17Understanding the Climate Change–Migration Nexus

17Understanding the Climate Change–Migration Nexus

Chapter 2

There is growing realization among researchers, policy makers, and the wider public that the impacts of climate change will have a large effect on global migration patterns in the coming decades. Although there is no consensus on the current number of climate migrants, most research recognizes that the rate of climate migration is increasing and that growing climate risks in the coming decades will accelerate this trend. This chapter presents the state of play of migration trends, with a focus on the impacts of climate change on these patterns, and provides context for the modeling approach and the analysis done in this report.

Migration has important impacts on people, places, and development. It can enhance adaptive capacity under certain conditions, but it can also destroy livelihoods if not adequately planned for and managed (McLeman 2016). A rich body of literature describes the conditions and complexities underlying migration (including climate change-induced) and the impacts of climate migration (albeit to a lesser degree). Research on climate migration first emerged in the 1980s. Output rose to about 10 peer-reviewed publications a year in the 1990s and to more than 100 a year since 2008, including at least 447 empirical case studies, mostly on Africa, the Asia-Pacific region, and the American continent (Ionesco, Mokhnacheva, and Gemenne 2016).5 The work carried out for this report builds on previous work.

This chapter overviews the scale of current internal migration, underlying causes, and its impacts.6 Overlaying this information with expected climate trends, the chapter aims to draw attention to the change in the pace of climate migration and its future impacts. This context informs the modeling approach in Chapter 3, and the interpretation and narrative of the results in the subsequent chapters.

5. Ionesco, Mokhnacheva, and Gemenne (2016) draw on studies recorded in the Climig Database by the Institute of Geography University of Neuchâtel, Switzerland (https://www.unine.ch/geographie/climig_database), which includes research in several languages.

6. The chapter also draws on background papers commissioned for this report, including McLeman (2016) and Hunter (2016).

Understanding the Climate Change–Migration Nexus

18 Groundswell: Preparing for Internal Climate Migration

TRENDS AND PATTERNS OF INTERNAL MIGRATION

Migration within countries generally entails people moving between administrative units within their country. The United Nations Development Programme (UNDP 2009) estimates that there were 740 million internal migrants globally at the turn of the millennium—three times the number of international migrants.7 Bell and Charles-Edwards (2013) estimate that in 2005, 763 million people were internal migrants. It is likely that the estimates were conservative and the number is higher today. In China alone, some 150 million migrants from rural areas were living in urban areas in 2009 (Chan 2013). Data gaps and methodological issues make it hard to estimate the current level of internal migration worldwide.

Internal migration shows significant heterogeneity across and within countries (Bell and Charles-Edwards 2013). For example, internal migration intensities—the total number of internal migrants as a share of the overall population —have fallen in Latin America but increased in China and some African countries. Part of this heterogeneity may be definitional: The smaller the administrative unit considered, the higher the rates of internal migration, because smaller units capture more of the overall movement of people. In 1995–2000, for example, internal migration intensities were as low as 0.8 percent in Indonesia (in eight provinces) and as high as 20 percent in New Zealand (in 75 territorial authorities) (Bell and Charles- Edwards 2013).

A significant proportion of internal migrants will end up in major cities. Internal migration is a large contributor to urbanization (Tacoli and others 2014). The share of the population living in urban areas has risen in almost all countries. It averages 32 percent, 49 percent, and 78 percent, respectively, in least developed, less developed, and developed countries (PRB 2016; UNDESA 2014). About 40 percent of the urban growth rate in low and middle income countries is related to migration (Montgomery 2008), although the pattern is not uniform across countries: Migration represents less than one-third of urban population growth in Sub-Saharan Africa but far more in Asia (Tacoli and others 2014).

This report focuses on internal climate migration but underlines that all forms of movements induced by climate change need to be considered and addressed holistically. For context, Box 2.1 on the next page briefly outlines patterns of international movements of people and linkages with climate change.

7. This estimate is conservative, based on census data and counting only movements across the largest administrative units in a country (usually states or provinces). Calculation of the number of internal migrants used census data from a sample of 24 countries covering 57 percent of the world’s population. It applied the regional patterns found in these data to estimate the number of about 740 million internal migrants in the world.

19Understanding the Climate Change–Migration Nexus

Box 2.1: Cross-border migration and its linkages with climate change

About nine million people a year migrate between countries (Abel and Sander 2014), and about 244–250 million people live outside their country of birth (UNDESA 2015; World Bank 2017a). Although absolute numbers are higher today than in the past, international migration has not changed greatly in percentage terms since 1990 (Abel and Sander 2014).

South–South migration accounts for more than 90 million migrants between countries in the global South, closely followed by South–North routes with about 85 million migrants (IOM 2015). A majority of international migration is to areas of economic opportunities: More than 60 percent of international migrants (150 million people) are classified as migrant workers (ILO 2015). Family relations, social networks, and many other factors also influence the motivations and direction of international migration.

Similar to internal climate migration, there is no single straight line from climate change impacts to international migration, and the economic, political, and social obstacles to such movement are often considerable (e.g., Czaika and de Haas 2013; Foresight 2011). Results from a systematic review of the small but growing body of peer-reviewed literature on the subject show that climate change may increase international migration, especially as it plays out through impacts on natural resource- based livelihoods. Yet it can also inhibit cross-border movements (Obokata, Veronis, and McLeman 2014). Nawrotzki and Bakhtsiyarava (2017), for example, find that excessive rain increases international migration from Senegal but that heat waves decrease it from Burkina Faso.

While current numbers of international climate migration are probably not large (Obokata, Veronis, and McLeman 2014), it is unclear what the future of international climate migration will look like. Cross-border movements provide opportunities for people affected by climate impacts but can also entail challenges (such as for health systems, see Watts and others 2017), especially when movements are large or rapid, or border spaces are sensitive (Foresight 2011). For example, competition over scarce resources in receiving areas can create tensions (Schleussner and others 2016; United Nations and World Bank 2017). Reuveny’s (2007) meta-review of 38 cases of migration induced by environmental change finds that factors influencing conflict included competition over resources, with ethnic tensions and distrust acting as exacerbating factors. Environmental changes can intensify tensions in regions where population growth and pressure on natural resources will increase competition between migrants and existing populations over access to arable land, water, and pasture (Marc and others 2015).

This highlights the need to take a holistic policy and planning approach that considers all forms of mobility in the context of climate change. Cross-border approaches integrated in anticipatory planning mechanisms can help. This includes preparing cross-border areas through tailored strategies to minimize risks (Warnecke, Tänzler, and Vollmer 2010). Participatory and inclusive resource management mechanisms can also reduce the potential for additional environmental stress (Martin 2005).

THEORIES AND DRIVERS OF MIGRATION

Theories of migration provide complementary insights into a complex decision-making process. This section focuses on three schools of thought—classical economic approaches, push-pull theories, and the New Economics of Labor Migration—to illustrate the factors involved when people decide to migrate (see Massey and others 1993 for a good overview of other theories).

