Labor migration, externalities and ethics: Theorizing the meso-level determinants of HIV vulnerability

Abstract

This paper discusses labor migration as an example of how focusing on the meso-level highlights the social processes through which structural factors produce HIV risk. Situating that argument in relation to existing work on economic organization and HIV risk as well as research on labor migration and HIV vulnerabilities, the paper demonstrates how analyzing the processes through which labor migration creates vulnerability can shift attention away from the proximate behavioral determinants of HIV risk and towards the community and policy levels. Further, it presents the concepts of externalities and the ethics of consumption, which underline how both producers and consumers benefit from low-waged migrant labor, and thus are responsible for the externalization of HIV risk characteristic of supply chains that rely on migrant labor. These concepts point to strategies through which researchers and advocates could press the public and private sectors to improve the conditions in which migrants live and work, with implications for HIV as well as other health outcomes.

Medical anthropologists have written frequently about structural violence and HIV risk (Farmer, 2001; Parker, 2002; Susser, 2009). The concept speaks powerfully to how material and social inequality constrain individual agency, but it can sometimes seem vague, all-inclusive, or hard to operationalize. This article advances work on structural violence by proposing a focus on meso-level social processes that contribute to the differential allocation of HIV risk. The term ‘meso’ denotes institutions, ideologies, or social processes that are 1) neither at the micro, individual or interpersonal level (such as beliefs or characteristics of interactions) nor at the macro-social level (such as socially-structured inequalities of race, gender, sexuality); 2) characterized by a ‘sociologically-plausible’ (Auerbach, Parkhurst and Cáceres, 2013) or empirically-described causal relationship to a health-relevant practice, and 3) conceivably modifiable through sustained strategically-organized collective action.

This idea of the meso-level resembles Robert Merton’s notion of ‘theories of the middle range’, which he contrasts with “a total system of sociological theory, in which observations about every aspect of social behavior, organization, and change promptly find their preordained place” (2007: 448). It focuses attention on modifiable population-level determinants of health inequalities, stimulating consideration of the processes through which inequalities translate into embodied suffering. A focus on the meso-level can move us from broad contentions about the political economy of HIV risk to a more policy- and program-relevant understanding of HIV vulnerability. The meso-level is exemplified here through a discussion of labor migration, with HIV vulnerability analyzed as an externality of specific modes of labor organization. (Externalities, which refers to “the unmeasured effects on third parties who were not at all involved in the primary economic transaction between buyer and seller” [Roberts, Mensah and Weinstein, 2010: 1718] calls attention to costs or benefits not reflected in the price of a good.) HIV vulnerability is not inherently an externality inherent to labor migration; rather, it results from policies at multiple levels as well as actions taken (or not) by the private sector and civil society.

Others have noted the urgency of understanding how macro-level inequalities create health inequalities. Writing about HIV prevention for migrants, Organista et al. propose a focus on influences they classify variously as structural, environmental, or situational, which lie between “super-structural factors” (2012: 11) and individual behavior. Also writing about migrants, Hirsch and Vasquez (2012) suggest that multi-sectoral determinants of risk include policy domains (such as transportation) infrequently considered relevant to HIV. Kippax, Stephenson, Parker and Aggleton (2013), arguing for theorizing “the middle ground”, talk about the need to understand “the specificities of…malleable social relations” (1373). Advancing those ideas, I propose the meso-level as an overarching concept that encompasses structural, environmental, or situational factors, as well as the “social drivers” noted by Auerbach, Parkhurst and Cáceres (2011). The goal is not to add another conceptual tool to an already substantial literature, but rather to simplify, providing a heuristic to distinguish between important but not easily modifiable axes of inequality and the institutions that reproduce those inequalities and are concrete enough to be the targets of action.

This is not just a theoretical question. “Combination prevention” (Kurth, Celum, Baeten, Vermund and Wasserheit, 2011) (HIV prevention combining biomedical, behavioral, and structural approaches) requires effective community and policy-level interventions to complement behavioral and biomedical interventions. Proponents of structural approaches have consistently noted the challenges of translating analyses of inequality into effective interventions. Parker, Easton and Klein (2000), for example, characterize the bulk of work on structural determinants as “poorly operationalizable” (S29), and Auerbach, Parkhurst and Cáceres call for the identification of “sociologically plausible causal chains between distal structural factors and specific individual or group practices” (2011: S305). The meso-level is broadly relevant to population health: whether the problem is obesity or gun violence, working at the meso-level can correct the tendency towards a behaviorist approach while forestalling the ‘public health nihilist’ argument that the only way to address health inequalities is to erase the injustices that produce them.

