Complex linkages between forced labor slavery and environmental decline in marine fisheries

abstract Recent media attention on human rights abuses in the fishing sector, precipitated by undercover investigations from nongovernmental organizations and investigative journalists (e.g., Environmental Justice Foundation [EJF] 2014, 2015a, 2015b; Mendoza, McDowell, Mason, and Htusan 2016), has prompted calls from the scientific community for increased transdisciplinary and empirical research of fisheries’ social dimensions, such as labor (Kittinger et al. 2017). Given views that social and ecological systems are interdependent (Ostrom 2009), the need for theory development to explicate pathways for how this interdependence occurs and the potential for using policy and practices for intervention and prevention exist. Integrating ecological data and economics and human rights theory, Brashares and colleagues’ (2014) wildlife decline and social conflict framework offered a hypothesis about the negative association between fish stock declines and child slavery. Yet, more precision in terminology, pathways, and feedbacks may be warranted. With the aim of exploring empirical, conceptual, and theoretical support for Brashares et al.’s (2014) pathways, the revised theory developed in this article posits how forced labor slavery and environmental decline in marine fisheries may be linked.

fisheries use. From the social systems perspective, if shocks to fisheries create economic pressures within a preexisting context of social vulnerability already known to yield slavery, then fish stock declines may exacerbate and accelerate the use of forced labor slavery in the sector by providing an economically rational tipping point into the practice. Thus, failing to recognize how environmental decline contributes to and compounds social vulnerabilities could undermine interventions aimed at eradicating slavery. These interventions risk being siloed within social systems and ignoring ecological feedbacks. From the ecological systems perspective, not considering the extent of unique environmental pressures produced by the use of forced labor slavery could subvert attempts to stabilize stocks and to efficiently maximize sustainable yields. Thus, when stocks decline, social consequences may increase vulnerabilities for fish-dependent persons (e.g., Perry and Sumaila 2007;Golden, Allison, Cheung, Dey, Halpern, McCauley, Smith, Vaitla, Zeller, and Myers 2016). As a result, these feedbacks create a multifaceted problem that requires multiple and holistic interventions and policies that address fisheries' environmental and labor challenges in conjunction. This article will consider such associations by defining what constitutes forced labor slavery in the twenty-first century. Brashares and colleagues' (2014) original framework and critique will be presented, along with a revised framework and diverse knowledge sources to support pathways.

