on how military 14. The arms exports data come from the Stockholm spending directly affects economic growth. The International Peace Research Institute (SIPRI) results differ across studies, partly depending on the Arms Transfers Database March 2021 Version sample, the analyzed time frame, model estimation (http://www.sipri.org/databases/armstransfers). techniques, and choice of military measures (e.g., SIPRI reports these data using a common unit, the Benoit 1973; Bullock and Firebaugh 1990; Kentor trend-indicator value, which allows for the overall and Kick 2008; Weede 1983). Whether such spend- comparison of arms trade between countries and ing directly increases or decreases economic growth over time. Given limited availability, including could lead to changes in carbon emissions growth arms exports reduces the sample to 709 observa- via mediation. This is distinct from our moderation- tions for 28 nations. oriented hypotheses, which we test with moderation 15. Oil production is measured in million metric tons, analysis. which we obtain from BP’s Statistical Review of 6. They include carbon dioxide produced during con- World Energy Database 2020 Version (https://www sumption of solid, liquid, and gas fuels and gas flar- .bp.com/en/global/corporate/energy-economics/ ing. statistical-review-of-world-energy.html). Given lim- 7. The three variables are grand mean-centered after ited availability, including oil production reduces the log transformation (see the Model Estimation Tech- sample to 801 observations for 32 nations. niques section). 16. In additional models, available from the lead author 8. Less frequently, manufacturing as percent of GDP upon request, we instead include the interaction is included as a control instead of services as per- between GDP per capita and military expenditures cent of GDP. However, manufacturing as percent of as percent of GDP. The estimated effect of the two- GDP has more missing data for nation-years than way interaction is not statistically significant, fur- does services as percent of GDP, and thus the latter ther supporting our focus on military expenditures is generally more desirable as a control for cross- per solider and military participation rate as mod- national panel analyses. erators. 9. For the equations, α refers to the nation-specific 17. The double-demeaned estimator estimates the fixed effects, u refers to the year-specific fixed within effect by demeaning each variable and then effects, and ε refers to the error term. demeaning the product of the two. i,t 10. The direct associations between national carbon 18. Seemingly unrelated regression allows for a set of emissions and the military measures are likely equations to be correlated with each other, and in underestimated, given varying practices and guide- this case, works by combining the results for the Jorgenson et al. 27 fixed-effects estimator and the double-demeaned Avant, Deborah. 2005. The Market for Force: The Con- estimator into a single model. sequences of Privatizing Security. Cambridge, UK: 19. These test statistics are available from the lead Cambridge University Press. author upon request. Baran, Paul A., and Paul M. Sweezy. 1966. Monopoly 20. For total emissions, the model also includes total Capital. New York: Monthly Review Press. population as an independent variable. Beckley, Michael. 2010. “Economic Development and 21. With this approach, the regressors are assumed to be a Military Effectiveness.” The Journal of Strategic function of unobserved common factors and an idiosyn- Studies 33(1):43–79. cratic error term that is independent of the model error Belcher, Oliver, Patrick Bigger, Benjamin Neimark, and term, and thus the defactored lags of the regressors are Cara Kennelly. 2020. “Hidden Carbon Costs of the valid instruments. We estimate models with two lags of ‘Everywhere War’: Logistics, Geopolitical Ecol- the covariates as the defactored instruments. ogy, and the Carbon Boot-Print of the US Military.” 22. If a heterogeneous model is more appropriate than Transactions of the Institute of British Geographers the pooled model, then the test of the overidentify- 45(1):65–80. ing restrictions in the pooled model will result in Belcher, Oliver, Benjamin Neimark, and Patrick Bigger. a statistically significant J-statistic (Kripfganz and 2020. “The U.S. Military Is Not Sustainable.” Science Sarafidis 2021). 367(6481):989–90. 23. These data come from SIPRI (2022) and the World Benoit, Emile. 1973. Defense and Economic Growth in Bank (2022). We provide the global measures for Developing Countries. Lexington, MA: Lexington 1990 to 2016 to correspond with our cross-national Books. analyses for the same period. Bigger, Patrick, and Benjamin Neimark. 2017. “Wea- ponizing Nature: The Geopolitical Ecology of the US Navy’s Biofuel Program.” Political Geography References 60:13–22. Black, Jeremy. 2008. War and the World: Military Power Adams, Gordon. 1982. The Iron Triangle. Piscataway, and the Fate of Continents, 1450–2000. New Haven, NJ: Transaction Publishers. CT: Yale University Press. Adua, Lazarus, Brett Clark, and Richard York. 2021. Block, Fred. 1980. “Economic Instability and Military “The Ineffectiveness of Efficiency: The Paradoxi- Strength.” Politics and Society 10(1):35–58. cal Effects of State Policy on Energy Consumption Bohr, Jeremiah, and Riley E. Dunlap. 2018. “Key Top- in the United States.” Energy Research & Social ics in Environmental Sociology, 1990–2014: Results Science 71:101806 (https://doi.org/10.1016/j.erss from a Computational Text Analysis.” Environmental .2020.101806). Sociology 4(2):181–95. Adua, Lazarus, Richard York, and Beth-Anne Schuelke- Boswell, Terry. 1989. “Colonial Empires and the Capital- Leech. 2016. “The Human Dimensions of Climate ist World-Economy: A Time Series Analysis of Colo- Change: A Micro-Level Assessment of Views from nization, 1640–1960.” American Sociological Review the Ecological Modernization, Political Economy 54(2):180–96. and Human Ecology Perspectives.” Social Science Boswell, Terry, and William Dixon. 1990. “Dependency Research 56:26–43. and Rebellion: A Cross-National Analysis.” Ameri- Alario, Margarita V., Leda Nath, and Steve Carlton-Ford. can Sociological Review 55(4):540–59. 2016. “Climatic Disruptions, Natural Resources, and Bradford, John Hamilton, and Alexander M. Stoner. Conflict: The Challenges to Governance.” Journal of 2017. “The Treadmill of Destruction in Comparative Environmental Studies and Sciences 6:251–59. Perspective: A Panel Study of Military Spending and Alic, John, Daniel Sarewitz, Charles Weiss, and William Carbon Emissions, 1960–2014.” Journal of World- Bonvillian. 