VoxEU Column Environment Migration

Migration’s response to increasing temperatures

Climate change can affect agricultural productivity and the incentives of people to remain in rural areas. This column looks at the effects of warming trends on rural-urban and international migration. In middle-income economies, higher temperatures increased emigration rates to urban areas and to other countries. In very poor countries, however, higher temperatures reduced the probability of emigration to cities or to other countries, consistent with the presence of liquidity constraints.

Climate change and related migration

The Intergovernmental Panel on Climate Change’s Fifth Assessment Report (AR5), which is the most comprehensive and relevant analysis of climate change, concludes that hundreds of millions of people will be affected by climate change. Its consequences will be felt directly and indirectly via resource availability and population movements, spreading consequences across the globe.

For this reason, the EU’s foreign and security policies, as well as official publications and strategies, have devoted increasing attention to climate-related factors. For instance, the joint report by Javier Solana and the European Commission defines climate change as a ‘threat’ multiplier, as it could be responsible for political and security risks affecting European interests (European Council 2008). Environmentally induced migration is quoted among the various threats identified in the report. According to the Council Conclusions on EU Climate Diplomacy, adopted in June 2011, climate change is a global environmental and development challenge with significant implications related to security and migratory pressures (European Council 2011).

The idea that climate-related migration could generate repercussions for European security is related to the possibility of large inflows of people from the areas adversely affected by climate change. Predictions of these flows, however, are extremely imprecise and based on a very wide range of hypotheses. The number of predicted migrants range wildly from 25 million to one billion over the next 40 years (IOM 2009). Vulnerability to climate change in poor countries, while certainly increasing the incentive to migrate, does not necessarily imply that migration will occur. Climate change, by decreasing the available resources, may constrain the ability to emigrate, and some vulnerable individuals may find themselves less mobile and less likely to migrate (Barrett 2008, Cattaneo and Massetti 2015, Gray and Mueller 2012, Foresight 2011).

New research

In a recent paper (Cattaneo and Peri 2015), we tackle the connection between increasing temperatures and migration by analysing the effect of differential warming trends across countries on the probability of migrating out of the country or migrating from rural to urban areas. A crucial insight is that by impoverishing rural populations and worsening their income perspectives, long-term warming affects migration in different ways, depending on the initial income of those rural populations. A decline in agricultural productivity, causing a decline in rural income, seems to have a depressing effect on the possibility of emigrating in extremely poor countries where individuals live on subsistence income. Lower income worsens their liquidity constraint, implying that potential migrants have a reduced ability to pay for migration costs and to afford travel and relocation costs. In this case, global warming may trap rural populations in local poverty. In contrast, in countries where individuals are not extremely poor, a decline in agricultural income strengthens the incentives to migrate to cities or abroad. Decreasing agricultural productivity may encourage a mechanism that ultimately leads to economic success of migrants, benefitting their country of origin and shifting people out of agriculture into urban environments.

Figure 1 provides correlations that corroborate this insight. The figure plots long-term changes (between 1960 and 2000) in temperature (horizontal axis) against long-term changes in emigration rates for poor countries (Panel 2) and for middle-income countries (Panel 1). The difference in the relationship between the two groups of countries is clear. Middle-income countries show a (small) positive correlation while poor countries show a negative correlation between temperature and emigration rate changes.

Figure 1. Change in emigration rates and in average temperature

Note: The graphs plot on the horizontal axis the natural logarithm of the average temperatures between 2000  and 1981 minus the natural logarithm of the average temperatures between 1960 and 1980. On the vertical axis we represent the natural logarithm of the average emigration rates between 1990 and 2000 minus the average emigration rates between 1970 and 1980.

The difference between middle-income and poor countries

In the empirical analysis we pursue more systematically the two effects presented in Figure 1. Using decade changes between 1960 and 2000 for 116 countries, ranging from very poor to middle income, we perform a regression analysis that controls for country effects, decade effects, and several other geographic variables and allows for a different impact of temperature on emigration and urbanisation rates in poor and middle-income countries.

  • We find that increasing temperatures are associated with lower emigration and urbanisation rates in very poor countries.
  • In contrast, in middle-income countries they are associated with positive changes in emigration and urbanisation rates.

The incentive effect driven by lower agricultural productivity prevails in middle-income countries, and rural population is driven to cities, speeding the country's structural transformation and ultimately increasing income per person. In poor countries, the worsening of the liquidity constraint due to lower agricultural productivity prevails, and urbanisation and emigration are slowed.