Defining forced labor slavery
The International Labour Organization's (ILO;1930) Convention Concerning Forced or Compulsory Labor defined forced labor as "all work or service which is extracted from any person under the menace of any penalty and for which the said person has not offered himself voluntarily" (Convention 29, Article 2). The ILO's Special Action Programme to Combat Forced Labour suggested eleven indicators to identify potential victims of forced labor slavery. Although the presence of just one indicator could constitute a case of forced labor, it is often thought of as a continuum, based on victims' vulnerabilities and the severity and number of indicators present (Special Action Programme to Combat Forced Labour 2012).
For this article, forced labor slavery will be defined as the involuntary entry and "holding of people at a workplace through force, fraud, or coercion for purposes of forced labor so that the slaveholder can extract profit" (Free the Salves 2017). Slavery was selected over human trafficking, a primarily legal term subjected to nuanced and often conflicting interpretations from diverse legal institutions influenced by various external forces (Bales 2017).
Instead, this definition centers victims' experiences versus legal frameworks (Bales 2017), and is holistic enough to encompass all aspects of forced labor slavery while also noting the shift from historical to modern slavery. In his theory of modern slavery, Bales suggested that slavery as a construct should still be defined by the relationship between victim and perpetrator (consistent with historical slavery), but that over time slavery has shifted from an owner-property relationship to a relationship in which the victim is paid little (an unfair value) or no money (labor exploitation) while the perpetrator's profits increase (2006). Other shifts include the transposition of unfreedoms from point of entry into the exploitative relationship (historical slavery) to the point of exit from the relationship (modern slavery; see Barrientos, Lothari, and Phillips 2013;Phillips and Mieres 2015;Stringer, Whittaker, and Simmons 2016) and control of the victim at point of entry into the relationship being exerted by a person (historical slavery) to socioeconomic conditions (modern; see O'Neill 2011). Indeed, although some victimized fishers are purchased by boat captains (Chantavanich, Laodumrongchai, and Stringer 2016), many exhibit the semblance of agency at point of entry caused by desperation to meet basic needs and exploited by brokers ' and/or recruiters' deception (O'Neill 2011).
The shift from historical to modern slavery has further complicated the identification of forced labor slavery, and often exploitation in the fishing sector has been minimized as poor labor practices (Stringer, Whittaker, and Simmons 2016). As a result, investigative journalism's contributions to eradicating abuses in the fishing sector exceed the scientific community's, requiring researchers to improve their consideration and inclusion of equity, equality, and social justice in environmental and sustainability research (Kittinger et al. 2017). Whereas investigative journalism has confirmed and generated mainstream media attention to the presence of labor exploitation in the fishing sector, the scientific community can advance this work by linking social and ecological processes to outcomes such as forced labor. Additionally, although multiple white papers have credibly postulated bidirectional relationships between fish stock declines and forced labor slavery (see United Nations Office on Drugs and Crime [UNODC] 2011; EJF 2015b), empirical evidence is limited, and improved research is needed to understand how strongly the issues are linked and the processes that facilitate these linkages.
Brashares's wildlife decline and social conflict framework Brashares et al. (2014) offered a specific theoretical framework hypothesizing how fish stock declines may be driving increases in child slavery through an amalgamation of human rights, political ecology, conservation biology and ecology, public health, and economic theory (see Figure 1). Building on the UNODC's (2011) investigation that posited a potential relationship between declining fish stocks and human trafficking (Section 1.5.4), the authors proposed that declining fish stocks force vessels to fish longer, farther from shore, and deeper in waters to maintain yields, increasing "production costs" (Brashares et al. 2014: 376). The framework hypothesizes cheap labor as an approach to offset increasing costs and continue harvesting fish species at a rate that would otherwise be cost-prohibitive, thus potentially increasing exploitative labor practices to the point of child enslavement (Brashares et al. 2014). Although Brashares et al. emphasized child slavery, it is expected that their hypothesis is more relevant to exploited adult laborers, as indicators for child slavery must consider differing cultural norms around child work, particularly between developing and developed countries.  Brashares et al. (2014, p. 377) offer a hypothesized feedback loop linking fishery declines with exploitative labor practices such as child slavery. It is hypothesized here that as demand for cheap labor increase, more people will be enslaved.