2010. “A New Strategy for Energy Inno- Systems Research 23(2):298–325. vation.” Nature 466:316–17. Brady, David, Jason Beckfield, and Martin Seeleib-Kai- Almeida, Paul, and Christopher Chase-Dunn. 2018. “Glo- ser. 2005. “Economic Globalization and the Welfare balization and Social Movements.” Annual Review of State in Affluent Democracies, 1975–2001.” Ameri- Sociology 44:189–211. can Sociological Review 70(6):921–48. Alvarez, Camila H. 2016. “Militarization and Water: A Brady, David, Jason Beckfield, and Wei Zhao. 2007. Cross-National Analysis of Militarism and Fresh- “The Consequences of Economic Globalization for water Withdrawals.” Environmental Sociology Affluent Democracies.” Annual Review of Sociology 2(3):298–305. 33:313–34. Alvarez, Camila H. 2021. “Military, Race, and Urban- Braswell, Taylor. 2022. “Extended Spaces of Environmen- ization: Lessons of Environmental Injustice from tal Injustice: Hydrocarbon Pipelines in the Age of Plan- Las Vegas, Nevada.” Sociological Perspectives etary Urbanization.” Social Forces 100(3):1025–52. 64(3):325–42. Brauer, Jurgen. 2009. War and Nature: The Environmen- Andreski, Stanislav. 1968. Military Organization and tal Consequences of War in a Globalized World. Lan- Society, 2nd ed. Berkeley: University of California ham, MD: Alta Mira Press. Press. 28 American Sociological Review 00(0) Bugden, Dylan. 2022. “Technology, Decoupling, and Clark, Brett, and Andrew K. Jorgenson. 2012. “The Tread- Ecological Crisis: Examining Ecological Modern- mill of Destruction and the Environmental Impacts of ization Theory through Patent Data.” Environmental Militaries.” Sociology Compass 6(7):557–69. Sociology 8(2):228–41. Clark, Brett, Andrew K. Jorgenson, and Jeffrey Kentor. Bullock, Brad, and Glenn Firebaugh. 1990. “Guns and 2010. “Militarization and Energy Consumption: A Butter? The Effect of Militarization on Economic and Test of Treadmill of Destruction Theory in Compara- Social Development in the Third World.” Journal of tive Perspective.” International Journal of Sociology Political and Military Sociology 18(2):231–66. 40(2):23–43. Bunker, Stephen G., and Paul S. Ciccantell. 2005. Glo- Clark, Brett, and Richard York. 2005. “Carbon Metabo- balization and the Race for Resources. Baltimore, lism: Global Capitalism, Climate Change, and the MD: Johns Hopkins University Press. Biospheric Rift.” Theory and Society 34(4):391–428. Burmaoglu, Serhat, and Ozcan Sarıtas. 2017. “Changing Clark, Robert, and Matthew Mahutga. 2013. “Explaining Characteristics of Warfare and the Future of Mili- the Trade-Growth Link: Assessing Diffusion-Based tary R&D.” Technological Forecasting and Social and Structure-Based Models of Exchange.” Social Change 116:151–61. Science Research 42(2):401–17. Burnett, Mackenzie, and Katherine J. Mach. 2021. CNA. 2007. “National Security and the Threat of Climate “A ‘Precariously Unprepared’ Pentagon? Climate Change.” Alexandria, VA: CNA Corporation (https:// Security Beliefs and Decision-Making in the U.S. www.cna.org/cna_files/pdf/national%20security%20 Military.” Global Environmental Change 70:102345 and%20the%20threat%20of%20climate%20change.pdf). (https://doi.org/10.1016/j.gloenvcha.2021.102345). Collins, Randall. 1981. “Does Modern Technology Burns, Thomas J., Byron L. Davis, and Edward L. Kick. Change the Rules of Geopolitics?” Journal of Politi- 1997. “Position in the World-System and National cal & Military Sociology 9(2):163–77. Emissions of Greenhouse Gases.” Journal of World- Commoner, Barry. 1971. The Closing Circle. New York: Systems Research 3(3):432–66. Alfred A. Knopf Publishers. Buttel, Frederick H. 2000. “World Society, the Nation- Condliffe, Jamie. 2017. “The Department of Defense State, and Environmental Protection: Comment on Wants to Double Down on Renewables.” MIT Tech- Frank, Hironaka, and Schofer.” American Sociologi- nology Review (https://www.technologyreview.com/ cal Review 65(1):117–21. s/603778/the-department-of-defense-wants-to-dou Cane, Mark A., Edward Miguel, Marshall Burke, Solo- ble-down-on-renewables/). mon M. Hsiang, David B. Lobell, Kyle C. Meng, Cooley, Alexander, Daniel Nexon, and Steven Ward. and Shanker Satyanath. 2014. “Temperature and Vio- 2019. “Revising Order or Challenging the Balance lence.” Nature Climate Change 4:234–35 (https://doi of Military Power? An Alternative Typology of Revi- .org/10.1038/nclimate2171). sionist and Status-Quo States.” Review of Interna- Caniglia, Elizabeth, Andrew Jorgenson, Stephanie Malin, tional Studies 45(4):689–708. Lori Peek, David Pellow, and Xiaorui Huang, eds. Correa, Jennifer G., and Joseph M. Simpson. 2022. 2021. Handbook of Environmental Sociology. Dor- “Building Walls, Destroying Borderlands: Reper- drecht, Netherlands: Springer. toires of Militarization on the United States-Mexico Carlton-Ford, Steve. 2010. “Major Armed Conflicts, Mil- Border.” Nature and Culture 17(1):1–25. itarization, and Life Chances: A Pooled Time-Series Custers, Peter. 2010. “Military Keynesianism Today: An Analysis.” Armed Forces & Society 36(5):864–89. Innovative Discourse.” Race & Class 51(4):79–94. Carlton-Ford, Steve, Katherine Durante, David Stearns, Cypher, James. 2015. “The Origins and Evolution of Mil- and Ciera Graham. 2019. “Guns and Butter: Child itary Keynesianism in the United States.” Journal of Mortality and the Mediators of Militarization.” Post Keynesian Economics 38(3):449–76. Armed Forces & Society 45(1):177–97. Cypher, James. 2022. “The Political Economy of Systemic Center for Systemic Peace and Societal-Systems U.S. Militarism.” Monthly Review 73(11):23–37. Research. 2018. “POLITY5 Dataset Version 2018” Davidson, Debra J. 2019. “Exnovating for a Renewable (http://www.systemicpeace.org). Energy Transition.” Nature Energy 4:254–56. Chase-Dunn, Christopher. 1998. Global Formation: Davidson, Debra J. 2022. “Climate Change Sociology: Structures of the World Economy. Lanham, MD: Past Contributions and Future Research Needs.” Rowman & Littlefield Publishers. PLOS Climate 1(7):e0000055 (https://doi.org/ Chase-Dunn, Christopher, Yukio Kawano, and Benja- 10.1371/journal.pclm.0000055). min Brewer. 2000. “Trade Globalization since 1795: Davidson, Debra J., and Jeffrey Andrews. 2013. “Not All Waves of Integration in the World-System.” Ameri- About Consumption.” Science 339(6125):1286–87. can Sociological Review 65(1):77–95. Deb, Nikhel. 