Most of the migration literature since the 1950s has been based on classical macroeconomic and microeconomic theories. Migrants are assumed to be opportunity-seekers who weigh the financial costs and benefits of potential destinations and make rational decisions based on the information available to them (see, for example, Todaro 1969, 1980). Classical economic approaches assume that migration typically flows from low-wage to high-wage labor markets and from low-opportunity to high-opportunity locales.

Other factors also affect decisions to migrate and the capability to carry them out. People seldom have perfect information. They are rational actors only to a degree, and decisions are not necessarily made just by the individual. The poorest people often have insufficient capabilities to access high-wage or high- opportunity markets.

20 Groundswell: Preparing for Internal Climate Migration

Going beyond the idea of individual cost-benefit analysis, the influential push-pull theory suggests that four factors shape migration: factors in the sending area, factors in the destination areas, “intervening” facilitators and obstacles, and personal factors (Lee 1966). Push factors (such as poverty, lack of opportunities, and migration networks) are associated with source areas. Pull factors (such as jobs, social services, family members, and social networks) are related to destination areas. The push-and-pull framework suggests that individual characteristics of affected populations have an important influence on migration decisions. Personal factors include age, gender, and ethnicity (Lee 1966; Geddes and others 2012; De Jong and Gardner 2013; Zickgraf and others 2016). Migration studies suggest, for instance, that younger people are more mobile (Hatton and Williamson 2005). Ethnicity, wealth, and gender all affect migration outcomes within populations similarly affected by climate change (Adger and others 2014).

Push-pull theories also show that people are rarely able to migrate wherever they wish. Movements can be restricted by natural geographic features (such as rivers and mountains) but also by entry requirements, travel costs, and sometimes physical obstacles (such as fences), particularly for international destinations or movement over long distances (Czaika and de Haas 2013). Barriers to internal migration also exist. For instance, concerns over loss of property rights in destination areas and other bureaucratic constraints prevent many individuals who would benefit from moving from doing so (Waldinger 2015). Tight financial resources and the paucity of information also limit potential migration, as does access to social networks to draw on for support (Palloni and others 2001).

Push-pull theories help explain why large urban areas, with their greater range of economic opportunities and better services, are often preferred destinations for migrants from rural areas and smaller towns and villages. They also show that alongside factors in sending and receiving areas, intervening and personal factors are important.

The New Economics of Labor Migration (NELM) suggests that not all migration decisions can be explained by individuals who weigh economic opportunities or are pushed from origin and pulled toward destination areas. Through surveys, it has demonstrated that migration decisions are often part of a household’s collective strategy to diversify its income sources and reduce its exposure to hardships and risks (e.g. Stark and Bloom 1985). In some dryland regions of West Africa, for example, households send young adults to cities to seek wages in the dry season to reduce the number of mouths to feed and benefit from remittances (Rain 1999). Household members may also migrate indefinitely, on the understanding that the migrants will repay the costs and absence from the household workforce by remitting money home (Mazzucato 2011). NELM emphasizes the importance of larger households in migration decision making and shows that migration may be related to risk reduction and livelihood diversification as much as economic opportunity-seeking. All these factors are important in understanding migration in the context of climate change, which is often understood to be a livelihood, coping, risk management, or adaptation strategy by households (Adger and others 2014).

These theories of migration mostly focus on more voluntary instances of movements, but mobility occurs on a continuum, from voluntary to forced migration and displacement (Hugo 1996). While not within the modeling scope of this report, Box 2.2 provides context on the dynamics of disaster displacement.

21Understanding the Climate Change–Migration Nexus

Box 2.2: Spotlight on disaster displacement

For people displaced by extreme events or conflict, both staying and leaving carry high risks; either choice may threaten survival (World Bank 2017a). Displacement can be said to be a function of the likelihood, severity, and nature of the hazard; the exposure of people; and preexisting vulnerabilities (IDMC 2015). It tends to emphasize “push” more than “pull” factors: People make decisions, often quickly and under duress, based on what they perceive to be the optimal coping strategy at a particular time. Violence is the strongest correlate of decisions to flee (World Bank 2017a).

Different types of climate-related hazards tend to generate different types of mobility. Key factors are the speed of onset of a climate hazard and the ability of the people affected to perceive the risk and adapt in anticipatory fashion (Renaud and others 2011). Rapid-onset events, such as extreme storms and floods, usually result in short-term displacement (followed by return to affected areas) (Kälin 2010; Brzoska and Fröhlich 2015; Black and others 2011). Such events also tend to generate migration “churn”—a mix of shorter- and longer-term displacement and out-migration and, potentially, sudden influxes of migrants from unaffected areas, such as reconstruction workers (De Waard, Curtis, and Fussell 2016). By contrast, slow- onset events, such as droughts, tend not to generate immediate changes in migration patterns, at least not initially (McLeman 2014). Drought-related changes in migration patterns, for instance, are more likely to emerge in areas where land degradation is severe (Neumann and Hilderink 2015) or drought is less common, such as Bangladesh (Gray and Mueller 2012a). Longer- term trends in crop yields or water availability are likely to contribute to out-migration (Nawrotzki and others 2017; Bohra- Mishra, Oppenheimer and Hsiang 2014). That said, the impact of successive climate shocks may erode household assets and therefore adaptive capacity in ways that can eventually influence decisions to migrate (IDMC 2016; Warner and others 2012), or even trap people in place (Black, Kniveton, and Schmidt-Verkerk 2013).

Displacement as a result of extreme and rapid-onset events, such as floods and storms, fluctuates from year to year but continues to be high. Between 2008 and 2015, the Internal Displacement Monitoring Center (IDMC) recorded an annual average of 26.2 million displacements caused by natural disasters. In 2015, about 19.2 million people in 113 countries were displaced by disasters caused by rapid-onset extreme events such as floods (8.3 million) and storms (6.3 million) —twice the number caused by conflict and violence. In 2016, that number was 24.2 million, with South and East Asia most affected (IDMC 2017).

According to the IDMC, disaster displacement has consistently affected poorer countries the most; low and middle income countries account for 95 percent of the global total. Smaller territories, such as Small Island Developing States (SIDS), suffer disproportionately. SIDS as diverse as Fiji and Tonga in the Pacific and Belize, Cuba, and Haiti in the Caribbean were among the 10 countries with the largest disaster displacements relative to population (IDMC 2014, 2016, 2017).

Many poor people have limited resources to confront disasters, and their livelihoods depend directly on increasingly threatened ecosystem goods and services (Hallegatte and others 2017). As a result, the poorest often bear climate impacts disproportionately (World Bank 2012a), with their well-being losses (equivalent to consumption) more than twice as large as the world’s average (Hallegatte and others 2017). The needs of displaced people are often far more pressing than those of other migrants, stemming from their lack of land, jobs, and homes; marginalization; food insecurity; and higher morbidity and mortality.