My argument has several parts. First, I situate labor migration and HIV vulnerability in relation to the broader literature on economic organization and HIV risk. Next, I review work on HIV and migration, enumerating the social processes linking migration to vulnerability to infection. Third, I discuss HIV and labor migration in relation to externalities and questions about ethical consumption. Fourth, I discuss meso-level intervention strategies, including how framing migration-related HIV vulnerability in relation to the ethics of consumption could contribute to social mobilization to modify contextual factors that produce vulnerability.

ECONOMIC ORGANIZATION AND HIV VULNERABILITY

Seminal work on economic determinants of HIV has argued for political-economic analyses of the epidemic’s causes, course and consequences (DeCosas 1996; Altman 1999). Lurie, Hintzen and Lowe (1995), similarly, have described how globally-coordinated (or imposed) social and economic development policies ironically exacerbated the spread of HIV, and Lee and Zwi (1996) have made an even broader argument (echoed recently in the work of Rowden [2009]) about the limitation of neoliberalism as a framework for the response to AIDS. Ethnographic research has contributed compelling illustrations; for example, Farmer’s searing description of how a hydro-electric dam built with international aid to provide electricity for Port-au-Prince propelled rural Haitians into risk-producing contexts forms part of his broader narrative about how international development policies and national economic strategies intersect with poverty, gender inequality, and racial discrimination to propel the rural poor to engage in risk-producing activities (Farmer, 2001, 2006). Others have delineated the HIV vulnerabilities generated by economic policies and projects in developing (de Zalduondo & Bernard, 1995) and developed-world settings (Wallace, 1988). Over the past two decades, this work on the political economy of HIV risk provided a vital conceptual counterweight to the hegemony of the focus on individual-level behavior. Insights about social, cultural, economic and political forces producing risk, however, were rarely followed by operationalizable recommendations for how to reduce HIV vulnerability.

A related body of research has described how poverty combines with other contextual factors to shape sexual networks (Adimora & Schoenbach, 2005) and to increase the likelihood both that women will engage in transactional sex (Ulin, 1992; Dunkle et al., 2004) and that other conditions will shape patterns of sexual partnering and constrain men’s and women’s ability to practice safer sex (e.g., Eaton, Flisher, & Aarø, 2003). Responding to concerns about the vulnerability of the poor (and particularly about the intersection of poverty and gender inequality), as well as calls for context-specific approaches, interventions have sought to reduce HIV vulnerability by ameliorating the effects of poverty or helping individuals escape it.

Other work, however, has found that the relation between income and HIV risk to be context dependent. Wojcicki argued that “marginally increasing select women’s access to funds may have the unintended result of increasing risk” and emphasized the importance of SES at the ecological level (2005: 28). Fox found that wealth’s effect was non-linear, with the wealth associated with risk in poorer settings, poverty associated with risk in wealthier settings, and regional income inequality overall more significant than individual income in shaping risk (2012). Further complicating that initially-widespread belief about the relationship between poverty and HIV, others have also found that patterns “are not uniform across, or even within, countries” (Gillespie, Kadiyala and Greener 2007: S15) and argued for “the complexities and non-deterministic nature of the relationship between structural factors such as poverty or wealth and the risk of HIV infection” (Parkhurst, 2010: 524).