Citing a lack of empirical evidence for all pathways, critics denounced the framework for oversimplifying a complex problem and making too big of a conceptual leap from fishery declines to slavery, thus conflating the environment's role in perpetuating slavery (Mauda and Scharks 2014). Although Brashares et al. were unable to support their pathways with empirical data at publication, it does not mean the phenomenon under investigation is implausible, nor does it preclude future studies from empirically confirming the model. Scientific precedent exists for empirically uncorroborated models having great utility in directing future research and advancing understanding of complex phenomenon (e.g., string theory; see Dawid 2006;Dawid, Hartmann, and Sprenger 2015). And in the social sciences, common scientific practice is to move from hypothesis generation to empirical testing by formulating a framework, model, or theory that organizes pertinent constructs (e.g., predictors, mediators, modifiers) into schemas to better predict outcomes and events under study (Jaccard and Jacoby 2010).
Although Mauda and Scharks (2014) cited literature from 2005 that did not consider environmental degradation as a driver of slavery (Surtees 2005), current literature is beginning to consider larger environmental and market-driven processes, including overfishing as a "key associated driver" of maritime crime (e.g., forced labor; Pomeroy, Pomeroy, Parks, Mrakovcich, and LaMonica 2016: 96) and how slavery may be escalating environmental degradation, which in turn increases slavery (Bales 2016). Investigations also suggest the fishing industry is one of the biggest users of slave labor, with an estimated 1.8 million people enslaved in the agriculture/fishing sector (ILO and Walk Free Foundation 2017), including on fishing vessels originating from and/or berthed in the United States, Thailand, New Zealand, and Peru among numerous others (International Transport Workers Federation [ITF] 2006;Bales 2016;ILO 2013;Yea 2014;FishWise 2014;EJF 2014EJF , 2015aMendoza et al. 2016; Verit e 2016). Mauda and Scharks's (2014) concerns about needlessly reallocating or misappropriating resources are germane. However, the persistence of forced labor slavery in marine fisheries (Shen & McGill, 2018) warrants a challenge to the dominant paradigm that has seemingly not produced reductions in the problem. This new discourse, instead, should consider contributing factors external to social vulnerabilities (e.g., environmental decline) that may influence the use of forced labor by creating a demand for free or cheap labor. And it should question the framing of slavery's contributions to illegal fishing and overfishing, wherein isolated environmental policies insufficiently attuned to forced labor slavery's contributions to environmental decline may inadvertently blame or punish slavery victims. A potentially more holistic understanding of social-ecological marine systems and subsequent appropriately targeted and multifaceted interventions could also advance the field by shifting it from reactive to preventive practices. No economic or other incentive is known to halt the use of forced labor slavery once such practices have generated increased profits. Therefore, to reduce the prevalence of this linked social-ecological injustice, interconnected interventions must prevent it before it happens.
Brashares's wildlife decline and social conflict framework revised Although Brashares and colleagues' (2104) framework is sufficiently developed to encourage empirical testing to confirm pathways and link previously disparate research fields, a more robust discussion of underlying theories may provide nonexperimental confirmation of pathways to repudiate previous critiques. Building on Brashares et al.'s original framework, a revised framework with altered constructs is presented in Figure 2. Tentative construct changes (see Table 1) were made to be consistent with human rights literature and theory and to emphasize power differentials that incite exploitative labor relationships, giving the powerful economic advantages while commodifying human beings (Manzo 2005;Phillips and Mieres 2015). Support for each construct and pathway in the revised framework will be described below.