2021. “Slow Violence and the Gas Peedit Clark, Brett, Daniel Auerbach, and Stefano B. Longo. in Neoliberal India.” Social Problems (https://doi.org/ 2018. “The Bottom Line: Capital’s Production of 10.1093/socpro/spab058). Social Inequalities and Environmental Degrada- Dietz, Thomas. 2015. “Prolegomenon to a Structural tion.” Journal of Environmental Studies and Sciences Human Ecology of Human Well-Being.” Sociology 8:562–69. of Development 1(1):123–48. Jorgenson et al. 29 Dietz, Thomas. 2017. “Drivers of Human Stress on the Foster, John Bellamy, and Robert McChesney. 2014. Environment in the Twenty-First Century.” Annual “Surveillance Capitalism: Monopoly-Finance Capi- Review of Environment and Resources 42:189–213. tal, the Military-Industrial Complex, and the Digital Dietz, Thomas, Kenneth Frank, Cameron Whitley, Jen- Age.” Monthly Review 66(3):1–31. nifer Kelly, and Rachel Kelly. 2015. “Political Frank, David John. 1997. “Science, Nature, and the Glo- Influences on Greenhouse Gas Emissions from US balization of the Environment, 1870–1990.” Social States.” Proceedings of the National Academy of Sci- Forces 76(2):409–435. ences 112(27):8254–59. Frank, David John, Ann Hironaka, and Evan Schofer. Dietz, Thomas, and Eugene A. Rosa. 1997. “Effects 2000. “The Nation-State and the Natural Environ- of Population and Affluence on CO Emissions.” ment over the Twentieth Century.” American Socio- Proceedings of the National Academy of Sciences logical Review 65(1):96–116. 94(1):175–79. Frey, R. Scott. 2013. “Agent Orange and America at War Dietz, Thomas, Rachael Shwom, and Cameron Whitley. in Vietnam and Southeast Asia.” Human Ecology 2020. “Climate Change and Society.” Annual Review Review 20(1):1–10. of Sociology 46:135–58. Gareau, Brian, and Cristina Lucier. 2018. “Neoliberal Ditzen, Jan. 2018. “Estimating Dynamic Common- Restructuring of the World Polity: The Weakening Correlated Effects in Stata.” The Stata Journal of the Montreal Protocol and Basel Convention in 18(3):585–617. Historical Perspective.” Environmental Sociology Downey, Liam. 2015. Inequality, Democracy, and the 4(3):325–42. Environment. New York: NYU Press. Giddens, Anthony. 1987. The Nation-State and Violence. Driscoll, Daniel. 2021. “Drivers of Carbon Price Volume Two of a Contemporary Critique of Histori- Adoption in Wealthy Democracies: International cal Materialism. Berkeley: University of California or Domestic Forces?” Socius 7 (https://doi.org/ Press. 10.1177/2378023121992252). Giddens, Anthony. 2011. The Politics of Climate Change. Dunlap, Alexander, and Andrea Brock, eds. 2022. Enforc- Cambridge, UK: Polity Press. ing Ecocide: Power, Policing, and Planetary Militari- Giesselmann, Marco, and Alexander Schmidt-Catran. zation. Dordrecht, Netherlands: Springer. 2022. “Interactions in Fixed Effects Regression Mod- Dunlap, Riley E., and Robert J. Brulle, eds. 2015. Cli- els.” Sociological Methods & Research 51(3):1100– mate Change and Society: Sociological Perspectives. 27. New York: Oxford University Press. Givens, Jennifer E. 2014. “Global Climate Change Nego- Durant, Robert. 2007. The Greening of the US Military: tiations, the Treadmill of Destruction, and World Soci- Environmental Policy, National Security, and Orga- ety.” International Journal of Sociology 44(2):7–36. nizational Change. Washington, DC: Georgetown Givens, Jennifer E. 2018. “Ecologically Unequal University Press. Exchange and the Carbon Intensity of Well-Being, Elliott, James R., and Scott Frickel. 2015. “Urbaniza- 1990–2011.” Environmental Sociology 4(3):311–24. tion as Socioenvironmental Succession: The Case of Givens, Jennifer E., Xiaorui Huang, and Andrew K. Jor- Hazardous Industrial Site Accumulation.” American genson. 2019. “Ecologically Unequal Exchange: A Journal of Sociology 120(6):1736–77. Theory of Global Environmental Injustice.” Sociol- Falkner, Robert. 2021. Environmentalism and Global ogy Compass 13(5):e12693. International Society. Cambridge, UK: Cambridge Givens, Jennifer E., and Andrew K. Jorgenson. 2013. University Press. “Individual Environmental Concern in the World Pol- Fisher, Dana R. 2022. “AnthroShift in a Warming ity: A Multilevel Analysis.” Social Science Research World.” Climate Action 1:9 (https://doi.org/10.1007/ 42(2):418–31. s44168-022-00011-8). Gould, Kenneth A. 2007. “The Ecological Costs of Mili- Fisher, Dana R., and William R. Freudenburg. 2004. tarization.” Peace Review 19(4):331–34. “Post Industrialization and Environmental Quality: Gould, Kenneth A., David N. Pellow, and Allan Schnai- An Empirical Analysis of the Environmental State.” berg. 2008. The Treadmill of Production: Injustice Social Forces 83(1):157–88. and Unsustainability in the Global Economy. Boul- Fisher, Dana R., and Andrew K. Jorgenson. 2019. “End- der, CO: Paradigm Publishers. ing the Stalemate: Toward a Theory of Anthro-Shift.” Grant, Don, Albert Bergesen, and Andrew Jones. 2002. Sociological Theory 37(4):342–62. “Organizational Size and Pollution: The Case of the Foster, John Bellamy. 1999. “Marx’s Theory of Metabolic U.S. Chemical Industry.” American Sociological Rift.” American Journal of Sociology 105(2):366–405. Review 67(3):389–407. Foster, John Bellamy, and Brett Clark. 2020. The Rob- Grant, Don, Andrew Jorgenson, and Wesley Longhofer. bery of Nature. New York: Monthly Review Press. 2020. Super Polluters: Tackling the World’s Largest Foster, John Bellamy, and Hannah Holleman. 2012. Sites of Climate-Disrupting Emissions. New York: “Weber and the Environment: Classical Foundations Columbia University Press. for a Postexemptionalist Sociology.” American Jour- Grant, Don, and Ion Bogdan Vasi. 2017. “Civil Society in nal of Sociology 117(6):1625–73. an Age of Environmental Accountability: How Local 30 American Sociological Review 00(0) Environmental Nongovernmental Organizations Hsiao, Cheng, M. Hashem Pesaran, and A. Kamil Tah- Reduce U.S. Power Plants’ Carbon Dioxide Emis- miscioglu. 2002. “Maximum Likelihood Estimation sions.” Sociological Forum 32(1):94–115. of Fixed Effects Dynamic Panel Data Models Cov- Greiner, Patrick T. 2022. “Colonial Contexts and the Fea- ering Short Time Periods.” Journal of Econometrics sibility of Mitigation through Transition: A Study of 109(1):107–50. the Impact of Historical Processes on the Emissions Huang, Xiaorui. 