Перейти к основному содержанию

Climate Change and International Migration: Evidence from Tajikistan

Дата публикации
Просмотреть оригинал


This paper investigates the impact of environmental factors that indicate climate change on household decisions to migrate abroad in the case of Tajikistan, an environmentally vulnerable and a labor-migrant source country in Central Asia. Both long-term climate variation (measured by weather anomalies) and short-term weather shocks (proxied by floods) are considered as environmental factors that could induce migration from Tajikistan. Using two waves of a nationally representative household survey and employing an empirical method within the New Economics of Labor Migration theory framework, the results highlight the differing effects of environmental factors (depending on their type and intensity) on the probability to migrate abroad. The findings show that a rise in air temperature from its longterm average reduces emigration, while changes in precipitation have a non-linear impact on emigration. There are substantial differences in seasonal weather anomalies, of which winter temperature and precipitation have the most significant impact on household decisions to migrate. Sudden onset environmental shocks appear to have a lagged impact on emigration.


Scientific arguments and evidence have drawn increasing attention to changes in the global climate in terms of rising global air temperature and sea level, retreating snow cover and glaciers, altered precipitation patterns, and more frequent extreme weather events (National Centers for Environmental Information 2020). While the Intergovernmental Panel on Climate Change (IPCC 2014) has argued that the greatest impact of climate change could be on human migration and has predicted that 200 million people will have migrated from their place of origin by 2050, the evidence for the effect of environmental factors on human migration, particularly on temporary migration as an adaptive strategy to confront climate stress, remains inconclusive. Because environmental factors affect both the incentive and the ability to migrate, the relationship is not straightforward and involves many complexities.

Environmental factors could exert substantially heterogeneous impacts on migration depending on the initial climatic and socio-economic conditions of the countries in question (Obokata et al. 2014; Berlemann and Steinhardt 2017). Many developing countries are predicted to be disproportionally affected by climate change due to their geography, agriculture-based economies, and lack of adaptive mechanisms (Beine |and Parsons 2017). Particularly, climate change poses tremendous challenges to livelihoods based on agriculture around the world (Cattaneo and Peri 2015), and people seek out informal ways of coping including migration when formal mechanisms such as insurance and credit markets are ill-functioning or inaccessible (Lewin et al. 2012).

Some studies view migration as a common mechanism, historically, for adapting and coping with climate and weather-related shocks, because migration helps mitigate the adverse effects of such events on livelihoods. Climate change and natural disasters could induce migration if individuals do not have any other mitigation strategies. If migrants send remittances home, migration could even have income diversification and consumption smoothing effects on the left-behind household members when formal safety nets and insurance markets are incomplete or absent (Wouterse and Taylor 2008). For example, Filipino households with overseas migrants can reduce adverse income shocks with receipt of remittances, while households without overseas migrants cannot (Yang and Choi 2007).

Conversely, the impact of environmental factors may be marginal if people cannot afford to migrate or if there are alternative coping strategies available that would allow them to alleviate the adverse effects of environmental stressors. Migration is costly both financially and psychologically. Studies have shown that people in the lowest income quantile do not necessarily migrate in the aftermath of natural disasters, as they lack the means to finance such migration. While environmental stressors may increase incentives to migrate abroad, these may not materialize if individuals are credit constrained. For instance, Bazzi (2017) has found that persistent income shock and liquidity constraints reduce emigration from rural Indonesia, while Bryan et al. (2014) have argued that liquidity constraints are a barrier to seasonal rural-urban migration in Bangladesh. The latter authors randomly assigned monetary incentives to rural farmers in Bangladesh and found that the incentive significantly increased rural-urban seasonal migration by easing such constraints. Similarly, individuals at the top of the income distribution are also less likely to migrate as a response to environmental shocks because they have other mitigating strategies available to them (Drabo and Mbaye 2015).
The intensity, frequency and the types of environmental shock play crucial roles in determining whether households engage in international or domestic migration, or stay in their place of origin. Studies have shown that sudden-onset and high-intensity environmental stressors such as floods, tsunamis, and earthquakes result in an immediate increase in short-distance or domestic migration. On the other hand, mediumand long-term environmental stressors such as long-term variations in temperature and precipitation and the subsequent degradation of natural resources—which cause repeated livelihood failures—often lead to long-distance or international migration (Brember and Hunter 2014). For example, Marchiori et al. (2012) have found that climate change, measured by long-term climatic variations (i.e., rising temperature and altered precipitation patterns from the historical averages) push migrants from developing countries to developed countries. Knowing what type of environmental stressors affect households and cause them to engage in international rather than internal migration would have implications for the cost and affordability of relocation, policy management of migrants, the size of future remittances, and the economic prospects abroad versus at home. These implications are likely to be country specific depending on the initial socio-economic conditions, geography, and capacity for adaptive mechanisms.