Contextual Constructs
Although forced labor slavery is a global phenomenon, not all fishers are enslaved, and context influences the labor relationship between fishers and employers. Important conditions underlie the persistence of forced labor slavery in the fishing industry, including geographic, regulatory, cultural, socioeconomic, and industry contexts-all constructs derived from Bales's (2006) and Crane's (2013) theories of modern slavery (see Figure 2). These constructs also provide the indicators used to estimate the prevalence of slavery. Geographic, regulatory, and cultural contexts create an environment that not only accommodates but enables slavery. They also interact with each other and with the socioeconomic context, creating the slave labor supply, and the industry context, creating the "demand" for slave labor. Although the operationalization of these broad concepts varies by country and region, key indicators of each that should be assessed in future empirical work are described. Moreover, although the importance of each contextual construct and indicator in driving slavery may vary across regions, it is the confluence of these factors in creating no viable alternatives (either actual or perceived) for victims that leads them into forced labor slavery experiences.

Geographic context
Although forced labor slavery occurs in developed, transitional, and developing countries, geographic factors such as a high density of migrant laborers and geographic isolation appear to increase the use of forced labor slavery (ILO 2005;Robertson 2011;Crane 2013). Empirical evidence suggests that migrants are the population most vulnerable to forced labor slavery (ILO 2005; International Organization for Migration [IOM] 2008;Wheaton, Schauer, and Galli 2010;Chantavanich et al. 2016). Their movement may be documented or undocumented, and is predicated on perceived opportunity in the form of higher wages or greater availability of work in the new area (Chuang 2006;Bales 2006Bales , 2007Wheaton, Wheaton, Schauer, and Galli 2010;Robertson 2011). Brokers and recruiters working for boat owners target them through formal and informal mechanisms, in markets before their migration (offering "assistance" with their Other pressures (e.g., population growth and environmental degradation) are also limit work opportunities further inland, driving more migrants to coastal areas (Creel 2003), and thus the fishing sector.
Like other industries with high forced labor usage (particularly natural resource extraction industries), fishing is an economic activity that occurs at specific sites, which can be hundreds of miles from shore (Crane 2013). This distinct separation and geographic distance minimizes contact between victims and law enforcement, family, civil society and aid organizations, and professional groups, intensifying fishers' dependence on their abusers, which increases the perpetrators' power (Fletcher, Bales, and Stover 2005). Over time, the experience becomes normalized, and as enslavers continue to gain power, it lowers the amount of resources needed to dominate the Fish stocks "The living resources in the community or population from which catches are taken in a fishery." (FAO, 2011, Glossary) Fish catch-per-unit-effort "The quantity of fish caught (in number of in weight) with one standard unit of fishing effort; e.g. number of fish taken 1000 hooks per day or weight of fish, in tons, taken per hour of trawling." (FAO, 2002, Glossary). Effort "The amount of fishing gear of a specific type used on the fishing grounds over a given unit of time e.g. hours trawled per day, number of hooks set per day, or number of hauls of a beach seine per day. Fishing effort would frequently be measured as the product of: (i) the total time spent fishing, and (ii) the amount of fishing gear of a specific type used on the fishing grounds over a given unit of time. When two or more kinds of gear are used, they must be adjusted to some standard type in order to derive an estimate of total fishing effort" (FAO, 2003, Glossary;FAO, 2011, Glossary). Profit margins Financial revenue from fishing sales exceeding the costs associated with fishing. Cheap labor demand Labor exploitation The desire to pay a workforce substandard or no wages in order to offset growing costs and expenses. When individuals involuntarily enter into a labor relationship (e.g., they cannot consent due to deceit, kidnapping, physical confinement, etc.); receive little (wages are unfair in relation to work) or no direct financial compensation for their work; and are unable to leave the exploitative situation due to the presence of threat of violence or other coercive actions (Bales, 2016;ITUC & ILO, 2008). Profits Financial gains at the expense or unfair treatment of others through forced labor. According to Bales' Theory of Modern Slavery, forced labor exploitation that leads to profit increases is the required economic relationship to constitute slavery (2006). Slavery The involuntary entry and "holding of people at a workplace through force, fraud, or coercion for purposes of forced labor so that the slaveholder can extract profit" (Free the Slaves, 2017). Effort "The amount of fishing gear of a specific type used on the fishing grounds over a given unit of time e.g. hours trawled per day, number of hooks set per day, or number of hauls of a beach seine per day. Fishing effort would frequently be measured as the product of: (i) the total time spent fishing, and (ii) the amount of fishing gear of a specific type used on the fishing grounds over a given unit of time. When two or more kinds of gear are used, they must be adjusted to some standard type in order to derive an estimate of total fishing effort" (FAO, 2003, Glossary;FAO, 2011, Glossary).
victims-further reducing the cost of using slaves (Crane 2013). This isolation also limits the reach of regulatory powers. In localities with effective labor laws, inspectors or enforcement agents often cannot access vessels at sea for compliance monitoring; thus, violations are unnoticed. And the plethora of geographic and physical boundaries transected by marine fishing vessels exposes loopholes in labor regulations even in developed countries (MacFarlane 2017). Vessels will also engage in transshipping, the use of mother ships or reefer ships to unload their catch, refuel, and restock in the middle of the ocean to prevent berthing in ports. Using transshipping, some vessels have reportedly remained at sea for several years (EJF 2015a). As a result, the geographic context shapes the regulatory context because illegal practices like slavery persist in industries that "operate beyond the oversight of regulations and other formal institutions" (Crane 2013: 54).

Regulatory context
Although the Universal Declaration of Human Rights illegalizes slavery everywhere, ineffective governance fosters environments in which forced labor slavery still thrives. The use of forced labor slavery most frequently occurs in countries characterized by government complicity, political instability, high levels of corruption, limited regulations, regulatory failures, and poor natural resource (fisheries) management (Crane 2013;Bales 2016;Pomeroy et al. 2016). And despite the Universal Declaration of Human Rights and subsequent conventions, international frameworks do not "translate into real protections unless they are incorporated into national legislation and implemented effectively" (ITF 2006: 9). Pomeroy et al.'s (2016) theoretical "fish wars" framework also posited that, in the absence of effective governance in fisheries, natural resource scarcity can result in increased social conflict (including human trafficking and forced labor). The social conflict contributes to the environmental degradation through feedback loops predicated on increased competition. Examples of weak governance in fisheries include corruption; lack of stakeholder participation, political will, and capacity; weak institutional capacity and capabilities; poor enforcement; and inadequate information and data. Investigative case studies have also suggested that vessels using forced labor are often disregarding international and national environmental regulations and engaging in illegal, unreported, and unregulated fishing (IUU) fishing (EJF 2015b). IUU activities may involve knowingly violating catch quotas; purposefully not reporting, underreporting, or falsifying catches; fishing without a license; fishing in protected areas; catching certain species that are illegal; and using illegal fishing gears. The persistence of IUU fishing is often facilitated by regulatory corruption, limited enforcement capacity, and loopholes in policies rendering the environmentally destructive fishing activities unregulated but not necessarily illegal (Global Ocean Commission 2013).
However, forced labor slavery is not just limited to fisheries in developing countries. It also occurs in developed and transitional countries because the fishing industry is often part of the informal economy-industries that lack employment security, benefits, and labor laws and regulations-making laborers more vulnerable to exploitation (Hart 1973).