2018. “Ecologically Unequal Exchange, Dynamics of Nation-States.” Global Environmental Recessions, and Climate Change: A Longitudinal Change 77:102609 (https://doi.org/10.1016/j.gloen Study.” Social Science Research 73:1–12. vcha.2022.102609). Huang, Xiaorui, and Andrew Jorgenson. 2018. “The Griffin, Larry, Joel Devine, and Michael Wallace. 1982. Asymmetrical Effects of Economic Development on “Monopoly Capital, Organized Labor, and Military Consumption-Based and Production-Based Carbon Expenditures in the United States, 1949–1976.” Amer- Dioxide Emissions, 1990–2014.” Socius 4 (http://doi ican Journal of Sociology 88(Supplement):S113– .org/10.1177/2378023118773626). S153. Huber, Joseph. 2010. “Upstreaming Environmental Gunderson, Ryan, Diana Stuart, and Brian Petersen. Action.” Pp. 334–55 in The Ecological Modernisa- 2018. “Ideological Obstacles to Effective Climate tion Reader, edited by A. Mol, G. Spaargaren, and D. Policy: The Greening of Markets, Technology, and Sonnenfeld. London, UK: Routledge. Growth.” Capital & Class 42(1):133–60. Hyde, Allen, and Todd Vachon. 2019. “Running With Haberl, Helmut, Dominik Wiedenhofer, Doris Virág, or Against the Treadmill? Labor Unions, Institu- Gerald Kalt, Barbara Plank, Paul Brockway, Tomer tional Contexts, and Greenhouse Gas Emissions in a Fishman, et al. 2020. “A Systematic Review of the Comparative Perspective.” Environmental Sociology Evidence on Decoupling of GDP, Resource Use and 5(3):269–82. GHG Emissions, Part II: Synthesizing the Insights.” Inglehart, Ronald, and Wayne Baker. 2000. “Modern- Environmental Research Letters 15:065003 (http:// ization, Cultural Change, and the Persistence of doi.org/10.1088/1748-9326/ab842a). Traditional Values.” American Sociological Review Hironaka, Ann. 2014. Greening the Globe: World Society 65(1):19–51. and Environmental Change. New York: Cambridge IPCC (Intergovernmental Panel on Climate Change). University Press. 2007. Climate Change 2007: Synthesis Report. Hirst, Paul. 2001. War and Power in the Twenty-First Geneva, Switzerland: IPCC. Century: The State, Military Power and the Interna- IPCC (Intergovernmental Panel on Climate Change). tional System. Cambridge, UK: Polity Press. 2013. Climate Change 2013: The Physical Science Hooks, Gregory. 1990. “The Rise of the Pentagon and US Basis. Cambridge, UK: Cambridge University Press. State Building: The Defense Program as Industrial Pol- IPCC (Intergovernmental Panel on Climate Change). icy.” American Journal of Sociology 96(2):358–404. 2022. Climate Change 2022: Mitigation of Climate Hooks, Gregory. 1994. “Regional Processes in the Hege- Change. Contribution of Working Group III to the monic Nation: Political, Economic, and Military Sixth Assessment Report of the Intergovernmental Influences on the Use of Geographic Space.” Ameri- Panel on Climate Change. Cambridge, UK: Cam- can Sociological Review 59(5):746–72. bridge University Press. Hooks, Gregory, and Leonard Bloomquist. 1992. “The Isiksal, Aliya Zhakanova. 2021. “Testing the Effect of Legacy of World War II for Regional Growth and Sustainable Energy and Military Expenses on Envi- Decline: The Cumulative Effects of Wartime Invest- ronmental Degradation: Evidence from the States ments on US Manufacturing, 1947–1972.” Social with the Highest Military Expenses.” Environmental Forces 71(2):303–337. Science and Pollution Research 28:20487–89. Hooks, Gregory, Michael R. Lengefeld, and Chad Smith. Jaccard, James, and Robert Turrisi. 2003. Interaction 2021. “Recasting the Treadmills of Production and Effects in Multiple Regression, 2nd ed. Thousand Destruction: New Theoretical Directions.” Sociology Oaks, CA: Sage. of Development 7(1):52–76. Jayaram, Dhanasree, and Marie Claire Brisbois. 2021. Hooks, Gregory, and Chad Smith. 2004. “The Tread- “Aiding or Undermining? The Military as an Emer- mill of Destruction: National Sacrifice Areas and gent Actor in Global Climate Governance.” Earth Sys- Native Americans.” American Sociological Review tem Governance 9:100107 (https://doi.org/10.1016/ 69(4):558–75. j.esg.2021.100107). Hooks, Gregory, and Chad Smith. 2005. “Treadmills of Johnson, Chalmers. 2004. The Sorrows of Empire. New Production and Destruction: Threats to the Environ- York: Metropolitan Books. ment Posed by Militarism.” Organization & Environ- Jorgenson, Andrew K. 2014. “Economic Development ment 18(1):19–37. and the Carbon Intensity of Human Well-Being.” Hsiang, Solomon M., Marchall Burke, and Edward Nature Climate Change 4:186–89. Miguel. 2013. “Quantifying the Influence of Climate Jorgenson, Andrew K., and Brett Clark. 2009. “The on Human Conflict.” Science 341(6151):1235367 Economy, Military, and Ecologically Unequal Rela- (https://doi.org/10.1126/science.1235367). tionships in Comparative Perspective: A Panel Study Jorgenson et al. 31 of the Ecological Footprints of Nations, 1975–2000.” Kick, Edward L., Byron L. Davis, David M. Kiefer, and Social Problems 56(4):621–46. Thomas J. Burns. 1998. “A Cross-National Analy- Jorgenson, Andrew K., and Brett Clark. 2012. “Are the sis of Militarization and Well-Being Relationships Economy and the Environment Decoupling? A Com- in Developing Countries.” Social Science Research parative International Study, 1960–2005.” American 27(4):351–70. Journal of Sociology 118(1):1–44. Klare, Michael. 2019. All Hell Breaking Loose: The Pen- Jorgenson, Andrew K., and Brett Clark. 2016. “The Tem- tagon’s Perspective on Climate Change. New York: poral Stability and Developmental Differences in the Metropolitan Books. Environmental Impacts of Militarism: The Tread- Kleykamp, Meredith. 2007. “Military Service as a mill of Destruction and Consumption-Based Carbon Labor Market Outcome.” Race, Gender & Class Emissions.” Sustainability Science 11:505–514. 14(3/4):65–76. Jorgenson, Andrew K., Brett Clark, and Jeffrey Kentor. Klinenberg, Eric, Malcolm Araos, and Liz Koslow. 2020. 2010. “Militarization and the Environment: A Panel “Sociology and the Climate Crisis.” Annual Review Study of Carbon Dioxide Emissions and the Ecologi- of Sociology 46:649–69. cal Footprints of Nations, 1970–2000.” Global Envi- Knight, Kyle, and Juliet Schor. 2014. “Economic Growth ronmental Politics 10(1):7–29. and Climate Change: A Cross-National Analysis of Jorgenson, Andrew K., Shirley Fiske, Klaus Hubacek, Jia Territorial and Consumption-Based Carbon Emis- Li, Tom McGovern, Torben Rick, Juliet B. Schor, Wil- sions in High-Income Countries.” Sustainability liam Solecki, Richard York, and Ariela Zycherman. 6(6):3722–31. 