Climate Impacts as Drivers of Migration
Media outlets, think tanks, researchers, and advocacy groups are increasingly raising the specter that climate change will cause mass migration via its spiraling impacts on agriculture, water resources, and infrastructure, particularly in the developing world. More than just speculation about the future, however, environmental migration is already here. Concern over mass displacement was an undercurrent in talks leading to the 2016 Paris Agreement on climate, and analysts have identified changes to the climate as a major driver of migration from sub-Saharan Africa to Europe as well as a factor for some of the movements from Central America to the United States in recent years. But what is the basis for such claims, and how likely are scenarios in which ecological threats push hundreds of millions of people or more to relocate?
Historically, there is ample evidence that climate factors have played a significant role in population movements. Climate variability hastened migration following the decline of the Western Roman Empire in 476 AD and drought was at work during the downfall of the Mayan Empire (660 to 1000 AD). Both lack and abundance of precipitation can spur population movements, as demonstrated by the Mongolian expansion westward in the 13th century, which was fueled by wetter conditions on the Eurasian Steppe. In the last century, the American Dust Bowl of the 1930s and the Sahelian droughts of the 1970s and 1980s drove many to migrate, respectively, to California and to regional urban centers in countries such as Burkina Faso, Mali, Mauritania, and Niger. A common factor in many of these cases is that a period of relatively humid conditions was upended by significant declines in precipitation, meaning the land could no longer sustain the same population.
Special Issue: Climate Change and Migration
This article is part of a special series about climate change and migration.
Global climate patterns have changed over the last century, triggering more extreme weather events including hurricanes, heat waves, and droughts. Looking forward, the Intergovernmental Panel on Climate Change (IPCC) has projected that, at the current rate, global temperatures are likely to average 1.5°C above pre-industrial levels at some point between 2030 and 2052. Even this seemingly modest increase, which is well within the range of daily variability, will result in cascading impacts on ice sheets, ecosystems, and productive systems that will fundamentally alter habitability when spread over the entire land surface of the planet. The effects will not be spread evenly, and already high latitudes are warming at twice the rate of the rest of the world while drylands are expanding.
Climate can be seen as the envelope in which all economic activities take place, and these changes could spell significant disruptions for modern society, both in low- and high-income settings. Yet humans have the ability to adapt as well as free will, so one must be careful not to engage in a sort of environmental determinism that draws a direct line from projected climate changes to future migration. Instead, climatic circumstances exist as one of several factors that drive the decision by an individual or community to migrate, and may compound those other drivers or be mitigated through policy action or individual circumstance.
This article draws on a growing evidence base for contemporary environmentally induced migration that includes individual case studies, sophisticated statistical analysis, modeling work, and the UK government’s benchmark 2011 Foresight report, Migration and Global Environmental Change. It summarizes key lessons from the evidence and assesses the implications for future migration under climate change.
The question of how climate factors influence migration is fraught with so many contextual specificities that it is helpful to begin with a clarification of terms as well as some general observations. On the migration side, researchers for more than a century have uncovered a range of so-called “stylized facts” governing how, why, and under what circumstances people move. Push and pull factors in origin and destination areas produce migration streams and counter-streams; at the same time, intervening obstacles such as costs of travel and border controls inhibit migration. Human movement tends to increase over time, and migrants are more likely to move to places where relatives or friends have preceded them. Migration is selective, meaning that, depending on the context, some people—such as those who are younger or males—are more likely to move than others. Finally, economic motives tend to dominate. Environmental factors can influence all these elements.
The three dimensions of migration are space, in terms of the distance migrants travel or the borders they cross; the duration of their stay; and volition of their movement, on a spectrum from fully voluntary to forced. In terms of volition, environmental migration is generally understood to fall on the forced end of the spectrum (see Figure 1). For this reason, terms often associated with climate migration include climate displacement, mass migration, distress migration, and climate refugees—a popular but problematic term, since “refugee” is a legal category limited to people fleeing persecution owing to factors such as their race, ethnicity, creed, or political beliefs.