Cultural context
Deep-rooted discriminatory beliefs and social inequalities within cultural contexts exclude groups of people from rights and protections and inhibit equitable development, particularly economically, by assigning these groups a subordinate status in society-thus making them more vulnerable to slavery (Crane 2013; Free the Slaves 2017). For migrants, this discrimination may occur in both their countries of origin and destination. In certain geographic regions, these forms of discrimination are institutionalized by the regulatory context. Even if labor laws and regulations exist, they may not be applied equitably, or they can legally discriminate against an individual based on gender, race, tribe, caste, religion, or immigration/migration status, resulting in most forced labor slavery victims identifying with at least one minority group (Upadhyaya 2008;Crane 2013). For example, in many countries, foreign-born, migrant workers are exempt from local labor laws. Additionally, in some countries, types of permissible work are determined by informal norms regarding social membership. As a result, entire groups of already vulnerable people become further marginalized, which means they will also disproportionately incur socioeconomic challenges.

Socioeconomic context
Geographic, regulatory, and cultural contextual variables combined create structural vulnerabilities that exacerbate socioeconomic inequalities. These socioeconomic inequalities may include wealth gaps and disparities in income, poverty and education levels, and access to financial institutions (Bales 2006(Bales , 2007Andrees 2008;Crane 2013). The inequalities may also occur intercountry (e.g., Thailand's economic growth relative to poorer neighboring countries like Cambodia and Laos, which facilitates migration from Cambodia into Thailand) or intracountry (e.g., Myanmar's economic growth, which is creating greater income inequality between skilled low-skilled works). Regardless, the powerful dominate the vulnerable by exploiting these inequalities (Barner, Okech, and Camp 2014). Factors that can moderate the relationship between socioeconomic vulnerabilities and forced labor slavery include access to affordable credit and education (including literacy and language skills). However, migrant populations often lack both, and culturally entrenched discrimination further limits their access, increasing their desperation and vulnerability to coercion (Andrees 2008).
Combined, socioeconomic inequalities create a surplus population vulnerable to forced labor slavery (the slave labor "supply"). The increased availability of "cheap labor" then causes the price of slaves to decrease, further inflating slaveholders' profits (Bales 2012;Crane 2013). This socioeconomic context also interacts with the contextual constructs in that migrant populations (geographic) are adversely incorporated into society-and thus the society's fishing sector-because of laws (regulatory) and discrimination (cultural). This adverse incorporation reinforces poverty; thus, a population vulnerable to slavery persists (Phillips and Mieres 2015).

Industry context
Whereas the socioeconomic context provides the supply, the industry context creates the "demand." Research suggests that modern forced labor slavery is most likely to occur in industries reliant on manual labor, exhibiting high labor intensity and low technological development, and operating in the informal economy under poor regulations (Chuang 2006;ILO 2009;Bales 2012Bales , 2016Crane 2013). Motivated by profits, forced labor slavery represents an economically rational decision to employers because it is "an opportunity to reduce the main costs driving profitability" (Crane 2013: 54; see also Wheaton et al. 2010). Because labor is often one of few production elements employers can control, the economic benefits of forced labor slavery are greater in labor-intense industries in which profit margins at the supply chain's source are increasingly narrow, such as fishing (Crane 2013;Hamilton-Hart and Stringer 2016). Because fishing already has a propensity for using forced labor slavery, it is not implausible to consider that fish stock declines and forced labor slavery are linked issues, especially considering that many adaptations to stock declines (e.g., increased effort) likely constrain profits further. Then the use of forced labor incentivizes increased pressures on stocks by delaying overfishing's unprofitability.