Assessing the HIV risk associated with an individual’s position in a market economy, however, is not the only possible application of political economy to HIV risk. An alternative is to study the impact of economic organization, as for example with research looking at HIV risk and the shift from a centrally-planned to a market economy in China (He & Detels, 2005; Liu, Lian and Zhao, 2006); Vietnam (Phinney, 2008), and Central Asia (Rhodes et al., 1999). Analyzing how economic regimes dependent on migrant labor generate vulnerability also illustrates this alternative approach to the political economy of HIV. Whether in reference to workers in export-oriented factories in China (Li et al., 2010; Sutherland, 2011), men from Swaziland working in South African mines (Hickel, 2012), rural-to-urban migrants within South Africa (Lurie, 2006; Lurie et al., 1995), Brazilians workers who have overstayed their visas in Japan (Komatsu & NSawada, 2007), agricultural workers in Canada (Preibisch & Hennebry, 2011), East African fishermen (Allison & Seeley, 2004), tourism workers in the Dominican Republic (Padilla, 2008) or Mexican migrants to the United States (Albarrán & Nyamathi, 2011), there is abundant evidence that labor migration and work-related mobility produces HIV vulnerability.

This research, however, generally fails to explore the organization of production as a modifiable element of the social context. The private sector response to HIV has primarily consisted of voluntarist efforts (behavioral prevention programs, HIV testing, and enhanced access to AIDS care), carried out under the rubric of corporate social responsibility; Sub-Saharan African mining companies, for example, have described their programs to ensure access to therapy as both cost-effective and “a moral imperative” (James, 2012; Nolen, 2006). The voluntary articulation of concern for the well-being of those whose labor they profit from, however, is hardly an admission that the organization of production shapes HIV vulnerability. Despite the widely recognized connection between, for example, mining and HIV (Campbell, 1997; Hargrove, 2008; Parker et al., 2000), there is no mechanism (either legal or conceptual) to force mining companies to improve living or working conditions. Critiques of the private sector response to HIV have made this point (Bendell, 2003: 34), as has the WHO Commission on the Social Determinants of Health, which noted the importance of the private sector in shaping the general conditions that produce health or disease (CSDH, 2008: 14–15).

Similarly, although markets involve both consumers and producers, little has been said about how consumption at the aggregate level shapes HIV vulnerability. Work on consumption and HIV has explored how an individual’s own desires, whether for alcohol (MacQueen et al., 1996) or modern consumer goods (Hunter, 2002; Parikh, 2007; Smith, 2007) influence engagement in practices that may expose them to the risk of infection. Much has been written about how injection drugs or purchased sex put an individual at risk for HIV infection, and about how products such as alcohol contributes to that risk, but the focus is on risk that recurs to the person doing the consuming, not risk that is produced by broader patterns of socially-organized consumption. Critiques of cause-related marketing such as the (RED) campaign have come close to linking consumption and HIV at the social level. Richey and Ponte, for example, argue that the Product (RED) campaign “masks the social and environmental relations of trade and production that underpin poverty, inequality and disease” (2008: 722). While raising important questions, neither Richey and Ponte nor other critics of HIV-oriented cause-related marketing (Wirgau, Farley and Jensen 2010; Youde, 2009) address how consumer practices might actually contribute to the creation of HIV risk.

Looking at HIV vulnerability in relation to the both production and consumption, therefore, would fill gaps in the political economy of HIV, which has explored risk in relation to an individual’s position within a market economy, a shift to a market economy, or an individual’s own consumption practices. Moreover, the connection between low-waged migrant labor and HIV has received less attention than other frequently-invoked structural determinants of HIV risk (gender inequality, poverty, legal and social discrimination against sexual minorities and injection drug users). Estimates place the number of migrants worldwide at over 214 million (not including rural to urban internal migrants) (Koser & Laczko, 2010), making “one out of every 33 people in the world” a migrant (Weine & Kashuba, 2012: 1605). Thus, researchers and policymakers responding to calls for structural interventions (Gupta, Parkhurst, Ogden, Aggleton and Mahal, 2008; Sumartojo, 2000) and developing strategies for combination prevention (Kurth, Celum, Baeten, Vermund and Wasserheit, 2011) must remember to consider labor migration and mobility as a structural determinant. Labor migration is hardly the only possible application of this meso-level approach; nonetheless, the prominence of labor migration as an element of the social context in regions with substantial existing and emerging burdens of HIV infection underlines the potential importance of thinking more – and more clearly – about labor migration and HIV.