Figure 1. The Mobility Continuum, from Forced to Non-Forced Movement
Source: Susana Adamo and Alex de Sherbinin, “Climate Change and Migration” (Presentation at CARE climate briefing, February 11, 2009, Washington, DC).
Even at the forced end of the spectrum, a migrant’s volition is worth considering. The aspirations and capabilities theory of migration recognizes that people move because they aspire to a better life than the one they have in their place of origin, and that their ability to act on that aspiration is highly contingent on their various individual and household capitals, including social, human, physical, natural, and financial ones, as well as legal and other barriers. Thus, a number of circumstances are possible: one may desire to migrate but lack the ability; one may desire to stay in place but nevertheless be forced to leave due to a natural disaster, conflict, or government intervention; or one may either desire to migrate or not, and be assisted or required to do so by a government in a process called planned relocation or resettlement.
Beyond voluntary and forced, there are a range of other mobility types, making it challenging to speak of migration in general terms. Domestic or internal migration requires different resources—including financial resources as well as human and social capital—and often has lower barriers than international migration. For this reason, the volume of regular internal migration is estimated to be at least three times larger than international migration. The rate of internal migration is probably even higher, yet data deficiencies make it difficult to know for sure. Even within domestic migration, there are differences in characteristics for people going from rural to urban areas, from one rural area to another, from an urban center to rural periphery, and between urban centers.
Permanent or long-term migration may be the most popular image of migration, but in many regions short-term, circular, or seasonal patterns of mobility predominate, especially for people such as migrant laborers and nomadic pastoralists. Finally, there are often distinct differences in migratory patterns between developed and developing countries.
In terms of the way that climate factors influence migration, risk frameworks such as the one introduced by the IPCC are helpful for understanding how climate hazards intersect with social vulnerability. Climate hazards can be classified by the location, timing, duration, and intensity of events. Social vulnerability is a function of the population’s sociodemographic characteristics such as age, sex, ethnicity, race, education, and major livelihoods, as well as its access to financial and other capitals and adaptive capacity. In general, greater frequency and intensity of climate hazards are more likely to prompt people to migrate when the population is more vulnerable and has a lower capacity to adapt.
Climate events can be divided into fast- and slow-onset events. Fast-onset events include climate extremes such as floods, storms, heat waves, and drought. Slow-onset events are gradual changes to climate regimes—such as increased temperatures or longer-term rainfall variation. Other slow-onset events include sea-level rise, ocean acidification, glacial retreat and related impacts, soil salinization, land and forest degradation, loss of biodiversity, and desertification. Each has a more or less direct relationship to climate change. Evidence suggests that fast-onset events are more likely to result in short-term displacement followed by a return to source areas, whereas slow-onset events are more likely to drive permanent migration. However, successive fast-onset events can reduce household assets in ways that may encourage long-term migration.
The Foresight report suggests that environmental (including climatic) factors may occasionally have direct impacts on population movements, but are more likely to operate through intermediate drivers, namely economic, social, demographic, and political ones (see Figure 2). Household characteristics and intervening obstacles can also influence the decision to migrate. A major contribution of the Foresight project was to emphasize that environmental factors rarely act in isolation, but rather exist as part of a broader constellation of macro-, meso-, and micro-level drivers.
Figure 2. Environmental Drivers of Migration Decision-Making
Source: Foresight, Migration and Global Environmental Change (London: The Government Office for Science, 2011), available online.
Another concept put forward by the Foresight project was that of “trapped populations,” which stay in place despite worsening conditions and a desire to move. At the low end of the income spectrum, people may lack the resources to move despite wanting to do so. Even the cost of bus fare or the prospect of leaving one’s local support network may represent a barrier to moving. Research from Manuela Angelucci, for instance, shows that offering poor families small cash payments can increase a family’s propensity to send at least one member away in order to diversify household income. Meanwhile, at the high end of the income spectrum, people have fewer incentives to migrate and are more likely to be “moored” in place. Migration is more likely for those in the middle of the wealth and capabilities spectrum, as confirmed by the United Nations Development Program’s Scaling Fences report.
