Modelling the effects of climate change on economic growth: a Bayesian Structural Global Vector Autoregressive approach
Maryam Ahmadi (Fondazione Eni Enrico Mattei); Chiara Casoli (Fondazione Eni Enrico Mattei); Matteo Manera (Fondazione Eni Enrico Mattei and Department of Economics Management and Statistics – DEMS, University of Milano-Bicocca); Daniele Valenti (Fondazione Eni Enrico Mattei and Department of Environmental Science and Policy – DESP, University of Milan)
C11, C32, O44, Q51, Q54
Climate econometrics, Bayesian Structural VARs, Identification theory, Global VARs
The identification of the effects of climate shocks on economic growth is central to design effective policies aiming at managing the future global climate change challenge. In this study, we investigate the effects of temperature and precipitation shocks on economic growth across different countries by means of a new methodology, namely a Bayesian Structural Global VARX model. This setup accommodates economic interpretation of the shocks and accounts for crosssectional spillovers among countries, as well as endogeneity of the climate variables with respect to the economy. Results show a high degree of heterogeneity, with some economies positively and some others negatively affected by climate shocks. In contrast with a consistent strand of the literature, according to which hot and poor countries bear the heaviest burden, we show that climate shocks may have severe effects for the economic growth of rich and cold countries as well. Furthermore, accounting for trade interdependence across countries, we document that, in response to unexpected temperature and precipitation shocks, some economies benefit from interconnections, while some others are damaged, depending on some key structural characteristics as the import-export mix, the relevant trade partners network and the level of economic development.
Suggested citation: M. Ahmadi, C. Casoli, M. Manera, D. Valenti, ‘Modelling the effects of climate change on economic growth: a Bayesian Structural Global Vector Autoregressive approach’, Nota di Lavoro 046.2022, Milano, Italy: Fondazione Eni Enrico Mattei