Climate migration in Europe? Evidence from drought exposure and internal migration in Italy

In recent years, studies on climate-induced migration have proliferated, but the focus has been mainly on low- and middle-income countries. This paper fills this gap by examining whether climate-driven migration patterns emerge in high-income settings, specifically Italy, one of the European countries most severely affected by climate change due to its Mediterranean location. Using annual bilateral migration flow data from the Italian National Institute of Statistics (ISTAT) for the period 2002-2023, drawn from the National Register of the Resident Population across 104 provinces (NUTS-3 level), and combining these data with drought severity indices (SPEI and PDSI), we examine how worsening climatic conditions affect internal migration within Italy. We adopt a panel data framework using a Poisson pseudo-maximum likelihood estimator, which allows us to capture both spatial and temporal variation in climate shocks and migration responses across provinces and regions. Our findings yield three main insights. First, we find robust evidence that drought exposure drives internal migration flows, showing that even high-income countries with greater adaptive capacity are not immune to climate-induced mobility. Translating the estimated effects into population counts shows that drought exposure over the past two decades is associated with several hundred thousand additional inter-provincial moves, with implied effects exceeding 560,000 migrants in high-exposure settings. Additional analyses also indicate higher international out-migration from drought-affected provinces, challenging the view that climate migration is primarily a concern of lower-income countries. Second, we reveal significant heterogeneity by citizenship status: foreign citizens display a markedly higher propensity to migrate in response to adverse climatic conditions than Italian citizens. Third, we find that local economic structure plays a crucial role, with drought-induced out-migration being strongest in middle-income regions rather than in the poorest or richest areas. Analyses of agricultural dependence reveal similar non-linear patterns, and further evidence indicates that drought affects migration by depressing local incomes and widening income gaps relative to potential destinations. Together, these results have important implications for climate adaptation policy, highlighting the need for proactive planning and early warning systems to anticipate and manage climate-induced movements.