Safety measures concerning the Coronavirus COVID-19. Read more about the measures.

x

ESM Research seminar: Weather Shocks, Climate Change, and Macro-Economic Projections with Emanuele Massetti (IMF)

-

Agenda

   

 

The ESM Research team is pleased to invite you to attend the hybrid seminar on:

 

Weather Shocks, Climate Change, and Macro-Economic Projections

 

Held by:

Emanuele Massetti

 (IMF)

Jointly with Matthieu Bellon

 

Thursday, 29 February 3 to 4.30pm CET

 

Please let us know by Wednesday, 28 February COB If you will attend virtually or in person.

 

If you attend virtually, please join via Teams link below

 

If you attend in person, please send an email confirmation to ESMResearch@esm.europa.eu   

The event will take place in our Conference area, make sure to come 15 minutes before the event allowing for a timely start!

 

  

  

Abstract: Projecting country economic growth is central to economic and policy planning and has been studied extensively. For a long time and in most countries, the literature has not paid attention to the role that climate conditions play in determining future growth.

We are now at a turning point when climate change is not only a sound scientific hypothesis but a well-known fact that can already be felt globally.  While many signals of climate change are still uncertain due to large natural weather variability, some trends are on an unambiguous trajectory globally. This new evidence must be reflected in both short- and long-term economic projections and scenarios.

In the short term, country growth projections are anchored by recent growth dynamics, economic policies, and external shocks, but are also affected by weather shocks, whose likelihood and intensity can be changing as a result of global warming. Over longer time horizons, the importance of recent dynamics and all the mean-reverting shocks, including weather shocks, fades away. Instead, long-term economic growth depends on fundamental drivers like demographic development, structural policies, investment in both technology and capital, long-run average weather – namely climate – as well as the distribution of shocks including the distribution of extreme weather.

The literature has made great strides to understand how trends in a common set of climate variables (typically temperature and precipitation averages) affect the economic growth. These studies typically focus on few climate variables and assume a globally uniform response. However, climate science suggests that climate variables are changing differently across countries, that the level of uncertainty surrounding these changes also differs by country, and that non-linearities for different which are ill-reflected in averages matter. Climate economics show that different countries respond differently to similar changes in climate variables and that climate economic impacts might be best captured by different climate variables depending on the country being studied. In short, the economic effects of climate change are widely heterogeneous across countries and more work is needed to reflect it when projecting economic growth. 

We take this heterogeneity seriously and contribute to the literature by developing and implementing a method to estimate the economic growth impact of climate change at the country level at short and longer horizons. Our approach is anchored in the empirical analysis of the relationship between climate and economic output in the past and makes use of climate models to project future climate. Our method approach consists of three steps.

First, we show how to best capture the relationship between climate and economic growth at the country level and at different horizons. Our approach follows Akyapi, Bellon and Massetti (2022) (ABM) to account for the need to capture non-linearities (e.g., the distinct role of extreme weather episodes occurring locally and for short periods within a year). We start with the dataset constructed in ABM collecting many climate variables that were put forward by the climate literature and that can reflect various moments of the distribution of weather within a country and a year.

Because the large number of climate variables would lead to over-identification issues in standard OLS estimators, we follow ABM and show how to us machine learning techniques to select the relevant climate variables that can best explain variations in economic growth. In this paper, we focus on country projections whereas ABM focused on examining the importance of weather shocks globally in historical data. Therefore, we implement the Least Absolute Shrinkage and Selection Operator (LASSO) at the country level, on arguably homogenous country groups, and explore the use of Classifier-LASSO (C-LASSO) – a variation of the LASSO that identifies relevant explanatory variables for groups of homogeneous countries and determines an optimal grouping of countries simultaneously (Su, Shi, and Phillips, 2016). There is a clear trade-off between selecting relevant climate variables in small versus large groups: implementing the LASSO in larger groups has the benefit of relying on more observations but comes at the expense of restricting the extent of heterogeneity within the full population. The C-LASSO allows us to find the optimal grouping that can capture country specificities while retaining some power.

Second, we discuss and propose a simple method on how to best recover climate change effects from historical data. Examining annual variations in weather and their effects has the advantage of leveraging long time series and plausibly random weather shocks. However, the effect of these short-term weather shocks may be different from the effect of a long-term change in average conditions (for example a temporary increase in annual weather of 1 °C may differ from the effect of a permanent increase in average temperature of 1 °C).

Finally, we use our empirical evidence and climate scenarios to illustrate the impact of observed trends and projections for a representative set of countries, illustrating different layers of uncertainty.

 

References

Su, Liangjun, Zhentao Shi, and Peter CB Phillips. "Identifying latent structures in panel data." Econometrica 84, no. 6 (2016): 2215-2264.

Akyapi, B., M. Bellon and E. Massetti. 2022. “Estimating Macro-Fiscal Effects of Climate Shocks from Billions of Geospatial Weather Observations.” IMF Working Papers 2022.156, International Monetary Fund, Washington, DC.

 

 

Short Bio: Emanuele Massetti is Senior Economist at the Climate Policy Division of the Fiscal Affairs Department of the International Monetary Fund. Before joining the Fund in 2021, he was Associate Professor at the School of Public Policy of the Georgia Institute of Technology in Atlanta, USA. Previously he was Postdoctoral Fellow at the Yale School of the Environment and Senior Researcher at the Euro-Mediterranean Center on Climate Change. While his research and policy analysis now focus on climate change impacts and adaptation, Emanuele has many years of experience in climate change mitigation policy. He contributed as Lead Author to the Fifth Assessment Report of the Working Group III of the International Panel on Climate Change, he was a member of the Committee on Carbon Utilization of the National Academy of Sciences, Engineering, and Medicine, and he is co-author of the Integrated Assessment Model WITCH.

 

Looking forward to seeing you!

 

ESM Research Team