Disasters through rapid-onset events already displace large numbers of people, and the numbers are expected to rise as such events become more frequent and severe because of climate change (Adger and others 2014; IDMC 2015, 2017). The impacts of disaster displacement will be added to those of climate migration. Although people often wish to return to their homes after a time, some remain in protracted situations for years or become displaced several times (IDMC 2017), underscoring the fact that all forms of mobility in the context of climate change require concerted and comprehensive planning and action.

22 Groundswell: Preparing for Internal Climate Migration

Photo Credit: Shutterstock

23Understanding the Climate Change–Migration Nexus

CLIMATE CHANGE AS A DRIVER OF MIGRATION

The decision to migrate is complex. The influencing factors change over time and vary across individuals and their larger households. Climate change adds complexity to the already difficult decision to stay in or leave one’s home. Understanding of four elements—the impacts of climate change on resource quality and availability; the implications for livelihood strategies in particular ecological regions; the role migration plays in supporting livelihoods; and how migration patterns may shift as people attempt to adapt to the changes they experience—is necessary. This section explores the ways in which climate change shapes livelihood activities, including migration decisions.

The Foresight Report (2011) developed a framework of climate migration that builds on theories and empirical studies on migration. It emphasizes that climate change influences migration decisions through existing migration drivers, especially economic, environmental, and to some extent political drivers—by, for instance, depressing rural wages, raising agricultural prices, shaping exposure to hazards, and stressing ecosystems. The framework conceptualizes environmental change as affecting macro, meso (intermediate) and micro levels of migration choices and capabilities. Macro-level influences are beyond the control of individuals, households, or communities but have ongoing direct effects on livelihoods, incomes, and well- being; they shape migration possibilities and opportunities. Macro-level factors include demographic (for example, the size of the population and the rate of growth); economic (for example, labor markets and commodity prices); political (for example, laws and conflict); social (for example, gender relations); and environmental variables. At the meso level, intervening facilitators and obstacles include employment agencies or migrant-smuggling networks. Micro-level factors include characteristics of individuals and households (for example, health, skills, and perceptions of risks). The migration decision is not a simple binary choice, but is also based on behavioral elements (Martin and others 2014). Such elements include the household’s perceptions of its vulnerability to climate change–related risks, its assessment of its future needs, household members’ beliefs and norms, and their experience (if any) of migration.

People can decide to migrate as a livelihood strategy when climate change affects overarching variables such as the economy, environment, and political system they live in. Climate change has the most direct impact on macro-level factors, especially through economic channels, but affects variables at all three levels. It can act as a push or pull factor by altering the relative attractiveness of locations. It can also affect inhibitors or facilitators of migration and the accessibility of places, or influence the livelihood capital of individuals and households.

Livelihoods depend on access to capital of various types, the quantities and qualities of which vary by geographic region and livelihood system. Scoones (1998) identifies the following forms of livelihood capital: natural (ecological goods and services); economic or financial; human (skills, knowledge, physical health and abilities); and social (benefits and opportunities that arise through social relationships). Climate change can affect livelihoods directly and indirectly, sometimes concurrently. Although the direct physical impacts of climate change may be the most visible, the less visible indirect effects can be just as pernicious (Gemenne, 2011). In Figure 2.1 the impacts of climate change represented by the orange arrows are not uniform across space or over time.

24 Groundswell: Preparing for Internal Climate Migration

CLIMATE CHANGE IMPACTS ON LIVELIHOODS AND MIGRATION PATTERNS THROUGH 2050

Building on the framework in Figure 2.1, this section highlights climate changes that are likely to have large impacts on livelihoods and migration patterns through 2050. The data draw on the IPCC Working Group I and II contribution to the Fifth Assessment Report and on peer-reviewed literature.

Climate change will influence migration through warming and drought, which will affect agricultural production and access to water; rising sea level, which makes coastal areas and island states uninhabitable; the increasing intensity and frequency of natural disasters; and competition over natural resources, which may contribute to drivers of conflict. The first three points are discussed in Box 2.3 below, and Box 2.4 reviews the climate change-conflict-migration nexus.

Figure 2.1: Foresight model adapted to illustrate climate change, livelihoods, and household migration behaviour

Source: Extended and adapted from Foresight (2011) and McLeman (2016)

Migrate

Household needs Risks perception

Norms, beliefs, trustPast experience

DECISION

Human Capital

Social Capital

Financial Capital

Physical Capital

Natural Capital

Political

Demographic

Economic

Social

Environmental

Political/legal frameworks

Cost of moving

Social networks

Recruitment agencies

Diasporic links

MACRO-LEVEL FACTORS

MICRO-LEVEL FACTORS

Personal/household characteristics

MESO-LEVEL FACTORS

Intervening obstacles and facilitators

Stay

C L

IM A

TE C

H A

N G

E

25Understanding the Climate Change–Migration Nexus

Box 2.3: How does climate change affect people and their livelihoods?

Four aspects of climate change affecting people and their livelihoods are reviewed here: temperature, water availability, sea level rise, and extreme events.

Temperature

By 2050, average global temperatures are expected to increase by between 0.3°C (the lower boundary [RCP2.6]) and 2.5°C (the upper boundary [RCP8.5]). Additional increases for the remainder of the century are projected to range from 0.3°C to 4.8°C, depending on future emissions (Box Figure B2.3.1).

Figure B2.3.1 Past and projected global average surface temperature change above the 1986–2005 baseline

Source: Adapted from IPCC (2013a). Note: See Chapter 3 for explanation of the Representative Concentration Pathways (RCPs), which are trajectories of greenhouse gas concentrations resulting from human activity, and modeling input of RCPs used in this report from 2020-50. The numbers of CMIP5 models used to calculate the multi-model mean is indicated.

Extreme heat events are occurring more frequently. Unusual and unprecedented heat extremes affect livelihoods primarily through their effect on agricultural yields and food security. Significant crop yield impacts are already being felt; if temperatures rise toward 4°C higher than the 1986–2005 baseline, climate change will add further pressure on agricultural systems. Crop yield decreases associated with a 2°C increase in temperatures relative to that baseline are expected to be as large as 50 percent in Central America and the Caribbean. Ecosystem shifts are projected with increasing temperatures. When coupled with changes in precipitation patterns, they greatly diminish ecosystem services. Substantial adverse effects on marine ecosystems and their productivity are also expected with rising temperatures, increases in ocean acidity, and likely reductions in available oxygen as a result of their combined effects. With increased warming by 2050, and without taking into account changes in ocean acidity, fishery catches in several locations are projected to decrease markedly as fish populations migrate toward cooler waters.