Empirical pathways
Declining fish stocks decrease fish catch-per-unit-effort Although the discourse about how dire stock declines are continues between conservation and fisheries management approaches (Worm et al. 2009), in 2013 an estimated 31.4 percent of marine fish stocks were overfished beyond biologically sustainable levels. This was a more-than-20 percent increase from 1974 (FAO 2016). Although management strategies have been effective in stabilizing some stocks, a global meta-analysis of overfished stocks determined that many stocks will likely take substantially more time to recover than initially predicted-even under aggressive, best-case scenario conservation and management approaches (Neubauer, Neubauer, Jensen, Hutchings, and Baum 2013).
When marine fish stocks decline, catch-per-unit-effort (CPUE) also decreases (Daskalov 2002;Schrank 2012, 2014;Watson, Cheung, Anticamara, Sumaila, Zeller, and Pauly 2013;Tsikliras, Dinouli, Tsiros, and Tsalkou 2015;Pauly and Zeller 2016). Although there are examples of CPUE increasing amid stock declines (Rose and Kulka 1999;Harley, Myers, and Dunn 2001), this outcome is dependent on numerous other factors, such as fishing fleet efficiency, changes in fishing technique and skill level, and technology advancements (Maunder et al. 2006)-all of which typically require substantial financial capital. Because forced labor slavery, in all industries, is used to increase profit as it requires limited financial investment, vessels using slave labor are unlikely to be motivated to expend financial capital into new technologies.
Others have suggested that declining catch may be an indicator of better fishery regulations (Worm et al. 2009). Although this may be relevant for some marine areas, it is not pertinent to the proposed framework, as all slavery activities are illegal and occur outside the realm of labor and other regulations. Furthermore, the observed co-occurrence of IUU fishing and forced labor slavery in investigative case studies from Thailand (see EJF 2015b) suggests that forced labor slavery may be more likely to occur on vessels fishing in unregulated areas or areas where fishery regulation enforcement is unfeasible or nonexistent-consistent with the use of forced labor slavery in deforestation (Bales 2016). Illegal fishing also frequently results in noncompliance with other environmental, national, and international regulations and actions (Agnew et al. 2009). And as the economic benefits of IUU already far exceed the risks of being apprehended or fined (Sumaila, Alder, and Keith 2006), the use of slave labor likely further reduces the risk. However, more robust empirical data are still needed.
Declining fish catch-per-unit-effort increases effort When CPUE decreases, vessels fish longer, deeper, and further, increasing effort (Hutchings and Myers 1995;Watson et al. 2013;Pontecorvo and Schrank 2014;Tsikliras et al. 2015;Bell, Watson, and Ye 2017;Gascuel, Coll, Fox, Guenette, Guitton, Kenny, Knittweis, Rasmus Nielsen, Piet, Raid, Travers-Trolet, and Shepard 2016). Critics of Brashares et al.'s original framework, citing contrary evidence, have argued that fish stock and catch declines lead to decreased effort (Smith 2014). However, Smith's cited findings were less precise and more speculative, relying on predictive modeling for future forecasting that used cross-sectional data (Liese, Smith, and Kramer 2007) or data from a fourteen-year period at best (Bjørndal and Conrad 1987). The study of Kenyan fishers questioned about their willingness to exit the artisanal fishing industry in light of hypothetical fish decline scenarios (Cinner, Daw, and McClanahan 2009) is also likely not applicable, as artisanal fisheries often diverge from industrial fisheries in terms of motivation (i.e., subsistence versus profit). Although investigations have uncovered slavery victims on short-and longhaul vessels and on boats fishing in territorial waters and on the high seas (Chantavanich et al. 2016;Mendoza, McDowell, Mason, Htusan 2016), the commercial nature of the fishing vessel appears to be a common characteristic.
In contrast, data suggesting increased effort amid stock and catch declines used more robust and rigorous methodologies, including longitudinal data sets with more time points. Bell, Watson, and Ye (2017), calculated effort and catch using a global sample generated by the FAO and other sources over a 62-year period (1950-2012). Watson et al. (2013) aggregated data from the European Union, FAO, global tuna commissions, and the Commission for the Conservation of Antarctic Marine Living Resources (CCAMLR), arguably one of the more reputable regional fishery management organizations, over a 56-year period . Furthermore, Hutchings and Myers (1995) created a 450-year historical reconstruction of Northern cod catch and effort based on iterative analyses of archival records, the North Atlantic Fishery Organization's database , and a spatial analysis of effort using historical documents. Although it is likely scenarios exist in which declining CPUE does not lead to increased effort based on contextual factors, the rigorous studies described suggested that numerous scenarios exist in which CPUE does lead to increased effort, and it is in these fisheries that forced labor slavery is likely to occur.