2019. “Social Science Perspectives on Drivers of and Krahmann, Elke. 2010. States, Citizens and the Privati- Responses to Global Climate Change.” WIREs Climate sation of Security. New York: Cambridge University Change 10(1):e554 (https://doi.org/10.1002/wcc.554). Press. Kelly, Orla. 2020. “The Silver Bullet? Assessing the Kramer, Ronald. 2020. Carbon Criminals, Climate Role of Education for Sustainability.” Social Forces Crimes. New Brunswick, NJ: Rutgers University 99(1):178–204. Press. Kelly, Orla, Ryan P. Thombs, and Andrew K. Jorgenson. Kraska, Peter. 2007. “Militarization and Policing – Its 2021. “The Unsustainable State: Greenhouse Gas Relevance to 21st Century Police.” Policing: A Jour- Emissions, Inequality, and Human Well-Being in the nal of Policy and Practice 1(4):501–513. United States, 1913 to 2017.” Socius 7 (http://doi.org/ Kripfganz, Sebastian, and Vasilis Sarafidis. 2021. 10.1177/23780231211020536). “Instrumental Variable Estimation of Large Panel Kennedy, Emily Huddart, and Jennifer E. Givens. 2019. Data Models with Common Factors.” The Stata Jour- “Eco-habitus or Eco-powerlessness? Examining nal 21(3):659–86. Environmental Concern across Social Class.” Socio- Lawrence, Michael, Holly Stemberger, Aaron Zolderdo, logical Perspectives 62(5):646–67. Daniel Struthers, and Steven Cooke. 2015. “The Kentor, Jeffrey. 2000. Capital and Coercion: The Eco- Effects of Modern War and Military Activities on nomic and Military Processes That Have Shaped the Biodiversity and the Environment.” Environmental World Economy 1800–1990. New York: Garland Pub- Reviews 23(4):443–60. lishing. Lengefeld, Michael R., Gregory Hooks, and Chad Smith. Kentor, Jeffrey, Rob Clark, and Andrew K. Jorgenson. 2021. “War and the Environment.” In Handbook of 2023. “The Hidden Cost of Global Economic Inte- Environmental Sociology, edited by B. Caniglia, gration: How Foreign Investment Drives Military A. Jorgenson, S. Malin, L. Peek, D. Pellow, and X. Expenditures.” World Development 161:106105 Huang. Dordrecht, Netherlands: Springer. (https://doi.org/10.1016/j.worlddev.2022.106105). Lengefeld, Michael R., and Chad Smith. 2013. “Nuclear Kentor, Jeffrey, and Andrew K. Jorgenson. 2017. “Mili- Shadows: Weighing the Environmental Effects tary Expenditures and Health: A Cross-National of Militarism, Capitalism, and Modernization in Study, 1975–2000.” International Journal of Sociol- a Global Context, 2001–2007.” Human Ecology ogy and Social Policy 37(13/14):755–72. Review 20(1):11–25. Kentor, Jeffrey, Andrew K. Jorgenson, and Edward Kick. Levine, Paul, Somnath Sen, and Ron Smith. 1994. “A 2012. “The ‘New’ Military and Income Inequality: A Model of the International Arms Market.” Defence Cross National Analysis.” Social Science Research and Peace Economics 5:1–18. 41(3):514–26. Levy, Barry, and Victor Sidel. 2007. War and Public Kentor, Jeffrey, and Edward Kick. 2008. “Bringing the Health, 2nd ed. New York: Oxford University Press. Military Back In: Military Expenditures and Eco- Light, Sarah. 2014. “The Military-Environmental Com- nomic Growth 1990–2003.” Journal of World Sys- plex.” Boston College Law Review 55(3):879–946. tems Research 14(2):142–72. Lim, Alwyn, and Kiyoteru Tsutsui. 2012. “Globaliza- Keyßer, Lorenz, and Manfred Lenzen. 2021. “1.5°C tion and Commitment in Corporate Social Respon- Degrowth Scenarios Suggest the Need for New Miti- sibility: Cross-National Analyses of Institutional and gation Pathways.” Nature Communications 12:2676 Political-Economy Effects.” American Sociological (https://doi.org/10.1038/s41467-021-22884-9). Review 77(1):69–98. 32 American Sociological Review 00(0) Lockie, Stewart. 2022. “Mainstreaming Climate Change McChesney, Robert. 2013. Digital Disconnect. New Sociology.” Environmental Sociology 8(1):1–6. York: New Press. Longhofer, Wesley, and Andrew K. Jorgenson. 2017. McGee, Julius Alexander, and Patrick Trent Greiner. “Decoupling Reconsidered: Does World Society 2018. “Can Reducing Income Inequality Decouple Integration Influence the Relationship between the Economic Growth from CO Emissions?” Socius 4 Environment and Economic Development?” Social (http://doi.org/10.1177/2378023118772716). Science Research 65:17–29. McNeill, William H. 1982. The Pursuit of Power. Chi- Longhofer, Wesley, and Evan Schofer. 2010. “National cago: The University of Chicago Press. and Global Origins of Environmental Association.” McQuaig, Linda, and Neil Brooks. 2012. Billionaire’s American Sociological Review 75(4):505–533. Ball. Boston: Beacon Press. Longo, Stefano, Ellinor Isgren, Brett Clark, Andrew K. Mejia, Steven. 2021. “The Climate Crisis and Export Jorgenson, Anne Jerneck, Lennart Olsson, Orla Kelly, Intensity: A Comparative International Study of David Harnesk, and Richard York. 2021. “Sociology Greenhouse Gas Emissions in the Global South, 1990– for Sustainability Science.” Discover Sustainability 2014.” International Journal of Sociology 51(1):1–22. 2:47 (http://doi.org/10.1007/s43621-021-00056-5). Mezey, Nancy J. 2020. “Presidential Address: Start Looney, Robert. 1990. “Military Expenditures, Military Spreading the News: Illuminating the Effects of Cli- Participation Rates, and Human Capital Development mate Change as a Social Problem.” Social Problems in the Arab World.” Current Research on Peace and 67(4):605–615. Violence 13(1):39–53. Mills, C. Wright. 1956. The Power Elite. New York: Lynch, Michael, Michael Long, Paul Stretesky, and Kim- Oxford University Press. berly Barrett. 2017. Green Criminology: Crime, Jus- Mitchell, Jon. 2020. Poisoning the Pacific. Lanham, MD: tice, and the Environment. Berkeley: University of Rowman & Littlefield. California Press. Mitchell, Ronald B., and Richard York. 2020. “Reduc- Mach, Katharine, Caroline Kraan, W. Neil Adger, ing the Web’s Carbon Footprint: Does Improved Halvard Buhaug, Marshall Burke, James Fearon, Electrical Efficiency Reduce Webserver Electricity Christopher Field, et al. 2019. “Climate as a Risk Fac- Use?” Energy Research & Social Science 65:101474 tor for Armed Conflict.” Nature 571:193–97. (https://doi.org/10.1016/j.erss.2020.101474). Machlis, Gary, and Thor Hanson. 2008. “Warfare Ecol- Mol, Arthur P. J. 2003. Globalization and Environmental ogy.” BioScience 58(8):729–36. Reform: The Ecological Modernization of the Global Magdoff, Harry. 1978. Imperialism. New York: Monthly Economy. Cambridge, MA: MIT Press. Review Press. Mumford, Lewis. 1963. Technics and Civilization. New Mahutga, Matthew C. 2006. “The Persistence of York: Harcourt, Brace & World, Incorporated. Structural Inequality? A Network Analysis of Newman, Nathan. 