ARTICULATING THE MESO-LEVEL: THE PRODUCTION OF HIV VULNERABILITY THROUGH REGIMES OF MIGRANT LABOR

Connections between migration and elevated risk of infection have been shown in rural-urban migrants (Pison, Le Guenno, Lagarde, Enel and Seck, 1993; Lurie et al., 2003; Mtika, 2007; Yang, 2004) as well as among those who cross international borders (Weine & Kashuba, 2012; Magis-Rodriguez et al., 2004). Elevated risk has been observed for both men and women, with research from Sub-Saharan Africa (Brockerhoff & Biddlecom, 1999; Camlin et al., 2010) and China (Yang & Xia, 2006) suggesting that the social processes through which migration contributes to HIV risk may differ by gender. The intersection between HIV and labor migration has drawn attention both as a general phenomena (e.g., UNAIDS & IOM, 1998; UNAIDS, ILO and IOM, 2008) and in relation to specific populations or migration streams (UNDP, 2008).

Many of the processes through which labor migration contributes to HIV vulnerability, ranging from the most micro-level to the broadly macro-social, have been extensively described. At the policy level, review articles (Albarrán & Nyamathi, 2011; Weine & Kashuba, 2012) point to export-oriented development (which exacerbate the already limited economic opportunities in rural communities) and migration policies, which force some migrants into a liminal legal status). Those syntheses also highlight the characteristics of migrants’ environments such as housing quality; dangerous and stressful working conditions; limited recreational options; financial, health-service, and language barriers; and different norms regarding sexual practices. At the interpersonal level, separation from family, mental health impacts of prolonged absence from home, and limited social support increase migrants’ likelihood of engaging in HIV risk behaviors. The substance use, stress, and depression that result from this noxious environment render migrant workers more likely to engage in sexual behaviors that further exacerbate the risk of HIV infection.

Despite identifying many factors beyond the behavioral level, the preponderance of migrant-oriented HIV prevention work consists of behavioral interventions to increase knowledge and condom use, frequently with an element of community participation or culturally-appropriate adaptation. The prevention interventions developed for migrant workers in the southeastern United States exemplify this (McCoy, Hlaing, Ergon-Rowe, Samuels and Malow, 2009; Sánchez, Silva-Suarez, Serna and De La Rosa, 2012; Rhodes et al., 2006). Even community-level interventions that reflect a keen sensitivity to migrants’ cultural forms, such as those described by Painter et al. (2012), neither articulate nor seek to transform the contextual factors that shape risk behaviors, as the authors themselves note (ibid: 369). (Carillo [2012] notes an important additional limitation of culturally-appropriate interventions for this population, arguing that they frequently rely on an outmoded and static notion of culture that obscures the role of contextual factors in shaping sexual practice.) Calls for understanding migrants’ social and cultural context (e.g., Parrado, Flippen, and McQuiston 2004) underline the importance of considering environmental determinants of migration-related risk as amenable to intervention rather than assuming that migration inevitably produces HIV vulnerability. A limited number of community-level interventions have followed this latter approach, seeking to increase access to health care services, including STI testing and treatment; to protect human rights and increase access to education and credit (Campbell & Williams, 1999; Mooney & Sarangi, 2005; Van Blerk, 2007; Williams et al., 2000); and even to change transport schedules (Lurie, Hintzen and Lowe, 1995).

Going beyond the widespread focus on migrants’ attitudes and behavior and the relative lack of efforts to transform the “working and living conditions that make migrants more vulnerable to HIV/AIDS” (UNAIDS & IOM, 1998: 452) necessitates a far-reaching and political critique of the role of the private sector in producing HIV risk. As the authors of the joint UNAIDS and Institute of Medicine [IOM] analysis of HIV and migration stated clearly fifteen years ago, “In countries receiving economic migrants, employers should be actively involved in improving the quality of life of their employees…[I]n isolated situations such as logging camps, plantations, and mines…it is often the employer (rather than public health authorities) who dictates the health and social services to which the migrants have access” (1998: 452).

It is already understood that labor migration can contribute to HIV vulnerability; what is missing is a strategy for ameliorating the multiple meso-level factors that contribute to that vulnerability. Speaking in generalities about capitalism and risk is insufficient, as is the further production of isolated (albeit moving) descriptions of the conditions in which labor migrants live and work. What is needed is an articulation of modifiable social processes through which labor migration creates HIV vulnerability. Externalities, supply-chain analysis, and the ethics of consumption offer a framework for including both the private sector and consumers as actors in the production of HIV risk for migrant workers.