Increased effort decreases profit margins
Applying Clark's (1990) "stock effect" to fishing, White, Kendall, Gaines, Siegel, and Costello (2008: 371) concluded that "the cost to catch a fish increases as the density of a fish population declines … thus more intensive fishing pressures [increased effort] may compromise profit." Fish stock declines in coastal areas have driven fishing vessels out to the high seas; yet the technology and other increased effort costs associated with high sea and deep sea fishing are so exorbitant that many types of fishing and fisheries are not profitable without government subsidies or other cost-cutting measures in relation to labor expenses (Gjerde, Currie, Wowk, and Sack 2013;Sala, Mayorga, Costello, Kroodsma, Palomares, Pauly, Sumaila, and Zeller 2018).
Research reliably estimating fishing costs, which directly impact profit margins, is limited. However, a longitudinal study coupling global catch and economic data from 1951 to 1999 found a 95 percent revenue decline during the study period (Sethi, Branch, and Watson 2010). If revenue is decreasing and cost is increasing, it is expected that profit margins would also decrease.

Theoretical pathways
Theoretical pathways are the relationships between constructs that are supported by an integrated set of disciplinary theories, which are used to explain the processes of why and how forced labor slavery persists.
Decreased profits increase demand for cheap labor Increases in demand for cheap labor are adaptations to profit losses driven by economic motives. The Domar serfdom model postulates that businesses are constantly seeking economic gains, which are typically made through production control. However, when production is scarce, it becomes more economically advantageous to own the labor force instead of production (Domar 1970). The hypothesis initially described increases in agricultural serfdom in the sixteenth and seventeenth centuries as a response to land scarcity, and was broadened by Domar to include production scarcity, with production encompassing resources and capital (1970). It is plausible the model would apply to natural resources, like fish, as other economists have identified fish functioning as capital (Brown 2000). Fish in marine ecosystems are considered a global common good that no particular party owns, and with the increase in high seas fishing and the presence of government subsidies (Gjerde et al. 2013), production ownership is frequently limited to a single element-labor. As such, unfree labor functions like government subsidies, delaying the tipping point from profitable to unprofitable (Sala et al. 2018).
Consistent with other commodity-oriented global value chains (GVCs), when fishing profits decrease, market pressures drive owners or operators to resort to measures to remain competitive (Chantavanich, Laodumrongchai, and Stringer 2016). Fishing is a profit-driven industry, in which net profit drives harvest decisions, creating a "race for fish" that rewards fishers who can harvest the most fish and maximize individual economic gains by reducing input costs (Sethi et al. 2010). And in the fishing industry cheap, migrant labor (regardless of how it is obtained) is considered a legitimate strategy-indeed a one of the primary strategies-for remaining competitive (ITF 2006).

Increased demand for cheap labor increases forced labor exploitation
Labor ownership (i.e., slave ownership) makes fishing more profitable and gives owners or operators a competitive advantage, essentially rewarding the use of cheap labor. Because decreasing fish stocks intensify effort and decrease profits, they indirectly increase demand for cheap labor and create market forces that make impoverished migrant workers more vulnerable to exploitative labor practices (UNODC 2011;Chantavanich et al. 2016). As noted in Crane's (2013) theory, in GVCs, when already narrow profits begin to dwindle because of increasing labor intensity, owners or operators will seek to zero their labor expenses, increasing the prevalence of forced labor exploitation because it becomes an economically rational decision.

Increased forced labor exploitation increases profits
Forced labor exploitation is an effective strategy for increasing profits because modern slavery is relatively inexpensive. Whereas slavery throughout the United States and Europe in the 1700s and 1800s required the purchased of slaves, modern forced labor slavery relies on coercion and deception instead of purchase (Bales 2016). Additionally, there are almost no costs associated with the coercion and deception, as intermediaries (i.e., recruiters and brokers) and boat captains exploit the preexisting vulnerabilities created by geographic, regulatory, and cultural factors and accentuated by the socioeconomic predictors. Intermediaries deceive vulnerable persons (most often migrants) into these schemes by offering employment agreements with competitive wages that are never paid, or agreements with advanced wages, debt repayment, or equipment loans in exchange for labor until the "debt" to the employer is satisfied (i.e., debt bondage schemes, which constitute one form of modern forced labor slavery). These schemes persist as employers continuously add new debts (e.g., for food and shelter), making repayment impossible, enslaving the laborer, and increasing the employer's profits (ILO 2005;Bales 2006Bales , 2007Bales , 2012IOM 2008;O'Neill 2011;Yea 2014). When labor exploitation increases profits, modern slavery exists (Bales 2006(Bales , 2007.