Climate Migration in Developing Countries
Researchers investigating the linkages between climate change and migration in developing countries often speak of migration—whether international or domestic, or responding to slow-onset or multiple rapid-onset events—as one of a number of forms of adaptation. In this context, migration serves several purposes.
First, migration to less risky or more stable environments can reduce individual or household exposure to climate hazards such as recurrent droughts hampering agricultural yields or floods inundating coastal areas. Second, at a household level, migration of one or more individuals can be part of a livelihood-diversification and risk-reduction strategy, whereby remittances from household members in destination areas offer the household some financial stability when hazards occur and choke off other income streams. Third, migration can increase household assets and thereby resilience to climate change. Fourth, migration can reduce the number of mouths to feed in a household, especially during the dry or lean season in agricultural regions, thereby increasing food security for those who remain behind. Lastly, returning migrants can bring new skills and technologies back to the communities they left, increasing their wealth and resilience to climate hazards.
A large and growing body of research has investigated the relative influence of climate factors in instigating migration. These range from anthropological case studies and survey research on climate perceptions and migration to statistical analyses based on census or survey data that control for factors known to influence migration and then introduce climate factors to determine their relative effect. Much of the large and growing body of research in this field focuses on the so-called “agricultural pathway,” by which the impact of climate on migration is moderated by changes in agricultural productivity. Analyses of this type in parts of Africa, South Asia, and Latin America show that increased or extreme temperatures and rainfall variability and extremes can negatively affect crop yields in ways that may induce migration. Yet the emigration may be limited in terms of who migrates, from which areas, and to which destinations. In one statistical study in Mexico, Raphael J. Nawrotzki et al. found that longer droughts and periods of anomalously high temperatures increased the propensity for rural-urban migration yet had relatively little impact on other forms of migration.
While these statistical analyses have sometimes found strong correlations between climate anomalies and migration, their effects vary from place to place, suggesting that economic drivers, policy, and cultural factors, such as migration as a rite of passage to manhood, still dominate migration decision-making. Some have argued that climate factors—much like the COVID-19 pandemic—merely expose underlying vulnerabilities and fractures in society. In other words, absent fundamental inequities and systems biased against smallholder agriculturalists or other vulnerable populations, they would not have an impetus to move. Climate factors simply tip the scales slightly in favor of migration. Jesse Ribot et al. found that narratives of impending climatic changes themselves, quite apart from actual experienced impacts, have engendered anxiety regarding the future potential for productive livelihoods in rural Senegal.
The agricultural pathway remains the most studied in terms of climate-migration linkages, but research has also been conducted on pastoralist livelihood systems, in coastal areas in relation to sea-level rise, in forest regions, and in urban areas. Findings similarly vary and are sometimes counterintuitive. For example, in pastoralist systems, climate “displacement” may in fact mean that nomadic herders become less mobile owing to the loss of livestock. Furthermore, many low-lying delta regions remain popular migration destination areas despite their exposure to storms and sea-level rise.
Climate Migration in Developed Countries
Most of the research focus on climate migration has occurred in developing countries, but there is emerging work in developed countries. One reason there has been less focus in developed countries is that their populations are highly urbanized and people’s incomes and livelihoods are less dependent on local environmental conditions owing to extended and international supply chains. In fact, modeling conducted by the author’s center for the Foresight project found high levels of net migration into climate-sensitive dryland and mountain ecosystems in North America over the past four decades, whereas similar ecosystems in all regions of the developing world experienced massive outmigration during the same period.