FROM ANALYSIS TO ACCOUNTABILITY: EXTERNALITIES AND THE ETHICS OF CONSUMPTION

The notion of externalities has been applied widely in population health, both in relation to specific products (e.g., analyses of the food chain [Barling, 2007]) or production processes (e.g., coal mining [Epstein et al., 2011]) and as a framework for thinking about the role of the private sector in shaping population health (Biglan, 2011; Sugarman, 2009). In infectious diseases, the concept has been applied infrequently, primarily in terms of the potential positive externalities that would result from expanding access to care and treatment; this case has been made for infectious diseases in general (Alsan, Westerhaus, Herce, Nakashima and Farmer, 2011; Roberts, Mensah and Weinstein, 2010) as well as for HIV (Stillwaggon, 2009).

Researchers have highlighted health externalities affecting either the consumer (e.g., related to tobacco or junk food [Barling, 2007]) or individuals external to both production and consumption (e.g, as with the impact of pollution on children [Currie, 2011]). Rather curiously given the vigorous public debate about the hidden costs of inexpensive products (e.g, Wyatt, 2012), negative externalities that affect workers themselves are less frequently discussed. Work on coal mining, for example (e.g., Epstein et al., 2011; or Saha, Pattanayak, Sills and Singha, 2011) has emphasized population-level impacts rather than worker health, despite mining being one of the world’s most dangerous occupations (Hambly, 2012). Exceptions exist: a review of externalities associated with pesticide use included the costs of harm to worker health, as well as the impact on agricultural productivity and the environment (Soares & de Souza Porto, 2012). Even there, however, concern about health externalities for workers is sometimes absent; a recent meta-analysis of the health impacts of organic produce which received substantial coverage in the popular press focused on nutrition and cancer risk for the consumer, making no mention of the impact on workers of the pesticide exposure associated with typical industrial agriculture (Chang, 2012; Smith-Spangler et al., 2012).

Perhaps the existence (at least in some places) of laws mandating safe workplaces has made the idea of externalities seem superfluous as a tool to advocate for better protections for migrant workers; however, occupations with significant concentrations of migrant laborers (mining, agriculture, construction) continue to be associated with exceptionally high rates of work-related morbidity and mortality (ILO, 2003), suggesting that laws by themselves are hardly sufficient to guarantee a safe working environment. Moreover, the migration-HIV connection is distal, falling entirely outside current regulatory structures; the private sector may be legally obligated to provide a safe working environment, but no such obligation exists to provide living conditions (family housing, salutary recreational options, etc.) that might mitigate the vulnerability to HIV infection faced by so many migrant workers. Making explicit the ways in which the profits of large-scale employers of migrant labor depend on the low wages and lack of benefits provided to those workers creates a logical mechanism for thinking about the well-being of migrants in a way that goes beyond workplace safety.

In public health, the consuming public is most frequently configured as the object of externalities, rather than the agent of them. The idea of externalities (which is little known outside of academic social science), however, intersects with much more widely-used discourses about ethical consumption, in which individual purchasing choices are framed as “citizenly acts” (Clarke, Barnett, Cloke and Malpass, 2007: 232). Concern for one’s fellow humans is sometimes a prominent part of ethical consumption and sometimes not; a major theme of writing on food ethics is the “plurality of moral claims” (Andersen, 2011: 448) under consideration, with some arguing that in the North American context in particular, “environmental issues tend to overshadow issues related to hunger, social justice, or agricultural labor” (Johnston, Szabo and Rodney, 2011: 295).

The notion of externalities is implicit in discourses of ethical consumption: as one author notes, “A general principle, almost, of ethical food is that it internalises costs externalized by conventional foods” (Lang, 2010: 1821). Vibrant consumer movements advocating conflict-free diamonds, sweat-shop-free campuses, or eggs laid by cage-free chickens operate on the principle that consumers can mitigate collectively the structural violence built into specific commodity chains; moral responsibility is attributed to the consumer who enjoys a product for harm caused in its production. Discourses of ethical consumption have been used by consumers in the global North to reshape international commodity chains for cut flowers (Hale & Opondo, 2005), sweatshirts with University logos (Heintz, 2004), and mobile phones (Wyatt, 2012) and it has been suggested that this “caring at a distance” (Trentmann, 2007) has also caused those who participate to consider the connection between their consumption practices and the well-being of workers closer to home (Silvey, 2004: 192).