Water availability

Freshwater resources are at risk from climate change in several regions. The impact of climate change on freshwater resources is felt through changes in rainfall patterns, increasing temperatures, and increased evapotranspiration (Abtew and Melesse 2013). In Central America, some central parts of South America, and southern central Asia, overall surface water availability is, with relatively high confidence, projected to decline substantially (Schewe and others 2014). A risk of decreasing water availability also exists, albeit with somewhat lower confidence, for Central and Southern Africa. Models also project increased water availability for India and the Horn of Africa, but with lower confidence. In many of these regions, population growth and rising incomes will have a greater impact on water resource availability than climate change; the two factors may yield declining per capita supplies of water even in areas projected to have higher rainfall (World Bank 2016).

6.0

4.0

2.0

−2.0

0.0

(o C)

42 32

39

historical RCP2.6 RCP8.5

Global average surface temperature change

RC P2

.6

RC P8

.5

Mean over 2081–2100

1950 2000 2050 2100

Period covered in this report

26 Groundswell: Preparing for Internal Climate Migration

Box 2.3: How does climate change affect people and their livelihoods? continued

Regions that are already dry can expect to become dryer and experience greater frequency or intensity of extreme heat events (Trenberth and others 2014). This pattern is likely to affect Ethiopia and Mexico, which have extensive drylands (see Chapter 5). Climate change is likely to undermine food security through decreased crop yields and increasing risk of crop failure caused by drought.

Climate change is also diminishing winter snowpack and leading to the retreat of glaciers. Snowpack and glaciers act as buffers between precipitation and runoff. Their retreat will affect the magnitude and shift the seasonality of river discharge in some of the most populous river basins on earth (Arnell and Gosling 2013; Bozkurt and Sen 2013). Major rivers such as the Ganges, Indus, and Brahmaputra are highly susceptible to glacier melt and reductions in snowfall, but are also densely populated (World Bank 2013). These three river basins provide water to about 750 million people, indicating immense climate vulnerability in these areas (Immerzeel, Van Beek, and Bierkens 2010). Already a substantial loss of glacier volume and extent has been observed in the Andes and Central Asia. Since the 1960s, the surface area of Central Asian glaciers has fallen by 3–14 percent, depending on location; losses of about 50 percent and up to 80 percent are projected, respectively, for a world 2°C and 4°C warmer than the 1986–2005 baseline (World Bank 2014). Increasing glacial melt poses a high risk of flooding, severely reduces freshwater resources during crop-growing seasons, and can have a negative impact on hydropower supply.

Sea level rise Changes in sea level—currently averaging three millimeters a year and increasing (IPCC 2013b; Hay and others 2015; Dieng and others 2017)—will be the most problematic for settlements on very low-lying land (within one meter of sea level), especially areas that experience subsidence, erosion, or tropical cyclones and extreme storm events. Sea level rise can salinize groundwater and surface freshwater bodies. It can also increase salt-water intrusion into freshwater aquifers (particularly in the Middle East and North Africa), a process made worse by other climate impacts (such as reduced water availability). Together, these impacts present substantial problems to agriculture and human consumption.

Sea level rise poses a particular threat to large urban settlements and major infrastructure along the coast, particularly in Small Island Developing States (SIDS), where the ability to retreat to higher ground is limited. Rising sea level will greatly increase risks from storm surges and tropical cyclones, in particular for highly exposed SIDS and low-lying coastal zones. In the country examples in Chapter 5, large areas of Bangladesh are vulnerable.

Extreme events All populations should expect increasing variability in climatic norms and conditions and be prepared for abrupt, nonlinear changes (Streets and Glantz 2000; Schneider 2004). Areas that already experience extreme storm events, especially in mid- latitudes and the wet tropics, should expect to experience as great or greater frequencies of such events, with the potential for more high-intensity events through 2050 (IPCC 2013b). The intensity of extreme storms is likely to increase, creating greater risks of flooding and damage to homes and coastal infrastructure (Chan and others 2012). The greater prevalence of extreme events will affect rural and urban communities, particularly on sloping lands and in coastal regions. Bangladesh falls into this category. SIDS face some of the highest levels of exposure and vulnerability to extreme storm events (World Bank 2013).

This report models climate impacts on internal migration through water availability and crop productivity, as well as sea level and storm surge. The literature shows that water scarcity and declining yields, along with sea level rise, are among the major impacts facing low income countries, and that these impacts will be important drivers of migration (see Henry, Schoumaker, and Beauchemin 2004; Feng, Krueger, and Oppenheimer 2010; Nawrotzki, Riosmena, and Hunter 2013; and Chapter 3). Water deficits can also have wider impacts on the economy, as households curtail spending and agricultural-processing industries and other businesses retrench. There is less evidence of what may happen to regions that experience increases in rainfall over time. Households may send one or more members to cities or other rural areas in search of an alternative livelihood, or they may abandon farming or other rural livelihoods altogether, as has happened in rapidly urbanizing countries such as China. The literature suggests that extremes are more likely to result in short-term displacement than sustained migration (Kälin 2010; Black and others 2011; Brzoska and Fröhlich 2015).

27Understanding the Climate Change–Migration Nexus

Box 2.4: The climate–conflict–migration nexus

The review of the empirical evidence in the latest IPCC report concludes that climate change can affect known drivers of conflict, such as unstable institutions, but finds little agreement on direct causal pathways or a strong relationship with armed conflict (Adger and others 2014). There is some agreement that existing patterns of conflict could be reinforced under climate change—in, for example, already fragile regions or areas with ethnical divides (Schleussner and others 2016; Buhaug 2016). Major security actors such as the U.S. Department of Defense (2015) have tended to frame climate change as a “threat multiplier,” which can exacerbate existing risks. Nonetheless, the world’s development and sustainability trajectory significantly influences how climate will influence conflict drivers and risks (Hegre and others 2016).

Conflict is both a process and an inherent feature of human interaction, with a raft of expressions, stages, locations, arenas, and impacts (Galtung 2008). However, its emergence and evolution are always “the result of an individual context-specific mixture of interconnected factors,” (Schleussner and others 2016, 9216). Climate change is hypothesized to drive or exacerbate conflicts through, for example, pressure on resources, natural disasters, and sea level rise. These factors can threaten livelihoods, increase competition, intensify cleavages, reduce state capability and legitimacy, trigger poorly designed climate action with unintended consequences, and lead to large movements of people that may have negative impacts in receiving areas (Buhaug, Gleditsch, and Thiesen 2010).

Some qualitative studies examine the influence of environmental stress on specific conflicts, such as insurgencies in Assam in India (Burrows and Kinney 2016; Homer-Dixon 2010). Cross-case quantitative studies find significant statistical correlations between climate change and violence or conflict (Burke, Hsiang, and Miguel 2015; Carleton, Hsiang, and Burke 2016; Mares and Moffett 2016). They posit that if human responses to climate change remain unchanged, climate change has the potential to increase violence and conflict.