That said, there is growing interest among scholars in managed retreat from climate-exposed areas, especially coastal zones, as well as fire-prone forest areas and flood-prone riparian environments in developed countries. Research in this area is focused on the mechanics of home buyouts, engineering solutions, environmental justice, understanding local planning needs, and how best to modify existing government policies such as the U.S. National Flood Insurance Program. Thus far, evidence does not suggest that there is wholesale migration out of the near-coastal or fire-risk zones. On the contrary, at least in the United States, these areas generally have seen above-average population growth as so-called “amenity migrants” move to the coast and into the urban-wildland interface. This has occurred despite high-profile natural disasters such as Hurricane Sandy in 2012 and major wildfires in California in 2019 and 2020, among many others. But that may begin to change.
Future Projections and Prospects
This article has provided an overview of the major strands of research on climate change-induced migration. Returning to the question posed at the beginning, how likely is it that we will witness mass migration as result of climate change in coming decades?
Researchers have used a variety of techniques to try and predict numbers of future migrants and, to some degree, source and destination areas. At the simplest level, exposure models identify the number of people who will likely be exposed to a given hazard—most often sea-level rise, but also recurrent flooding or drought—and estimate the proportion of people likely to move. For example, researchers Scott Kulp and Benjamin Strauss estimate that 1 billion people now occupy land less than 10 meters above current high-tide lines, including 230 million below one meter who will presumably need to relocate as sea levels rise. At a more sophisticated level, statistical models of populations’ past tendencies to migrate in response to climate anomalies project possible numbers of migrants under various future scenarios. Newer modeling approaches include agent-based models (ABM), system dynamics, radiation models, and population gravity models. A gravity model developed by the author and colleagues for the World Bank’s Groundswell report finds that by 2050 the world could see 143 million internal climate migrants in Latin America, East Africa, and South Asia alone. Each modeling method has its strengths and weaknesses, and not all can be easily applied in all contexts.
A critical assumption of many models is that the impacts of climate change will progress along a more or less smooth arc, with no major tipping points or thresholds that flip the climate system—and in turn the agricultural, economic, and other systems—onto new and less predictable pathways. This could be a limitation if the future does not play out as predicted. At broader spatial scales, these models also do not adequately account for various possible political upheavals, conflicts, or pandemics which could intersect with climate impacts to add unexpected shocks to existing systems. This is one reason why “resilience” has become a term of art among scholars; resilience-building will be critical for societies to navigate the various emerging challenges in this century.
To conclude, climate change and variability are already affecting mobility of all types, including longer-term migration. As with all migration, most climate migration and displacement will be internal, though even the relatively small international fraction could be sizeable given growing populations and the potential scale of climate impacts. People migrate for a variety of reasons, with economic factors predominating. But in cases of large-scale migration, concerns about local safety and security along with a prevailing hopelessness seem to be driving increasingly perilous journeys.
In this sense, it is a misnomer to speak of “climate migrants” as a distinct group of people forced to move solely because of climate factors. Perhaps this label is appropriate for a few migrants from low-lying areas and small island states, but in many areas climate factors could be better characterized as an additional nudge out of marginal environments for people who also have other reasons to leave. In some regions, people are fleeing increasingly desperate situations and mobility is one of the few forms of personal agency available to them. In others, a perception that climate change will make life more difficult in the future, combined with a desire for a better life, prompts them to move for their own sake or that of their children.
In all of this, it is important to note that society and institutions are ill-prepared to meet the coming challenge. Legal instruments such as the 1951 Refugee Convention set narrow criteria for protection, for example requiring that an individual fear persecution, effectively excluding those who would nevertheless truly suffer if forced to return to their home countries. Donors are increasingly devising programs to promote in situ adaptation strategies, some of which are designed to reduce outmigration from rural areas, as well as migrant-centered interventions. Some are designed to facilitate the integration of migrants into urban environments or host countries, while others seek to repatriate migrants from developed countries, in some cases providing them with job training and financial assistance. Programs to limit migration generally have had a spotty record of success in achieving their desired aims. In fact, one of the byproducts of development is greater freedom of choice; where people are able to fully realize their aspirations and exercise their capabilities they often choose to move. This will be no less the case in a world transformed by increasingly severe climate impacts.