2002. Net Loss: Internet Prophets, Pri- International Trade, 1965–2000.” Social Forces vate Profits, and the Costs to Community. University 84(4):1863–89. Park: Pennsylvania State University Press. Mahutga, Matthew C., and David A. Smith. 2011. “Glo- Nickell, Stephen. 1981. “Biases in Dynamic Models with balization, the Structure of the World Economy and Fixed Effects.” Econometrica 49(6):1417–26. Economic Development.” Social Science Research Norgaard, Kari Marie. 2018. “The Sociological Imagina- 40(1):257–72. tion in a Time of Climate Change.” Global and Plan- Malin, Stephanie, Tara Opsal, Tara O’Connor Shelley, etary Change 163:171–76. and Peter Hall. 2019. “The Right to Resist or a Case Norkutė, Milda, Vasilis Sarafidis, Takashi Yamagata, and of Injustice? Meta-Power in the Oil and Gas Fields.” Guowei Cui. 2021. “Instrumental Variable Estimation Social Forces 97(4):1811–38. of Dynamic Linear Panel Data Models with Defac- Mann, Michael. 2012. The Sources of Social Power, Vol. tored Regressors and a Multifactor Error Structure.” 3, Global Empires and Revolution, 1890–1945. New Journal of Econometrics 220(2):416–46. York: Cambridge University Press. Nowak, Peter. 2011. Sex, Bombs, and Burgers: How War, Mann, Michael. 2014. The Sources of Social Power, Vol. Pornography, and Fast Food Have Shaped Modern 4, Globalizations, 1945–2011. New York: Cambridge Technology. Guilford, CT: Lyons Press. University Press. O’Rourke, Ronald. 2021. Navy Large Unmanned Surface Marquart-Pyatt, Sandra. 2012. “Contextual Influences and Undersea Vehicles. Washington, DC: Congres- on Environmental Concerns Cross-Nationally: A sional Research Service. Multilevel Investigation.” Social Science Research Pathak, Swapna. 2020. “Ecological Footprints of War: An 41(5):1085–99. Exploratory Assessment of the Long-Term Impact of Marzec, Robert. 2016. Militarizing the Environment: Violent Conflicts on National Biocapacity from Climate Change and the Security State. Minneapolis: 1962–2009.” Journal of Environmental Studies and University of Minnesota Press. Sciences 10:380–93. Mazur, Allan. 2013. Energy and Electricity in Industrial Pellow, David N. 2007. Resisting Global Toxics: Trans- Nations: The Sociology and Technology of Energy. national Movements for Environmental Justice. Cam- Oxford, UK: Earthscan. bridge, MA: MIT Press. Jorgenson et al. 33 Pellow, David N., and Hollie Nyseth Brehm. 2013. “An Samaras, Constantine, William J. Nuttall, and Morgan Environmental Sociology for the Twenty-First Cen- Bazilian. 2019. “Energy and the Military: Conver- tury.” Annual Review of Sociology 39:229–50. gence of Security, Economic, and Environmen- Pesaran, M. Hashem. 2006. “Estimation and Inference in tal Decision-Making.” Energy Strategy Reviews Large Heterogeneous Panels with a Multifactor Error 26:100409 (https://doi.org/10.1016/j.esr.2019.100 Structure.” Econometrica 74(4):967–1012. 409). Pesaran, M. Hashem, and Ron Smith. 1995. “Estimating Sanders, Barry. 2009. The Green Zone: The Environmen- Long-Run Relationships from Dynamic Heteroge- tal Costs of Militarism. Oakland, CA: AK Press. neous Panels.” Journal of Econometrics 68(1):79– Sanderson, Matthew, and Vivian Hughes. 2019. “Race to 113. the Bottom (of the Well): Groundwater in an Agri- Pickup, Mark. 2015. Introduction to Time Series Analy- cultural Production Treadmill.” Social Problems sis. Thousand Oaks, CA: Sage. 66(3):392–410. Podobnik, Bruce. 2006. Global Energy Shifts: Foster- Scanlan, Stephen, and Craig Jenkins. 2001. “Military ing Sustainability in a Turbulent Age. Philadelphia: Power and Food Security: A Cross-National Analysis Temple University Press. of Less-Developed Countries, 1970–1990.” Interna- Raleigh, Clionadh, and Henrik Urdal. 2007. “Climate tional Studies Quarterly 45(1):159–87. Change, Environmental Degradation and Armed Schnaiberg, Allan. 1980. The Environment. New York: Conflict.” Political Geography 26(6):674–94. Oxford University Press. Renner, Michael. 1991. “Assessing the Military’s War on Schofer, Evan. 2003. “The Global Institutionalization of the Environment.” Pp. 132–52 in State of the World, Geological Science, 1800 to 1990.” American Socio- edited by L. Starke. New York: W.W. Norton & Com- logical Review 68(5):730–59. pany. Schofer, Evan, and Ann Hironaka. 2005. “The Effects Rice, James. 2007. “Ecological Unequal Exchange: of World Society on Environmental Protection Out- International Trade and Uneven Utilization of Envi- comes.” Social Forces 84(1):25–47. ronmental Space in the World System.” Social Forces Scott, Lauren, and Erik Johnson. 2017. “From Fringe to 85(3):1369–92. Core? The Integration of Environmental Sociology.” Rice, James. 2015. “Downwind of the Atomic State: Environmental Sociology 3(1):17–29. US Continental Atmospheric Testing, Radioactive Shandra, John M., Bruce London, Owen P. Whooley, Fallout, and Organizational Deviance, 1951–1962.” and John B. Williamson. 2004. “International Non- Social Science History 39(4):647–76. governmental Organizations and Carbon Dioxide Rieger, Annika. 2021. “Does ICT Result in Dematerial- Emissions in the Developing World: A Quantita- ization? The Case of Europe, 2005–2017.” Environ- tive, Cross-National Analysis.” Sociological Inquiry mental Sociology 7(1):64–75. 74(4):520–45. Roberts, J. Timmons, Peter E. Grimes, and Jodie L. Sharkey, Amanda J., and Patricia Bromley. 2015. “Can Manale. 2003. “Social Roots of Global Environ- Ratings Have Indirect Effects? Evidence from the mental Change: A World-Systems Analysis of Car- Organizational Response to Peers’ Environmental bon Dioxide Emissions.” Journal of World-Systems Ratings.” American Sociological Review 80(1):63– Research 9(2):277–315. 91. Rosa, Eugene A., and Thomas Dietz. 2012. “Human Shaw, Martin. 1988. Dialectics of War. London, UK: Drivers of National Greenhouse-Gas Emissions.” Pluto Press. Nature Climate Change 2:581–86. Shaw, Martin. 2002. “Risk-Transfer Militarism: Small Rosa, Eugene A., Richard York, and Thomas Dietz. 2004. Massacres and the Historic Legitimacy of War.” “Tracking the Anthropogenic Drivers of Ecological International Relations 16(3):343–60. Impacts.” Ambio 33(8):509–512. Shaw, Martin. 2005. The New Western Way of War: Risk- Royal Society and U.S. National Academy of Sciences. Transfer War and Its Crisis in Iraq. Cambridge, UK: 2020. Climate Change: Evidence and Causes. Wash- Polity Press. ington, DC: National Academics Press. Shorette, Kristen. 