Weather risks and international migration: Panel-data evidence from Tajikistan

Takeshima, Hiroyuki; Lambrecht, Isabel Brigitte; Akramov, Kamiljon; and Ergasheva, Tanzila. 2025. Weather risks and international migration: Panel-data evidence from Tajikistan. IFPRI Discussion Paper 2340. Washington, DC: International Food Policy Research Institute. https://hdl.handle.net/10568/175059

Climate change and increased frequency of abnormal weather are becoming growing threats to people’s livelihood, including in Central Asia. These threats are particularly challenging in Tajikistan, the poorest country in the Central Asia region. Despite the fact that migration is prevalent and remittances account for a significant share of GDP, evidence is scarce as to whether the decision to migrate is driven by weather shocks, whether migration is used as mitigating tool against adverse weather shocks, and how much of the loss in welfare is actually mitigated by such migration. This study aims to narrow this knowledge gap by providing evidence based on a unique panel dataset from one of the poorest and agriculturally dependent regions in Tajikistan (Khatlon province), combined with a detailed set of various climate data. In doing so, we apply a novel approach through the least absolute shrinkage and selection operator (LASSO) to identify key weather shock variables among a vast set of potential variables associated with outmigration decisions in different districts. Our results show that different types of weather shocks are associated with outmigration decisions in different districts within the province, ranging from rainfall, temperatures, drought, and windspeed in different subperiods throughout the year. Regardless, more abnormal weather is almost universally associated with more outmigration, and outmigration significantly mitigates the potentially adverse effects on household consumption and food/nutrition security in the origin households. However, more abnormal weather in the origin location is also associated with reduced remittances per month per migrant sent to the origin location. Thus, the capacity of migration to mitigate against weather shocks is still limited. Combined with migration policies that increase net earnings during migration, supplementary support to enhance climate resilience in home locations, such as climate-smart agriculture and development of the non-farm sector, remains critical.