Climate Change, Inequality, and Human Migration

Michał Burzyński, Christoph Deuster, Frédéric Docquier, and Jaime de Melo

Journal of the European Economic Association, Volume 20, Issue 3 (2022), Pages 1145–1197 


This paper examines the long-term implications of climate change for migration and inequality. The authors investigate: (i) the scale of climate migration; (ii) the characteristics of climate migrants including their age and educational attainment; (iii) their origins and destinations, including local displacements, migration within their country of origin, or international migration; and (iv) the socio-economic implications of climate migration.  

The empirical model accounts for three types of climate change: (1) slow-onset changes in temperatures, which affect productivity in both agricultural and non-agricultural sectors; (2) rising sea levels, which cause flooding in low-elevation coastal areas; and (3) frequency and intensity of fast-onset natural disasters (heatwaves, droughts, floods, and severe storms) that result in utility and productivity losses. The model incorporates a spatial representation of the world’s land surface divided into millions of pixels measuring about 5 square kilometers at the equator.  

Main results: 

  • Increasing temperatures will increase migration from lower-productivity rural to higher-productivity urban localities in less-developed countries. Median projections show that agricultural productivity will decrease more than 50 percent in regions close to the equator, while agricultural productivity will decline to a smaller degree (or even increase) at high latitudes. Non-agricultural productivity is projected to decrease by 30–40 percent in low-latitude areas and increase slightly at high latitudes. These slow onset changes, which are easier to anticipate, are projected to induce migration from lower- to higher-latitude locations. While climate change is likely to increase the distance of migratory movements (since temperature changes are spatially correlated), it is not projected to induce a massive relocation across borders. 
  • Rising sea levels are projected to generate local and regional forced displacement. Median projections suggest that sea level will rise by approximately 0.7 meters by 2100, forcibly displacing an estimated 47 million people, destroying capital infrastructure, and reducing productivity in low lying areas. Highly populated coastal areas in the Mexican Gulf, eastern United States, and northern Europe are likely to be affected but are also expected to benefit from mitigation measures that will limit forced displacement. However, populations of many developing countries in Africa, Asia, and South America lack mitigation capacity and will be vulnerable to forced displacement. Around 25–30 percent of forced migrants are expected to migrate internationally, while 40–50 percent are expected to move inter-regionally within their home country. Rising sea levels also reduce the attractiveness of coastal areas, affecting the allocation of global migrants across destinations. 
  • Fast-onset climate shocks will be concentrated in developing regions and are expected to lead to increased international migration. Fast onset climate shocks will be concentrated in developing regions, inducing adverse effects on global GDP (a decline of 12.7 percent in 2100), inequality and poverty. The economic impact of these shocks will increase incentives to migrate, but internal migration responses will be limited due to the spatial correlation of fast-onset shocks. Consequently, the number of international migrants is projected to increase significantly, by a factor of 2–3. 
  • Climate change will deepen the gap between developing and developed countries and between rural and urban areas, increasing extreme poverty in many developing regions. Climate change is projected to intensify global inequality, with losses in gross domestic product (GDP) concentrated in Africa, Asia, and South America. Climate change is also projected to accelerate urbanization, especially in developing countries. Median projections indicate that 9.5 percent of the world’s population will be living in extreme poverty, compared to 4 percent without climate change. 
  • Overall, median projections indicate that climate change will induce voluntary and forced migration of 62 million working-age individuals over the course of the 21st century. In this mid-range scenario, climate change is estimated to induce 62 million working-age migrants including 57 million international migrants (22 million people from Africa, 27 million from Asia, and 6 million from South America). It is projected that 24 million climate migrants will go to Europe, 17 million to North America, and 5 million to Oceania. 
  • Climate change raises the world’s stock of human capital, as people tend to move from poorer regions to richer regions, where there is almost universal access to education. Compared to those staying in home countries, climate migrants are more educated (especially migrants leaving Africa), and tend to move longer distances, substituting within-country for cross-border movements. 

The authors argue that aggregate numbers of international migrants are relatively small from the perspective of sending countries, indicating that international climate migration will be a costly and unlikely adaptation strategy for most people affected by climate change. The authors conclude that it is unlikely that climate shocks will induce massive international flows of migrants, except under combined extremely pessimistic climate scenarios and highly permissive migration policies. Rather, moderate migration responses to climate change imply that many will be trapped in impoverished and troubled regions, inducing significant increases in extreme poverty.