2012. “Outcomes of Global Environ- Rudel, Thomas K. 2009. “How Do People Transform mentalism: Longitudinal and Cross-National Trends Landscapes? A Sociological Perspective on Suburban in Chemical Fertilizer and Pesticide Use.” Social Sprawl and Tropical Deforestation.” American Jour- Forces 91(1):299–325. nal of Sociology 115(1):129–54. Shwom, Rachael. 2011. “A Middle Range Theorization Rudel, Thomas K., J. Timmons Roberts, and JoAnn Car- of Energy Politics: The Struggle for Energy Efficient min. 2011. “Political Economy of the Environment.” Appliances.” Environmental Politics 20(5):705–26. Annual Review of Sociology 37:221–38. Siebold, Guy. 2001. “Core Issues and Theory in Military Running, Katrina. 2013. “World Citizenship and Con- Sociology.” Journal of Political and Military Sociol- cern for Global Warming: Building the Case for a ogy 29(1):140–59. Strong International Civil Society.” Social Forces Sills, Peter. 2014. Toxic War: The Story of Agent Orange. 92(1):377–99. Nashville, TN: Vanderbilt University Press. 34 American Sociological Review 00(0) Simpson, Joseph M., Riley E. Dunlap, and Andrew S. tial Vulnerability to Climate Change: A Social Sci- Fullerton. 2019. “The Treadmill of Information: ence Review.” WIREs Climate Change 10(2):e565 Development of the Information Society and Car- (https://doi.org/10.1002/wcc.565). bon Dioxide Emissions.” Sociology of Development Thombs, Ryan P. 2017. “The Paradoxical Relationship 5(4):381–409. between Renewable Energy and Economic Growth: Singer, Peter. 2008. Corporate Warriors: The Rise of A Cross-National Panel Study, 1990–2013.” Journal the Privatized Military Industry. Ithaca, NY: Cornell of World-Systems Research 23(2):540–64. University Press. Thombs, Ryan P. 2018a. “The Transnational Tilt of the SIPRI (Stockholm International Peace Research Insti- Treadmill and the Role of Trade Openness on Car- tute). 2022. “Military Expenditure Database” (https:// bon Emissions: A Comparative International Study, www.sipri.org/databases/milex). 1965–2010.” Sociological Forum 33(2):422–42. Smart, Barry. 2016. “Military-Industrial Complexities, Thombs, Ryan P. 2018b. “Has the Relationship between University Research and Neoliberal Economy.” Jour- Non-Fossil Fuel Energy Sources and CO Emissions nal of Sociology 52(3):455–81. Changed Over Time? A Cross-National Study, 2000– Smith, Chad, and Michael R. Lengefeld. 2020. “The 2013.” Climatic Change 148(4):481–90. Environmental Consequences of Asymmetric War: Thombs, Ryan P. 2021. “In-and-Beyond State Power: A Panel Study of Militarism and Carbon Emissions, How Political Equality Moderates the Economic 2000–2010.” Armed Forces & Society 46(2):214–37. Growth-CO Emissions Relationship, 1990–2014.” Smith, David A. 2017. “Globalizing Social Problems: An Sociological Quarterly 62(3):528–47. Agenda for the Twenty-First Century.” Social Prob- Thombs, Ryan P. 2022. “A Guide to Analyzing Large lems 64(1):1–13. N, Large T Panel Data.” Socius 8 (http://doi.org/ Smith, Kurt. 2020. “Environmental Protection, the Mili- 10.1177/23780231221117645). tary, and Preserving the Balance: Why It Matters, in Thombs, Ryan P., and Xiaorui Huang. 2019. “Uneven War and Peace.” Seattle Journal of Technology, Envi- Decoupling: The Economic Growth–CO Emissions ronmental & Innovation Law 11(1): Article 5. Relationship in the Global North, 1870 to 2014.” Soener, Matthew. 2019. “Profiting in a Warming World: Sociology of Development 5(4):410–27. Investigating the Link between Greenhouse Gas Thombs, Ryan P., Xiaorui Huang, and Jared B. Fitzgerald. Emissions and Capitalist Profitability in OECD 2022. “What Goes Up Might Not Come Down: Mod- States.” Sociological Forum 34(4):974–98. eling Directional Asymmetry with Large-N, Large-T Soeters, Joseph. 2018. Sociology and Military Studies: Data.” Sociological Methodology 52(1):1–29. Classical and Current Foundations. New York: Rout- Tilly, Charles. 1990. Coercion, Capital, and European ledge Press. States, AD 990–1992. Oxford, UK: Blackwell. Sohag, Kazi, Shaiara Husain, Shawkat Hammoudeh, and Turse, Nick. 2008. The Complex: How the Military Normah Omar. 2021. “Innovation, Militarization, and Invades our Everyday Lives. New York: Henry Holt Renewable Energy and Green Growth.” Environ- and Company. mental Science and Pollution Research 28:36004–17 Turse, Nick. 2015. Tomorrow’s Battlefield: US Proxy (https://doi.org/10.1007/s11356-021-13326-6). Wars and Secret Ops in Africa. Chicago: Haymarket Spaargaren, Gert, Arthur P. J. Mol, and Frederick H. But- Books. tel. 2006. Governing Environmental Flows: Global U.S. Army. 1999. Staging, Onward Movement, and Inte- Challenges to Social Theory. Cambridge, MA: MIT gration Field Manuel. Washington, DC: Headquar- Press. ters, Department of Army. Staples, Steven. 2000. “The Relationship between Global- USDOD (United States Department of Defense). 2010. ization and Militarism.” Social Justice 27(4):18–22. Quadrennial Defense Review Report. Arlington, VA: Stuart, Diana. 2021. What Is Environmental Sociology? United States Department of Defense. Cambridge, UK: Polity Press. USDOD (United States Department of Defense). 2020. Swed, Ori, and Thomas Crosbie, eds. 2019. The Sociol- Department of Defense Annual Energy Management ogy of Privatized Security. London, UK: Palgrave and Resilience Report (AEMRR) Fiscal Year 2019. Macmillan. Washington, DC: Office of the Assistant Secretary of Thayer, George. 1969. The War Business: The Interna- Defense for Sustainment. tional Trade in Armaments. New York: Simon and Vandenbergh, Michael, and Jonathan Gilligan. 2017. Schuster. Beyond Politics: The Private Governance Response Thee, Marek. 1990. “Science-Based Military Technol- to Climate Change. Cambridge, UK: Cambridge Uni- ogy as a Driving Force behind the Arms Race.” Pp. versity Press. 105–20 in Arms Races: Technological and Political Vasi, Ion Bogdan, Edward Walker, John Johnson, and Dynamics, edited by N. Petter Gleditsch and O. Njøl- Hui Fen Tan. 2015. “No Fracking Way! Documen- stad. London, UK: Sage. tary Film, Discursive Opportunity, and Local Oppo- Thomas, Kimberley, R. Dean Hardy, Heather Lazrus, sition against Hydraulic Fracturing in the United Michael Mendez, Ben Orlove, Isabel Rivera-Collazo, States, 2010 to 2013.” American Sociological Review Timmons Roberts, et al. 2019. “Explaining Differen- 80(5):934–59. Jorgenson et al. 35 Vesia, Danielle J., Matthew C. Mahutga, and Bonnie been published in outlets such as the American Journal of Khánh Hà Buì. 2021. “Flattening the Curve? The Sociology, Conservation Biology, Global Environmental Structure of the Natural Resource Exchange Network Change, Environmental Research Letters, Theory and and CO Emissions.” Social Networks (http://doi.org/ Society, and Social Problems. His most recent book, with 10.1016/j.socnet.2021.07.004). John Bellamy Foster, is The Robbery of Nature. Vine, David. 