  • We find consistently that emigration in middle-income countries induced by higher temperatures is associated with growth in GDP per person.
  • The slowing of emigration and urbanisation associated with climate warming in poor countries is associated with lower average GDP per person.

This connection between temperatures and GDP growth was first pointed out by Dell and Olkien (2012). Our study provides an important channel to explain it.

Urbanisation and industrialisation are crucial mechanisms for GDP growth. For countries with intermediate levels of income per person, warming can push towards these gains. However, for countries where agricultural productivity is so low as to trap rural populations at subsistence levels, warming may instead slow economic transformation. These effects could contribute to divergence of income between poor and middle-income countries.

Where do people migrate to in response to warming?

Does warming produce large scale movements of individuals from middle-income countries in Africa, Asia, and Latin America to rich countries in Europe and North America? Or does it produce more local migrations in the regions?

  • We find that growing temperatures are mainly associated with emigration to non-OECD destinations that are close to the countries of origin (especially those within a 1,000km radius).  
  • Emigration to OECD (i.e. rich) countries does not seem affected.

This result is consistent with the idea that climate-driven emigration is associated with a worsening of local opportunities and migrants move where they have better chances of finding a job given their current constraints. This ‘push’ factor (decreased rural income) increases migration to similar economies rather than to OECD economies. On the other hand, the migration-reducing effect for poor countries (due to worsening opportunities) affects both types of destination, as potential emigrants become less likely to leave the country altogether. Combining the effect on poor and middle-income countries, it appears that increases in average temperatures may actually decrease overall emigration to OECD countries. Middle-income countries are not more likely to experience emigration towards those destinations, while poor countries experience a reduction in emigration rates altogether. These findings suggest that climate change is unlikely to be the driver of large migrations to Europe as the impact on poor countries seems negative and climate-related migrations seem more local.  

Migration and natural disasters

Climate change is also expected to bring an intensification of extreme weather events. For this reason, we tested whether temperature anomalies and natural disasters such as droughts, floods, and storms influence emigration rates in middle-income and poor countries. We find that long-run emigration rates in poor or middle income countries are not significantly affected by the occurrence of these events. It is likely that natural disasters drive different types of migration, more akin to local mobility and temporary. Given their relatively rare occurrence and temporary nature in the considered period, extreme weather episodes did not affect significantly long-run rural-urban and international migration.


In this column we have focused on the potential impact of growing average temperatures on rural-urban and international migration. We found that in very poor countries, warming implies less emigration. Rural populations may be stuck in deeper poverty with fewer resources to migrate. In contrast, in countries where income is not as low, lower agricultural productivity increases the incentives to migrate, producing higher emigration rates. Through these different responses temperature changes may contribute to a divergence of income and opportunities between very poor and middle-income countries. Finally, a future of increased migrations to Europe or to the US driven by global warming is not a scenario supported by our analysis.


Barrett C (2008), "Poverty traps and resource dynamics in smallholder agrarian systems", in R Dellink and A Ruijs (Eds), Economics of poverty, environment and natural-resource use, Dordrecht, The Netherlands: Springer; pp. 17-40.

Cattaneo C and E Massetti (2015) “Migration and Climate Change in Rural Africa", FEEM Working Papers N. 2015.29

Cattaneo C, and G Peri (2015), “The Migration Response to Increasing Temperatures” NBER Working Paper 21622.

Dell M, B Jones and B Olkien (2012), "Temperature Shocks and Economic Growth: evidence from the last half Century," American Economic Journal: Macroeconomics, 4(3), pp. 66-95

European Council (2008), “Climate change and international security”, Joint Paper from the High Representative and the European Commission to the European Council, S113/08.

European Council (2011), “Towards a renewed and strengthened European Union Climate Diplomacy”, Council Conclusions on EU Climate Diplomacy, 3106th Foreign Affairs Council Meeting, Brussels, 18 July 2011, Joint Reflection Paper by the High Representative and the Commission.

IOM (2009), “Migration, Environment and Climate Change: assessing the evidence”, International Organization for Migration, Geneva

Foresight (2011) Migration and Global Environmental Change: Future Challenges and Opportunities. Final Project report, UK Government Office for Science, London

Gray, C L and V Mueller (2012), “Natural disasters and population mobility in Bangladesh”, Proceeding of the National Academy of Science, 109(16), pp. 6000-6005.

1,995 Reads