Adamo, Susana and Alex de Sherbinin. 2009. Climate Change and Migration. Presentation at CARE climate briefing, February 11, 2009, Washington, DC.
Angelucci, Manuela. 2015. Migration and Financial Constraints: Evidence from Mexico. Review of Economics and Statistics 97 (1): 224-28.
Ayamga, Jennifer, Shouvik Das, Sumana Banerjee, Robert Nicholls, Craig Hutton et al. 2019. Climate Change, Migration, and Adaptation in Deltas: Key Findings from the DECCMA Project. N.p.: Deltas, Vulnerability, and Climate Change: Migration and Adaptation (DECCMA). Available online.
Barnett, Jon and W. Neil Adger. 2018. Mobile Worlds: Choice at the Intersection of Demographic and Environmental Change. Annual Review of Environment and Resources 43 (1): 245-65.
Black, Richard, Stephen R.G. Bennett, Sandy M. Thomas, and John R. Beddington. 2011. Migration as Adaptation. Nature 478 (7370): 447-49. Available online.
Black, Richard, W. Neil Adger, Nigel W. Arnell, Stefan Dercon, Andrew Geddes et al. 2011. The Effect of Environmental Change on Human Migration. Global Environmental Change 21: S3-11. Available online.
Borderon, Marion, Patrick Sakdapolrak, Raya Muttarak, Endale Kebede, Raffaella Pagogna et al. 2019. Migration Influenced by Environmental Change in Africa: A Systematic Review of Empirical Evidence. Demographic Research 41: 491-544. Available online.
Büntgen, Ulf and Nicola Di Cosmo. 2016. Climatic and Environmental Aspects of the Mongol Withdrawal from Hungary in 1242 CE. Scientific Reports 6: 25606. Available online.
Center for International Earth Science Information Network (CIESIN). 2011. Estimating Net Migration by Ecosystem and by Decade: 1970–2010. London: Foresight Environmental Migration Project. Available online.
Chan, Sewell. 2015. Paris Accord Considers Climate Change as a Factor in Mass Migration. The New York Times, December 13, 2015. Available online.
Collyer, Michael. 2010. Stranded Migrants and the Fragmented Journey. Journal of Refugee Studies 23 (3): 273-93. Available online.
De Haas, Hein. 2011. The Determinants of International Migration: Conceptualizing Policy, Origin and Destination Effects. Working Paper 32, International Migration Institute, Oxford, United Kingdom, April 2011. Available online.
De Sherbinin, Alex and Ling Bai. 2018. Geospatial Modeling and Mapping. In Routledge Handbook of Environmental Displacement and Migration, eds. Robert McLeman and François Gemenne. New York: Routledge.
De Sherbinin, Alex, Marc Levy, Susana Adamo, Kytt MacManus, Greg Yetman et al. 2012. Migration and Risk: Net Migration in Marginal Ecosystems and Hazardous Areas. Environmental Research Letters 7 (4): 045602. Available online.
De Sherbinin, Alex, Marcia Castro, François Gemenne, M. M. Cernea, Susana Adamo et al. 2011. Preparing for Resettlement Associated with Climate Change. Science 334 (6055): 456-57. Available online.
Drake, B. Lee. 2017. Changes in North Atlantic Oscillation Drove Population Migrations and the Collapse of the Western Roman Empire. Scientific Reports 7 (1): 1227. Available online.
Economist, The. 2020. The Idea that Aid and Development Slow Migration Is Wrong. The Economist, August 27, 2020. Available online.