This large-scale quantitative work has received criticism with respect to the definition of conflict, sample selection, statistical methods, and lack of explanation of causal mechanisms at work (Buhaug and others 2014; Buhaug 2010, 2014; Selby 2014). Other empirical work and quantitative analyses challenge their findings (Buhaug 2010; O’Loughlin and others 2012; Bergholt and Lujala 2012; Slettebak 2012; Klomp and Bulte 2013; Bernauer, Böhmelt, and Koubi 2012). These analyses argue that adaptation capacity, institutions, and existing vulnerabilities mediate the effects of climate change on factors that may drive conflict, and that other factors than environmental ones are often more important in igniting conflict.

For climate impacts and their influence on migration, understanding uncertainty, variability, thresholds and “tipping points” is key. Most regions will become hotter and drier. There is uncertainty as to the specific manifestations of climate change, however, partly because natural variability will remain the dominant factor in observed climate trends in the near term and partly because the period through 2050 represents a transition during which anthropogenic changes to the atmosphere become a more dominant influence on climate systems (IPCC 2013b). The spatial and temporal variability of the impacts of climate change make it difficult to have precise predictions of the frequency or extent to which specific areas, countries, or regions will experience particular climate risks or hazards. Chapter 3 explains how this uncertainty affects the modeling effort.

28 Groundswell: Preparing for Internal Climate Migration

A tipping point is a particular moment at which a component of the earth’s system enters into a qualitatively different mode of operation, as a result of a small perturbation (Levermann and others. 2012; Schellnhuber 2009). Abrupt climate change occurs when the system crosses this tipping point, triggering a transition to a new state at a faster rate (Michaelowa 2002). Tipping points exist because of nonlinearity—the fact that there is no simple proportional relationship between cause and effect (IPCC 2013b). Systems such as the Arctic summer sea ice, the Greenland ice sheet, the El Niño-Southern Oscillation, and the Amazon rainforest may reach their tipping points at different times (Lenton and others 2008; World Bank 2013). For instance, in about 50 years, the tipping point for the Amazon rainforest may be triggered by changing precipitation and dry seasonal length that would make it difficult for the rainforest to reestablish itself, resulting in substantial biodiversity loss and further decreases in rainfall (Lenton and others 2008). This report’s modeling cannot capture such relationships. Their existence may rapidly affect the scale of climate migration after a system reaches a tipping point. An indication of the approach of a tipping point, or rapid movement toward an observed change in climate, livelihood, or climate migration trend, underscores the urgency to scale up climate change mitigation and adaptation efforts now.

Independently of uncertainties associated with climate change projections, and potential future tipping points, evidence shows that climate migration is already taking place (Ionesco, Gemenne, and Zickgraf 2015). There is also consensus that the future may see the emergence of areas where combinations of exposure, sensitivity, and adaptive capacity put whole communities at risk (de Sherbinin 2014). People have moved and will continue to move toward zones of environmental risk, such as coastal cities (de Sherbinin and others 2012; Geddes and others 2012). Changed climate patterns will expand patterns of cyclical, seasonal, and rural-to-urban migration (Ammer and others 2016). The largest form of movement will be internal; when people cross borders, most will move to neighboring countries, often from one low-income country to another (Foresight 2011; McLeman and others 2016; Groschl and Steinwachs 2016). Table 2.1 presents the findings of a systematic review of 160 peer-reviewed publications on climate migration.

Table 2.1: Summary of research findings on the interconnectivity between climate change, livelihoods and migration

Finding Characteristics/implications Strength* of evidence

Migration is a key component of sustainable livelihoods and household adaptive capacity in low and middle income countries

Migration is one of many possible ways in which households adapt to and cope with climatic and nonclimatic risks/ uncertainty.

In countries with weak institutions, migration may be the only form of adaptation available.

Households that lack migration options are inherently more vulnerable and less adaptable to the impacts of climatic variability and change.

Broad base of evidence from wide range of empirical studies in multiple regions

Reliance on migration to meet rural livelihood needs is growing

Growing reliance on migration is a common trend across less developed countries.

Seasonal migration is common in regions with highly seasonal climates.

Migration for longer durations is becoming increasingly common.

Most migrants are young adults.

Broad base of evidence from wide range of empirical studies in multiple regions

Most climate-related migration takes place across short distances within countries or across contiguous borders

Local migration has relatively low costs.

The destination may be urban or rural, depending on wage- earning opportunities.

Broad base of evidence from wide range of empirical studies in multiple regions

29Understanding the Climate Change–Migration Nexus

Finding Characteristics/implications Strength* of evidence

Households that receive remittances from migrants have greater long-term social and economic prospects

Remittances from international migrants are typically of greater value than remittances received from internal migrants.

Within communities, remittances increase socioeconomic inequality.

Remittances can help improve prospects for disaster recovery and to some extent for preparedness and adaptation, but the overall contribution to building climate resiliency is uncertain.

Broad base of evidence from wide range of empirical studies in multiple regions

Rural to urban migration rates are high and growing

Rural to urban migration rates are growing especially rapidly in Sub-Saharan Africa, although that region is less urbanized than others.

Conclusive statistical evidence

Households that participate in rural to urban resource- sharing networks have greater food security

Such networks are likely to grow in importance as impacts of climate change on food production systems strengthen.

Households lacking access to such networks are more vulnerable.

Strong empirical evidence from Sub-Saharan Africa and South America

Migration is a key means of recovering from extreme weather events, including floods and droughts

After an extreme event, households send young adults to seek wage labor opportunities to rebuild lost/damaged housing and livelihood assets.

The duration and destination of migration vary by context.

Strong empirical evidence from Central America, East Africa, and South and Southeast Asia

People in remote areas have worse migration opportunities and adaptation prospects

Remoteness and isolation are important factors in the vulnerability of mountain populations and some dryland areas.

People living in areas with good access to roads, markets, and social infrastructure have a greater range of adaptation options and potential migration destinations.

Strong empirical evidence from dryland areas of Africa, the Andes, and Nepal

Gender dimensions can change over time, but have important implications for climate migration

Where only men migrate, women, children, people with disabilities, and the elderly left behind are at greater risk of food insecurity and personal safety.

Land degradation and climatic variability can force higher levels of gendered migration or longer-duration migration.

Gender dimensions can change over time.

Case study evidence from Bangladesh, Nepal, and Pakistan; more research needed in other countries and regions

Empirical research shows persuasively that climate resilience must be built urgently given the known risks, but there remains uncertainty about which strategies hold the greatest promise and how to implement them

The impacts of extreme heat, dryness, and variability in precipitation on regional migration patterns will grow through 2050 but can be moderated by sound development strategies.

Climate impacts on food systems, water resources, and livelihoods accelerate in nonlinear fashion after 2050, as will the risks of large-scale displacements and distress migration.

Business as usual will fall far short of meeting future adaptation needs.

Climate model evidence increasingly strong

Models of future crop yields vary but decline for several crops

Systems understandings of general climate–migration dynamics are increasingly strong, with strong agreement across the research literature

Further empirical research on resilience-building needed across most regions

Source: McLeman (2016). * Green = strong evidence few uncertainties. Yellow = reasonably strong evidence, uncertainties in some regions or areas.

