Centre de formation sur l’environnement et la société

Accueil > Annuaire > Anciens > GROTH, Andreas > Projects > Current project

Current project

Vulnerability of Economic Systems to Climatic Impacts

supported by Réseau de Recherche sur le Développement Soutenable (R2DS) of the Région Ile-de-France

Principal Investigator : Michael Ghil
Partner : Centre International de Recherche sur l’Environnemnt et le Développement (CIRED), Nogent-sur-Marne
Department of Economics "S. Cognetti de Martiis" and Department of Economic and Financial Sciences "G. Prato", Turin, Italy

This project aims to shed light on the vulnerability of economic systems with respect to climatic impacts, including extreme events. The world economy typically exhibits complex dynamical behavior on different time and space scales and there are numerous difficulties in trying to study this behavior across the scales, as well as the coupling between the economic and climate system. In this project we will concentrate on the analysis of macroeconomic indicators in order to better understand the underlying dynamics of economic subsystems. In general, macroeconomic indicators, such as the gross domestic product (GDP), exhibit a long-term upward trend (Solow, 1956), on which shorter-term fluctuations are superimposed. There is substantial evidence for these fluctuations exhibiting a recurrent or cyclical structure, referred to as business cycles. To understand the response of economic systems to exogenous impacts, including those of climatic and other natural hazards, requires one to understand the origin of these business cycles. The main goal of this project to understand socio-economic consequences of natural disasters and the dynamics of reconstruction. We aim to raise questions about the assessment of climate change damages or natural disaster losses that are based purely on long-term growth models. Using the non-equilibrium dynamical model (NEDyM) of Hallegatte et al. (2008), Hallegatte and Ghil (2008) found a remarkable “vulnerability paradox“ : the indirect costs caused by extreme events during a growth phase of the economy are much higher than those that occur during a deep recession. This apparent paradox, however can be easily explained. Disasters during highgrowth periods enhance pre-existing disequilibria. Inventories are low and cannot compensate the reduced production ; employment is high, and hiring more employees induces wage inflation ; and the producer lacks financial resources to increase his/her investment. The opposite holds during recessions, as mobilizing investment and labor is much easier. The motivation of this project is to provide an experimental verification of a phase dependent vulnerability and to draw conclusion about the sensitivity of economic system with respect to exogenous shocks. Vice versa with the knowledge of concrete extreme events and the economic response we aim to gain information about the underlying dynamics. This means to overcome the gap between the output of such a toy model as NEDyM and the complex macroeconomic data. Although the model does reproduce typical business cycle behavior, it is still too simple to describe the complete state of a national, let alone global economy. On the other hand, it is precisely its simplicity that allows us to better understand the significance of its parameters. The fluctuations of the macroeconomic indicators, however, are variable in amplitude, period, and shape ; they are nonstationary and exhibit a high degree of complexity. Due to these substantial differences between the NEDyM model and the data available to calibrate its parameters it is not possible to directly assimilate the data into the model by standard algorithms, such as the Kalman filter and its various generalizations and extensions (Ghil and Malanotte-Rizzoli, 1991). Therefore, in the preceding CNRS-supported project of Andreas Groth, “Climate change and coupled climate-economics modeling,” the problem of suitable data preparation has been addressed by extracting an average business cycle for the U.S. economy, with all its variables besides GDP (Groth et al., 2009) ; such a smooth, suitably averaged cycle is easier to assimilate into an idealized model like NEDyM. Unsurprisingly, a gap remains which requires also a further refinement of the model. Moreover, this project aims to generalize the question about vulnerability to a scenario of multiple countries. In a strongly connected monetary union, such as the European Union, it is also of great interested to gain information about the individual’s country vulnerability in a composite of multiple countries in order to find a common policy that will be optimal for all countries concerned. Therefore, questions about synchronization of business cycles and common stylized facts arise and are still controversial (De Haan et al., 2008). We propose to help resolve these controversies by applying the methods of mutivariate singular spectrum analysis (M-SSA, Ghil et al., 2002) and synchronization of chaotic oscillators (Boccaletti et al., 2006) to draw conclusion about common or different behavior of individual countries. Since different countries can be effected by different climate events, this allows us, further, to cross-validate our coupled climate-economic findings.