Empirical evidence is needed to inform policymakers about the different migration responses to climate change and natural disasters in the countryspecific context.

In this area, Tajikistan is an interesting case for at least two reasons. First, temporary labor migration from Tajikistan plays a vital role in keeping the Tajikistan economy afloat via remittances from Tajik migrants abroad (particularly in the Russian Federation). A recent report on a nationally representative household survey in Tajikistan found that about 40% of households have at least one member who has migrated abroad, while only 3% of households had internal migrants in 2018 (Japan International Cooperation Agency 2020). According to the report, out of all total international migrants, almost 99% chose the Russian Federation as their destination due to the well-established migration corridor and the historical tie through the former Soviet Union. Remittances from migrant workers have been a major contributor to Tajikistan’s economy, constituting 30%–50% of its GDP since the mid-2000s, and making the country one of the top remittancedependent countries in the world (World Bank 2020).

Second, Tajikistan is highly vulnerable to climate change due to its specific orography and climatic conditions, as well as low capacity for adaptation and coping (World Food Programme 2017). About a half of Central Asia’s glaciers are located in Tajikistan, stretching over 8.5 thousand square kilometers (UNDP 2012). These glaciers serve an important climatic role in retaining water, controlling flows, and regulating the climate not only in Tajikistan, but also in Central Asia. According to the World Bank (2020), there has been a significant increase in the annual average air temperature (a rise by 0.2°C–1.2°C since 1940) in Tajikistan, which has caused melting of seasonal snow cover and glaciers. The IPCC (2014) has predicted that the average surface temperature in Tajikistan will rise a further 1°C–3°C by 2050. Tajikistan has also experienced a sharp increase in the intensity and frequency of extreme weather events, particularly floods and mudflows related to the melting of glaciers. These environmental stressors threaten livelihoods in Tajikistan, especially among those dependent on agriculture and natural resources. Environmental mismanagement and poor infrastructure are ill suited to mitigating such environmental shocks. Due to the lack of coping mechanisms, labor migration could serve as a way to diversify household income and reduce economic stress triggered by environmental shocks in Tajikistan.

This paper considers the impact of environmental factors that gauge climate change on household decisions in Tajikistan to send one or more household members abroad temporarily to work. The paper uses data from the Tajikistan Living Standards Survey (TLSS) conducted in 2007 and 2009. The contribution of the paper is twofold. First, it examines the effects of both long-term climatic shocks1 and short-term natural disaster shocks on the migration decision. Most studies analyze either long-term climatic variability or short-term weather shocks for their effects on migration decision. However, long- and short-term environmental factors are perceived differently by households and may have different associations with their migration decisions. It is therefore important to see how the impact of slow-onset or long-term climate variability differs from suddenonset or short-term natural disasters on the decision to migrate under the same household setting. A further contribution is that this paper considers the individual impacts of seasonal climate variations measured by seasonal temperature and precipitation anomalies in addition to annual average weather anomalies. For short-term environmental shocks, data on the incidence of floods at the district level were used.

Second, this is, to the best of the author’s knowledge, the first study to analyze the impact of climate variability and weather shocks on international migration in Tajikistan. Given the importance of Tajikistan’s geography for normalizing weather in Central Asia, climate change could have an extensive impact on the livelihoods not only of the Tajik people, but also of citizens in neighboring Central Asian countries. Understanding whether environmental stressors lead to international migration is crucial to help formulate policy recommendations to manage climate change and consequent natural disasters and to support the livelihoods of those affected.

The findings suggest that environmental factors have some liquidity constraint–related effects on international migration decisions, as the long-term increase in temperature and precipitation from historical averages are found to reduce emigration. The findings also show that seasonal weather changes have differing impacts on migration.

Throughout the course of a year, the most significant changes in weather have been occurring in winter, which coincides with the return of temporary migrants from the Russian Federation and the planting of crops such as winter wheat, which is a staple food in Tajikistan. Consequently, the rise in winter temperatures and precipitation increases emigration from Tajikistan and suggests the lack of other coping mechanisms for migrants.

This paper is organized as follows. Section 2 describes the data sources used and the current state of climate change and migration in Tajikistan. Section 3 explains the theoretical model and empirical strategy applied in this study. Section 4 presents and discusses the results. Section 5 summarizes the main findings and gives perspectives for future research.

Asian Development Bank: © Asian Development Bank