30 Groundswell: Preparing for Internal Climate Migration

POSSIBLE INFLUENCES OF CLIMATE IMPACTS ON MIGRATION IN SPECIFIC ECOSYSTEMS

This section explores what the literature says about possible future climate migration for key ecosystems. The findings are based on a systematic review of the literature conducted in a background paper for this report (McLeman 2016).

Migration patterns will change across ecosystems when facing climate change impacts alongside other stressors such as deforestation, land degradation, and biodiversity loss. In drylands, labor migration and rural to urban migration can be expected to increase, and extreme events will probably lead to more migration under distress. In tropical and temperate forest regions, churning migration patterns in forest frontier areas are likely, as are growing rates of young adult labor migration out of more established farming areas. In coastal zones, by midcentury there may be increased rates of rural to urban migration, temporary and indefinite, as well as displacement from smaller atolls and the seaward edge of deltas as a result of erosion and salinization. For montane regions, though case-dependent, further increases in already high rates of rural to urban migration and to international destinations are projected. Smallholder cropping regions can expect increased rates of rural to urban migration across all regions, both temporary and indefinite (Table 2.2).

Table 2.2: Overview of climate change and macro trends on livelihoods in key ecosystems, and possible migration trends by 2050

Main climatic influences on migration by ecosystem type

Other macro trends Impact on livelihoods Possible migration trends

Strength* of evidence

Drylands

Extreme heat events, droughts, dryness, and precipitation variability

Growing intensification and market-orientation of agriculture

Land degradation, water scarcity, and depletion of soil nutrients

Increasing enclosures and land grabs

Pastoralism declines

Specialized livestock pasturing increases

Average farm size shrinks

Small-scale farms lack capital to intensify

Outside interests acquire farms on best, most accessible land

Labor migration increases

Rural to urban migration increases

Extreme events increase distress migration

Very strong case-based evidence from range of Sub-Saharan African countries

Smaller range of studies for dryland migration in other regions

Tropical and temperate forest regions

Increased extreme heat events and precipitation variability, which may increase fires and adversely affect forest health

High rates of deforestation in the Americas, Central Africa, and Southeast Asia

Loss of biodiversity

Land degradation

Smallholder farmers are key drivers of forest change in Africa

Commercial enterprises are key drivers of forest change in Americas and Asia

Indigenous and customary forest users squeezed out

Small farms created, bought, and sold in dynamic process

Larger, intensive farm operators acquire best lands

Churning migration patterns appear in forest frontier areas

Rates of young adult labor migration out of more established farming areas increase

Strong case-based evidence of migration processes in Americas and Southeast Asia

More evidence from Africa desirable

Considerable variability in expected impacts of climate change across regions

31Understanding the Climate Change–Migration Nexus

Main climatic influences on migration by ecosystem type

Other macro trends Impact on livelihoods Possible migration trends

Strength* of evidence

Coastal zones

Sea level rise and increased intensity of storms, causing floods, erosion, soil salinization, aquifer salinization; warming ocean temperatures, affecting reef health; possible increased variability of precipitation

Rapid urbanization and industrial development in deltaic regions of Asia

Loss of protective features (mangroves, marshes)

Expanding aquaculture across coastal Asia

Declining offshore fish stocks

Economic opportunity increasingly urban based

Small-scale fishing declines in many regions

Coastal farms under increasingly intensive cultivation

Some farms in Southeast Asia converted to shrimp aquaculture

Rates of rural to urban migration, temporary and indefinite, increase

Displacement from smaller atolls and seaward edge of deltas occurs, as a result of erosion and salinization

Strong statistical and case-based evidence for coastal states in Asia and Oceania; fewer data and cases for Africa

Montane regions

Increased precipitation variability and warming temperatures, leading to growing seasonal and interannual water scarcity; flash floods and landslides of increasing magnitudes; effects vary widely across regions because of inherent heterogeneity of mountain environments

Reinforced endemic poverty through remoteness, lack of physical and social infrastructure

Deforestation problematic around many mountain settlements

Pastoralism declines, but larger numbers of livestock are kept by less mobile farmers

Subsistence farmers lack food security

Livelihoods become highly diversified

Seasonal and longer- term migration becomes essential to livelihoods

Further increases in already high rates of rural to urban migration occur within mountain countries and to international destinations

Strong case-based evidence from Ecuador, India, Nepal, Pakistan, and Peru

Statistical evidence on migration weak for Nepal

Strong confidence in region-wide predictions, but local experience will vary considerably

Smallholder cropping regions

Increased frequency of extreme heat events, dryness, droughts in most regions; more intense extreme precipitation events in some regions; greater variability of precipitation in some regions; water scarcity in some regions

Global and regional population growth drives growing food demand

Greater pressure for intensification of production in all regions, especially Africa and South Asia, where yield gaps are greatest

Small-scale farmers, pastoralists increasingly squeezed off best land

Need to intensify production increases need for cash incomes, capital to invest in new technologies, crop varieties, irrigation

Some areas in Asia under irrigation may need to revert to rainfed agriculture, reducing household incomes

Productivity in some regions and of some crops may decline, rates to vary by region and crop type

Rural to urban migration temporary and indefinite, increase across all regions

Strong statistical evidence on current crop production

Models of future crop productivity display variability but are consistent across regions

Quantitative studies from several countries in Latin America, Asia, and Sub-Saharan Africa provide strong evidence of climate–rural income– migration linkages

Large array of case- based evidence from all continents

Source: McLeman (2016). * Green = strong evidence few uncertainties. Yellow = reasonably strong evidence, uncertainties in some regions or areas.

32 Groundswell: Preparing for Internal Climate Migration

Small Island Developing States are particularly vulnerable to climate impacts, making them an important location of future climate migration. These states are likely to experience more frequent and more severe storms, coastal erosion, and sea level rise (IPCC 2013b). Because of their small size, they could not be included at the resolution of the model applied in this report.

IMPACTS OF MIGRATION ON PEOPLE AND AREAS

Migration affects the well-being of people and surrounding systems in complex ways that evolve over time (UNDP 2009; Milan and others 2015; Melde, Laczko, and Gemenne 2017). Migration is a common coping, income diversification, risk management, and adaptation strategy for people facing economic stress and adverse climatic conditions (Ellis 2000; Adger and others 2014; Melde, Laczko, and Gemenne 2017). Climate migration is not good or bad in itself; the cost-benefit calculus is heavily dependent on perspective (Gemenne and Blocher 2017). Understanding the conditions under which positive or negative outcomes can result is necessary for managing the possible future trends outlined in this report in Chapters 4 and 5. As studies on the impacts of climate migration are still limited, this section also reviews evidence drawn from the general migration literature on systemic issues that arise from migration.

Impacts on migrants and their households

Outcomes of movements depend on the conditions under which people move (UNDP 2009). Climate extremes erode household livelihoods and assets (Zomers and others 2016; Warner and Van der Geest 2013), diminishing the capabilities needed for successful establishment in destination areas. Hugo (2009) argues that displaced people are more likely to find it difficult to adjust to their new destination for multiple reasons, connected with the unexpectedness of the move, their emotional state, and the unfamiliarity of their new surroundings.