IMPLICATIONS FOR HIV PREVENTION

Intervening at the meso level requires a complex set of analytically distinct yet necessarily interdependent elements. These elements include the ways in which health problems are understood to be related causally to conditions or actors; the coalitions that mobilize to transform those conditions, and the policies and programs enacted through this collective action. Shared understandings of who is responsible either for causing or fixing a public problem fundamentally shapes (or limits) the range of policy responses (Gusfield, 1981), and so advocating for a particular framing of a problem must be understood as one element of intervening at the meso-level. The variability in the extent to which clean needles are made available to injection drug users, despite the strong evidence for the effectiveness of needle exchange as a meso-level intervention, underlines the importance of articulating and attending to ideological and political conditions as part of the meso-level. Similarly, in the case of anti-bullying legislation, which has been shown to affect mental health among LGBT youth (Hatzenbuehler, 2011), a meso-level determinant of health became modifiable at least in part due to the social mobilization that resulted from (and perhaps contributed to) an outpouring of public concern about bullying as a social problem. Mapping out who is “responsible” for the vulnerability of migrants to HIV and delineating the modifiable social institutions that create those vulnerabilities sets the stage for collective action for redress. The inclusion of the migrant-employing private sector and the consuming public among the contextual determinants of migrants’ vulnerability to HIV should therefore be a vital component of intervening at the meso-level. Reframing the problem of migration-related HIV is part of creating the conditions for collective engagement to improve the conditions in which the world’s most vulnerable workers live and work.

Civil society is in and of itself a crucial part of the ‘middle range’ shaping the response to HIV (Kippax et al., 2013), and so coalition-building is a second element of meso-level intervention; researchers, advocates, and policymakers concerned about HIV should ally with those working to improve migrants’ living and working conditions, including migrants themselves. The global response to HIV has demonstrated powerfully what is achievable through organized civil society action (Parker, 2011). Although there are certainly instances in which effective collective action may have resulted from the relative class and racial privilege of the actors, the effective coalitions for policy change that have responded to the needs of (and included) sex workers and injection drug users underlines the importance of coalition building for any kind of policy change – perhaps even more so for these acutely marginalized groups. Regarding migrants, there is unrealized potential to collaborate with advocates for labor rights, for local laws that create a climate less hostile to migrant workers, or for food justice, who are already seeking to ameliorate meso-level determinants of HIV risk. Researchers should explore empirically whether advocacy efforts such as the Workers Defense Project, the Coalition on Immokalee Workers, Domestic Workers United, or The Food Chain Workers Alliance (Food Chain Workers Alliance, 2012; Greenhouse, 2013) affect behavioral determinants of HIV risk, either through transforming the social context or through creating pro-social environments for the participants. There is also the possibility of coalitions within public health, between those concerned with a range of disease outcomes; interventions or policies intended to improve mental health or occupational safety and health should be assessed for their impact on behavioral determinants of HIV risk.

Short of the goal of wholesale immigration reform, law and policy at the national, state and local levels can create important gradations between full citizenship and vulnerable invisibility. Initiatives such as the Obama Administration’s Deferred Action for Childhood Arrivals program (which allows for prosecutorial discretion in relation to deportation for individuals who migrated to the US as children without legal documentation [Singer & Svajlenka, 2013]) may affect the meso-level context of risk. A less persecutory legal and administrative context, as manifest in eligibility for college scholarships, access to publicly-financed health services, repeal of English-only laws, and limits on police authorization to check immigration status is – at least in some locations – an achievable policy goal. Many other aspects of law and policy at the state and local level affect the social context of migrant life; increases in the minimum wage, improvements in public transportation, parks and libraries, and low-income housing could all conceivably affect migrants’ vulnerability to HIV. Safer working conditions and (in places where the employers provide housing) improved housing would all address HIV risk, as well as other health risks, at the meso level. Family housing may improve mental health and reduce participation in HIV risk behaviors in nascent epidemics, but it would be less likely to be effective, and may even have adverse consequences, in more disseminated epidemics (Gebrekristos, Resch, Zuma and Lurie, 2005).