Field, Christopher B., Vicente Barros, Thomas F. Stocker, Qin Dahe, David Jon Dokken et al., eds. 2012. Managing the Risks of Extreme Events and Disasters to Advance Climate Change Adaptation: Special Report of the Intergovernmental Panel on Climate Change. Cambridge, UK: Cambridge University Press.
Foresight. 2011. Migration and Global Environmental Change. London: The Government Office for Science. Available online.
Friedman, Thomas L. 2016. Out of Africa. The New York Times, April 13, 2016. Available online.
Fussell, Elisabeth. 2012. Space, Time, and Volition: Dimensions of Migration Theory. In Oxford Handbook of the Politics of International Migration, eds. Marc R. Rosenblum and Daniel J. Tichenor. Oxford, UK: Oxford University Press.
Hauer, Mathew E. 2017. Migration Induced by Sea-Level Rise Could Reshape the US Population Landscape. Nature Climate Change 7 (5): 321-25.
Hugo, Graeme. 1996. Environmental Concerns and International Migration. International Migration Review 30 (1): 105-31.
Kennett, Douglas J., Sebastian F.M. Breitenbach, Valorie V. Aquino, Yemane Asmerom, Jaime Awe et al. 2012. Development and Disintegration of Maya Political Systems in Response to Climate Change. Science 338 (6108): 788-91.
Kulp, Scott A. and Benjamin H. Strauss. 2019. New Elevation Data Triple Estimates of Global Vulnerability to Sea-Level Rise and Coastal Flooding. Nature Communications 10 (1): 1-12. Available online.
Lustgarten, Abrahm. 2020. The Great Climate Migration Has Begun. The New York Times Magazine, July 23, 2020. Available online.
McLeman, Robert and Barry Smit. 2006. Migration as an Adaptation to Climate Change. Climatic Change 76 (1-2): 31-53.
McLeman, Robert and François Gemenne. 2018. Environmental Migration Research. In Routledge Handbook of Environmental Displacement and Migration, eds. Robert McLeman and François Gemenne. New York: Routledge.
Milman, Oliver, Emily Holden, and David Agren. 2018. The Unseen Driver Behind the Migrant Caravan: Climate Change. The Guardian, October 30, 2018. Available online.
Nawrotzki, Raphael J., Jack DeWaard, Maryia Bakhtsiyarava, and Jasmine Trang Ha. 2017. Climate Shocks and Rural-Urban Migration in Mexico: Exploring Nonlinearities and Thresholds. Climatic Change 140 (2): 243-58. Available online.
Nawrotzki, Raphael J. and Maryia Bakhtsiyarava. 2016. International Climate Migration: Evidence for the Climate Inhibitor Mechanism and the Agricultural Pathway. Population, Space and Place 23 (4): e2033. Available online.
Ravenstein, Ernest George. 1889. The Laws of Migration. Journal of the Royal Statistical Society 52 (2): 241-305. Available online.
Ribot, Jesse, Papa Faye, and Matthew D. Turner. 2020. Climate of Anxiety in the Sahel: Emigration in Xenophobic Times. Public Culture 32 (1): 45-75.
Rigaud, Kanta Kumari, Alex de Sherbinin, Bryan Jones, Jonas Bergmann, Viviane Clement et al. 2018. Groundswell: Preparing for Internal Climate Migration. Washington, DC: World Bank Group. Available online.
Shah, Sonia. 2020. Climate Change Will Drive People Across Borders: But the Real Crisis Would Be Making Them Stay. Foreign Affairs, July 29, 2020. Available online.
United Nations Development Program (UNDP). 2019. Scaling Fences: Voices of Irregular African Migrants to Europe. New York: UNDP. Available online.
Warner, Koko, Charles Ehrhart, Alex de Sherbinin, Susana Adamo, and Tricia Chai-Onn. 2009. In Search of Shelter: Mapping the Effects of Climate Change on Human Migration and Displacement. Bonn, Germany: United Nations University, CARE, and CIESIN-Columbia University. Available online.