People with greater capabilities—and therefore greater freedom to choose migration timing and destinations—often improve their economic situation after migrating. Harttgen and Klasen (2009) find that in 14 of 16 developing countries they studied, the Human Development Index was higher for internal migrants than for nonmigrants. For cross-border migrants moving from low- to high-income countries, a World Bank study (2017b, 15) finds on average “a 15-fold increase in income, a doubling of school enrollment rates, and a 16-fold reduction in child mortality.” Access to better health services, water quality, and sanitation at the destination, combined with a longer duration of stay, can have profound positive health impacts on migrants (UNDP 2009; Yabiku and others 2009). People who migrate to urban centers with better access to health services improve their chances of survival relative to rural residents, although their health outcomes are worse than those of urban nonmigrants (UNDP 2009).

Migration generally has positive impacts on households in sending areas, although those benefits are unevenly distributed. Respondents in surveys conducted through the Migration, Environment and Climate Change: Evidence for Policy (MECLEP) project, implemented by the International Organization for Migration (IOM) through a consortium of six universities, find migration to have generally positive, or neutral, impacts on the household economy. Only a very small proportion of respondents indicate that migration had negative impacts on income, employment, safety, and education (Figure 2.2).

33Understanding the Climate Change–Migration Nexus

Migration can be a positive adaptation strategy for people who migrate, but displacements challenge adaptation, and planned relocation shows mixed results. In Haiti seasonal migration helped reduce vulnerability. In all six countries studied by Melde, Laczko, and Gemenne (2017), migrant households reported mainly positive impacts of migration on income growth and diversification. Many also reported having learned, applied, and taught new skills through migration. Most migrant households also stated that they improved their health conditions and education thanks to migration. However, migrant households also experience more discrimination, exclusion from employment and access to important social services and systems than nonmigrants and are more likely to face insecurity; some also live in less robust houses than locals. Although these factors can undermine human development and adaptation prospects, the report finds that migration can increase adaptive capacities, especially if policy enables better access to social services, action against discrimination and exclusion, and protection of migrants.

Migration often generates remittances that can buffer environmental and economic shocks faced by household members who remain behind. Such transfers can increase and diversify assets and resources available to the sending household, allowing for more flexibility in livelihood strategies and less reliance on the local environment (World Bank 2006; Acosta 2007; Adams 2011; Acharya Leon-Gonzalez 2012; Le De, Gaillard, and Friesen 2013; Guadagno 2017). Remittances are often spent primarily on food, but better-off households also spend sizable shares on education, health care, and housing (UNDP 2009). Remittances are part of livelihood and insurance strategies in the event of sudden shocks and slow-onset disasters. In response to rainfall shocks in the Philippines, remittances decrease when the recipients’ income increases and rise when local income falls. About 60 percent of exogenous declines in income are replaced by remittance inflows from migrants overseas (Yang and Choi 2007). Remittances can help improve prospects for disaster recovery and to some extent preparedness and adaptation; the contribution to resilience is less clear (for example, Mohapatra, Joseph, and Ratha 2009; Joseph, Wodon, and Blankespoor 2014; Manandhar 2016; Banerjee and others 2017).

Figure 2.2: Perceived impact of migration on households according to a 2016 survey conducted by the International Organization for Migration

Source: IOM (2016).

Pe rc

en ta

ge o

f s ur

ve y

re sp

on de

nt s

PERCEIVED IMPACT

34 Groundswell: Preparing for Internal Climate Migration

Out-migration and remittances are not altogether beneficial for sending areas, however. Remittances may increase inequality, as the poorest households may be unable to send migrants—and even if they do, they may not be able to send remittances (Zickgraf and others 2016). Migration can also lead to remittance dependence and loss of intellectual capital (de Haas 2007), although the literature has cast doubt on the universal validity of the brain drain (de Haas 2010). In terms of environmental impacts, out-migration sometimes leads to a reduction in environmental pressure, through agricultural deintensification and land abandonment, because of the decline in the available labor force and the increase in cash income through remittances (Reichert 1981; Zimmerer 1993; Rudel and others 2005; Qin 2010). In Albania, Bolivia, and Mexico, out-migration is associated with an increase in vegetation cover (Preston, Macklin, and Warburton 1997; Lopez and others 2006; Muller and Sikor 2006). But out-migration can also weaken traditional forms of community-based natural resource management, as a result of declines in the human capital necessary to maintain it (Robson and Nayak 2010), and remittances can finance the use of environmentally destructive methods of natural resource extraction.

Impacts on migrant receiving areas

Migration has positive and negative impacts on receiving or destination areas. International in-migration replenishes the labor supply and skills, spurs entrepreneurship and innovation, eases strains on pension systems, and helps care for the elderly (KNOMAD 2017). However, it is frequently associated with localized population growth and attendant environmental impacts through, for example, changes in land use, deforestation, and land degradation (Starrs and Wright 1995; Hugo 1996; Bilsborrow 1992). Migration-related shifts in land use may alter the local environment, with negative impacts on health and forests. In the Brazilian Amazon, it can increase malaria risk (de Castro and others 2006). Environmental degradation is also associated with conflict, although the linkages are complex (see Box 2.4).

Photo Credit: Shutterstock

35Understanding the Climate Change–Migration Nexus

The effect of in-migration on the wages and employment of existing employees depends largely on whether migrants are substitutes for or complements of the existing labor force (Borjas, Freeman, and Katz 1997; Card 2001; Borjas 2003; Borjas 2006; Federman, Harrington, and Krynski 2006; Boustan, Fishback, and Kantor 2010). In many receiving areas, migrants may take “dirty, difficult, and dangerous” jobs, seasonal work, and informal sector work (IOM 2005). Some studies find no local wage effects on in- migration. Others find evidence of downward pressure (Kleemans and Magruder 2012; Maystadt, Mueller, and Sebastian 2016; McIntosh 2008). Generally, however, the impact of migration on destination areas is context dependent, based on the characteristics of the migrants, the receiving areas, and the larger overall structures, creating scope for policy entry points.

Urbanization is among the most important impacts of migration in destination areas. By 2050, it is expected that 70 percent of the world’s population will be urbanized (PRB 2016). Many rapidly growing cities have struggled with negative externalities, such as high unemployment, strained infrastructure, and environmental degradation (Bencivenga and Smith 1997; Beauchemin and Schoumaker 2005; Yin and others 2011; Kavzoglu 2008). By 2030, some two billion people—40 percent of urban residents—are expected to be living in slums (UNDP 2009). Most slum areas have inadequate access to safe water and sanitation; insecure land tenure; and inadequate waste management, transport, and electricity (de Sherbinin and others 2009; Connell and Lea 2002). Yet cities are often unfairly stigmatized; they may offer the best hope for improved conditions for migrants, as well as environmental conservation in rural areas (Marcotullio, Baptista, and de Sherbinin 2011).