Proposed pathways
Increased profit increases effort Once slaveholders in the fishing sector have perfected their strategy for minimizing cost and maximizing profit, they will continue to exploit this strategy as long as two conditions are met (Bales 2016). The first is the continued existence of a supply of slave labor, generated by socioeconomic and other inequities. The second is the continued global consumer demand for fish, driven primarily by industrialized nations. Between 1960 and 2010, the annual per capita fish consumption increased from 9.9 kilograms to 19.2 kilograms, driven primarily by industrialized countries and predicated on a growing obsession with exotic fish products (such as sushi), trade globalization, human population increases, and increasing scientific evidence of fish's health benefits (FAO 2014). Although the least developed countries saw an increase as well, their annual per capita consumption was almost 50 less than that of industrialized nations, despite the overreliance on fish for subsistence (FAO 2014). As such, Crane (2013: 56) hypothesized that enslaving migrants through forced labor exploitation will "effectively lock in low-price labor," institutionalizing the practice.

Increased effort decreases fish stocks
Because perpetrators of forced labor slavery are profit motivated and already operating outside laws and regulations, it is unlikely they will comply with binding laws that lack enforcement or international soft laws bereft of punitive consequences for overfishing or slavery violations, or that they will comply with equipment restrictions or catch limitations if their actions are increasing profitability (Bales 2016). Thus the use of slavery could perpetuate overfishing by delaying the unprofitability. Additionally, if forced labor does further decrease the risk associated with IUU fishing, the increased use of slavery could contribute to even higher rates of IUU fishing-already an identified driver of overfishing and marine fish stock declines (Global Ocean Commission 2013;FAO 2014). However, although the co-occurrence of IUU fishing and forced labor slavery has been documented (see EJF 2015b), the relationship between the two problems remains poorly understood and underresearched.

Conclusion
The proposed framework's utility is in connecting theories that have long been siloed in individual disciplines. These transformative connections thus have the potential to provide a strong foundation for future empirical testing; development of causal theories-including bidirectional and cyclic theories; identification of modifiable factors as intervention points; and elevation of the need for multiple entry points to address linked social-ecological problems. As the measurement of all forms of slavery evolves and continues to improve (Bales 2017;Larsen and Diego-Rosell 2017;Larsen and Durgana 2017), it is important to consider measurements beyond prevalence that will further understanding of forced labor slavery-including its external contributing factors, contributions to environmental degradation, and most efficacious modification points.
The proposed framework and suggested construct definitions and indicators provide the basis for modeling case scenarios and identifying feedback loops in social-ecological systems. Rather than emphasizing a singular causal factor (either direction), the framework's application is intended to account for the complex interrelatedness of variables in the linked system, thus elucidating an understanding of how social and ecological conditions interact to create feedback loops. Better identification of these feedback loops can also help model repercussions of policy and governance decisions-the most likely intervention mechanism. Too often policies are constructed to deal with problems in isolation, limiting the authority of enforcement mechanisms for addressing problems outside of this limited purview. Because multiple factors, which are themselves interrelated, contribute to slavery in the industry, research and interventions (including policy responses) need to be multifaceted and implemented in unison or in complementary approaches.
Although the proposed framework has not yet been empirically validated, and empirical evidence to support the relationship between natural marine environmental decline and forced labor slavery may be difficult to obtain, it is important to continue moving the field forward. Connecting theories that have long been segregated in individual disciplines provides a strong foundation for future empirical testing. Without this progress, the potential links between marine environmental decline and social conflicts will continue to perpetuate socio-ecological injustices such as slavery in the marine fisheries sector, which simultaneously perpetuates human rights violations and marine environmental degradation.

Funding
The researchers did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Neither author has a conflict of interest to report.

Notes on contributors
Jessica L. Decker Sparks is a graduate of the University of Denver's Graduate School of Social Work's doctoral program, a course director in the Cummings School of Veterinary Medicine at Tufts University's Conservation Medicine program, and a research associate at the Rights Lab at the University of Nottingham. Her research uses mixed-methods approaches to explore, understand, and quantify linkages between overfishing, marine fish stock declines, and forced labor slavery.
Leslie K. Hasche is an associate professor at the University of Denver's Graduate School of Social Work. Her gerontological and mental health services research informs her teaching of theory-based and contemporary issue courses related to social work, aging, and intergenerational justice.