Whatever their awareness of responsibility regarding the conditions that contribute HIV vulnerability, it is unlikely that consumers themselves will build better housing, inspect factories, or create salutary recreational opportunities for migrant laborers. Similarly, it is unlikely that the private sector, with its fiduciary responsibility to maximize profits, would voluntarily raise wages or improve working conditions: higher wages are only likely to result from external pressure, either from unions or from increases in the legal minimum wage, and improved occupational safety and health depends on both legislative action and subsequent funding to enforce that legislation. It is vital therefore to consider the circumstances under which either the private sector or the state might come address contextual determinants. Global supply chain activists’ “name and shame” tactics have produced notable successes of late in forcing the private sector to address unsafe working conditions (Alderman, 2013). An additional set of possibilities exist when, as in the case of food systems, elements of the commodity chains linked to the production of HIV are within the same country as the activists; participatory democracy offers numerous routes to press for legal and regulatory reform.

CONCLUSION

Overall, this discussion of meso-level processes, externalities and the ethics of consumption suggests the potential impact of increased dialogue between those concerned about HIV and the increasing numbers of consumers struck by the recognition that an iPad made with rare-earth metals mined in China or a factory farm-produced boneless chicken breast prop up a global system of production dependent on vulnerable migrant labor. Both the gold and the diamond in any engagement ring were likely mined in circumstances that produced HIV risk for a distant miner and that distant miner’s sexual partners; this bitter irony presents an opportunity to mobilize action to mitigate the health impacts of these supply chains. By incorporating consumers in the global North into a conversation about the unmeasured adverse impacts of products they enjoy, an analysis of the economic production of HIV risk can contribute to social mobilization to protect labor migrants. The recognition that “our actions collectively contribute to the complex structural processes that produce the working conditions we deplore” (2003: 40) contains the seeds of potential for change.

In treating cancer, a radiologist can identify a tumor, but then a surgeon is called on to remove it. Similarly, basic social science research can articulate the meso-level processes through which macro-level forms of social inequality create health risks, but distinct forms of expertise will be necessary to use that social diagnosis as the basis for meso-level interventions. An example would be the work of Hirsch et al. (2009), who identify work-related mobility and men’s leisure-time socializing as two “extramarital opportunity structures” (11) through which gender inequality and economic organization facilitate men’s participation in extramarital relations, and recommend HIV prevention policies and interventions aimed at those meso-level phenomena (210–218). The feasibility of those recommendations, or any ideas about meso-level interventions, should be considered in light of the substantial existing knowledge about the circumstances under which economic or social policies or institutions are amenable to change. Context continues to matter; what is modifiable in one context may be tilting at windmills in another, and the extent to which specific social institutions are modifiable is, in and of itself, an important area for research.

The notion of intervening at the meso-level can serve as a heuristic, providing a tool to distinguish between interventions that reflect a structural analysis – such as those addressing risk among labor migrants, most of which are a repackaging of behavior change approaches targeted towards a particular population of migrants – and those that actually seek to transform some element of those structures. For women, gay men, and drug users, it has been possible to transform some of the meso-level structures that express and reproduce vulnerabilities created by macro-level social processes: interventions to change inheritance laws or reduce gender-based violence (Jewkes & Morrell, 2010) address aspects of gender inequality, anti-homophobia campaigns (Lyra, 2008) challenge heteronormativity, and policies that facilitate access to sterile injection equipment (Des Jarlais et al., 1995) challenge a critical manifestation of the stigma faced by IDUs. Those engaged with HIV prevention for migrants must focus, similarly, on modifying the contextual determinants of vulnerability.

Research Highlights.

• HIV research on structural violence should delineate meso-level social processes that create risk

• Labor migration is one example of a meso-level process linking structural violence to HIV

• Externalities and ethical consumption are vital concepts for analyzing labor migration and HIV

• Low-waged migrant labor benefits producers and consumers and contributes to HIV vulnerability

• Seeing risk for migrants as an externality suggests new approaches to HIV prevention