2015. Base Nation. New York: Metropolitan Books. Ryan P. Thombs is a PhD candidate in the Sociology Weede, Erich. 1983. “Military Participation Ratios, Department at Boston College. His research examines Human Capital Formation, and Economic Growth: the drivers of global environmental change, the drivers A Cross-National Analysis.” Journal of Political and of health disparities and population health outcomes, Military Sociology 11(1):11–19. and quantitative methodology, particularly related to Wilcox, Fred. 2011. Scorched Earth: Legacies of Chemi- panel data and time-series modeling. His work appears cal Warfare in Vietnam. New York: Seven Stories in journals such as Sociological Methodology, Journal Press. of Health and Social Behavior, Sociological Forum, Wilcox, William. 2007. The Modern Military and the The Sociological Quarterly, Sociology of Development, Environment: The Laws of Peace and War. Lanham, Global Environmental Change, Environmental Research MD: Government Institutes. Letters, Science of the Total Environment, Climatic Wills, Jocelyn. 2017. Tug of War: Surveillance Capital- Change, and Energy Research & Social Science. ism, Military Contracting, and the Rise of the Secu- rity State. Montreal, CA: McGill-Queen’s University Jeffrey Kentor is a Professor of Sociology at Wayne Press. State University and founding co-editor of Sociology of World Bank. 2022. “World Development Indicators” Development. Kentor’s research focuses on long-term, (https://databank.worldbank.org/source/world-devel macro-level social change, from a global political- opment-indicators). economy perspective. His work considers the fundamen- York, Richard. 2012. “Asymmetric Effects of Economic tal economic and coercive forces that shape relationships Growth and Decline on CO Emissions.” Nature Cli- among countries, and how these dynamics impact eco- mate Change 2:762–64. nomic development, inequality, health, and the environ- York, Richard, and Julius A. McGee. 2016. “Understand- ment. Kentor’s research is published in the American ing the Jevons Paradox.” Environmental Sociology Journal of Sociology, American Sociological Review, 2(1):77–87. Social Forces, Social Problems, World Development, York, Richard, and Julius A. McGee. 2017. “Does Energy Research & Social Science, Global Environmen- Renewable Energy Development Decouple Eco- tal Politics, and elsewhere. nomic Growth from CO Emissions?” Socius 3 (http://doi.org/10.1177/2378023116689098). York, Richard, Eugene A. Rosa, and Thomas Dietz. Jennifer E. Givens is an Associate Professor of Soci- 2003. “Footprints on the Earth: The Environmental ology at Utah State University. She is interested in Consequences of Modernity.” American Sociological relationships between the environment, development, Review 68(2):279–300. well-being, and various inequalities from the local to Zierler, David. 2011. The Invention of Ecocide. Athens: the global, especially in the context of climate change. University of Georgia Press. Jennifer has published in journals such as Environmental Sociology, Sociology of Development, Social Science Research, Society & Natural Resources, Social Indica- Andrew K. Jorgenson is a Professor of Sociology at tors Research, and Environmental Research Letters, University of British Columbia (UBC) and a Research and she has received funding for her interdisciplinary Fellow at Vilnius University. Prior to UBC, he was research from the U.S. National Science Foundation. faculty at Boston College. Much of his research focuses on the societal dimensions of the climate crisis. His work appears in such venues as American Journal Xiaorui Huang is an Assistant Professor of Sociology of Sociology, Nature Climate Change, Social Forces, at Drexel University. His research examines the climate Environmental Research Letters, Social Problems, and change and human well-being implications of economic Science of the Total Environment. He is a recipient of the development, international trade, renewable energy, Fred Buttel Distinguished Contribution Award from the and income inequality. He has also conducted work American Sociological Association’s Section on Envi- on quantitative methods and agricultural communities’ ronmental Sociology, and co-author of the recent book, engagement with natural resource and climate change Super Polluters. issues. His published research appears in such venues as Social Science Research, Climatic Change, Sociological Brett Clark is a Professor of Sociology at the University Forum, Sociological Methodology, Energy Research & of Utah. His research focuses primarily on the human Social Science, Ecological Economics, Socius, and Sci- dimensions of environmental change. His scholarship has ence of the Total Environment. 36 American Sociological Review 00(0) Hassan El Tinay is a PhD student in the Department of Matthew C. Mahutga is a political economist at the Sociology at Boston College. His research focuses on the University of California-Riverside. His research utilizes nexus between political economy and political ecology, theories of global and comparative political economy to including topics such as the human drivers of climate study inequality and stratification, economic develop- change and ecologically unequal exchange. ment, environmental degradation, economic organiza- tion, labor, labor markets, and political attitudes and preferences. Beyond that, he enjoys talking to people Daniel Auerbach is an Assistant Professor of Sociology about matters that are important to them, mentoring hard- at the University of Wyoming. His research examines working and ambitious graduate students, and promoting the political-economic and militaristic drivers of envi- the upward mobility of underrepresented students. ronmental change. Daniel has published in journals such as International Critical Thought, Environmental Sociol- ogy, Climatic Change, and Organization & Environment.

Journal

American Sociological ReviewSAGE

Published: Jun 1, 2023

Keywords: climate change; environmental sociology; political-economic sociology; development sociology; militarization

There are no references for this article.