Gendered impacts of migration patterns and impacts on vulnerable groups

This section draws from a review of the literature conducted in a background paper for this report (Hunter, 2016).

The impacts of shocks to ecosystems and livelihoods from climate change are not distributed evenly across population subgroups (Soares and others 2012). In many places, social groups in vulnerable situations—the poor and underprivileged, women, the elderly, children—are also the most severely affected by climate change impacts (for example, Reckien and others 2017). These subgroups are generally less able to cope with and respond to hazards or shocks because of their disadvantaged position: socially because of marginalized status; economically because they are poorer; and politically because of lack of independence, decision-making power, and underrepresentation (Gaillard 2010; Alston 2015). Beyond embodied vulnerabilities, the Office of the United Nations High Commissioner for Human Rights reports that vulnerabilities can arise from the circumstances that people on the move encounter in transit, at borders, and at reception or destination, particularly in lengthy and fragmented journeys.

Migration shows wide regional variation by gender. African countries and Mexico have seen increased migration by men in recent decades (Donato and Gabbacia 2015); other parts of the world have seen increases in migration by women. Asian countries have recorded increases in the proportion of female emigrants since 1960; emigrants from the Philippines include more women than men (IOM 2014). These streams are shaped by globalization, which has shifted labor demands and resulted in new migration “pulls” for women. The increase in dual-income households and aging populations in destination countries, for example, has fueled a rise in demand for care-giving services, for which women are most often sought.

Internal rural to urban migration patterns have led to shifts in traditional gender roles and some empowerment of women in their communities and households. In Vietnam, for example, women’s migration to urban areas has increased since the mid-1990s, influenced in part by the economic reforms of the late 1980s (Thao and Agergaard 2012). Rural women, particularly single women, are now heavily employed in light manufacturing, social work, and health care sectors in urban areas. Married female migrants are more likely to be employed in the informal sector. In their wives’ absence, husbands have taken on greater responsibilities for child care, housework, and agriculture. These changes in the division of labor have seen women gain influence in household purchasing and investment decisions, especially concerning remittances (World Bank 2012b). Where male household heads migrate for work, women may increasingly manage household finances. In agriculture, for example, male migration leaves many women

36 Groundswell: Preparing for Internal Climate Migration

as the de facto head of household, with more responsibility for overseeing and maintaining subsistence production for the family (Griener and Sakdapolrak 2013; Deere 2005; Radel and Schmook 2008).

A handful of studies offer insights on the effect of climate-related migration on gender issues. Gender norms often shape the ways in which men and women engage in agricultural and natural resource tasks; changes in these aspects of the local environment have gendered impacts, including on migration. In rural Nepal, for instance, men are responsible for gathering fuelwood and women for gathering fodder (Bohra-Mishra and Massey 2011). In these areas, environmental changes that make wood collection more time consuming are associated with local migration by men but not women; increases in the collection time for fodder are associated with increases in local migration by women but not men (Massey, Axinn, and Ghimire 2010).

Land tenure, tenure security, and productivity play roles in the gender–migration–climate change association. In the Loja province of the southern Ecuadorian Andes, for example, land-poor households are less likely than others to migrate abroad, although both women and men migrate internally, to diversify livelihoods. Land-rich households, in contrast, are more likely to send male but not female migrants abroad (Gray 2010).

Women and girls experience greater disadvantage than men from pervasive discrimination and structural inequalities in access to, and control of, resources. For instance, women with low socioeconomic status are more likely to get injured or die in natural disasters (Cannon 2002; Frankenberg and others 2011; Ikeda 1995; Hunter and others 2016; Paul 2010); greater socioeconomic gender equality mitigates this dynamic (Neumayer and Plümper 2007). Such inequalities also shape postdisaster experience. Women sometimes have less access to relief resources (Bradshaw 2004) and may be subject to additional workloads (Enarson, Fothergill, and Peek 2002). Women displaced by disasters are often at increased risk of sexual violence (Felten-Biermann 2006; Anastario, Shehab, and Lawry 2009) and mental health problems (Dell’Osso and others 2013; Viswanath and others 2013), in part because of exposure to violence (Anastario, Shehab, and Lawry 2009) and heavy caregiving burdens, which can increase anxiety and posttraumatic stress (Mills, Edmondson, and Park 2007).

But men and boys can also be in vulnerable situations (Demetriades and Esplen 2008). Gender norms for migration influence who is more likely to migrate when climate pressures intensify. Cultural norms that encourage male migration can put them in risky situations. Masculine gender norms such as those related to risk perceptions and risk-taking behavior can increase the vulnerability for men during natural disasters, as seen during Hurricane Mitch in Central America in 1998 (Bradshaw 2004; West and Orr 2007).

Photo Credit: nuttavut sam m

ongkol , Shutterstock

37Understanding the Climate Change–Migration Nexus

FIVE MAIN POINTS EMERGE FROM THIS CHAPTER

• Internal migration is already happening at scale and is expected to increase. • Migration decisions are the product of complex human decision making. • Climate change will affect multiple aspects of human livelihoods that feed into migration decisions. • The impacts of internal migration can be substantial in both sending and receiving areas. • Migration is an adaptation strategy and must be managed for both its opportunities and challenges.

Climate change affects livelihoods both directly and indirectly. People can decide to migrate as a livelihood strategy when climate change affects overarching variables such as the economy, environment, and political system they live in. Climate change can also affect inhibitors or facilitators of migration, and people’s natural, financial, human, and social capital. Ecosystems and associated livelihoods can shape current and future outcomes. The impacts of migration are often substantial; they need to be carefully investigated and managed within development and adaptation frameworks. Migration can be an important strategy for livelihood diversification and poverty reduction, and it is a major part of adaptation. At the same time, migration presents multiple challenges. For example, migrants and their families can face high risks and costs, such as discrimination and violence. Many migrants are also moving to areas that will be increasingly vulnerable. Remittances have the potential to create or increase inequalities in some cases. Furthermore, under certain conditions, in-migration can have adverse impacts on destination areas, including on social cohesion, infrastructure, and the environment.

Policy makers need to bear in mind the opportunities and challenges of internal climate migration when designing responses and long-term plans. Much can be done to improve the outcomes of migration (DRC 2009; UNDP 2009); bolster its adaptation potential (Melde, Laczko, and Gemenne 2017); and mitigate or prepare for climate impacts that lead to migration under distress, displacement, entrap people or require planned relocation. For example, integrated spatial planning that considers climate migration in terms of areas of destination or origin, as well as targeted interventions such as social protection measures and diversification of livelihoods, can help communities adapt in place or move in safety and dignity (see Chapter 6). People in vulnerable situations should be prioritized in responses and planning, as they will bear the brunt of climate change impacts. The gendered dimensions of climate migration also require special attention.