An integrated approach for the estimation of agricultural drought costs
with Gabriele Standardi (CMCC) and Andrea Staccione (CMCC and Ca’ Foscari University). Work in progress.
The objective of this research is to assess the overall (direct and indirect) macroeconomic effects (%GDP) of agricultural drought impacts in Italy. through a coupled CGE-econometric approach. fAPAR-based statistical crop models for major Italian crops will be fitted at a local-level (Regione Agraria) and subsequently aggregated (spatially-weighted) at the Regional (NUTS-2) level. Regional economic impacts will feed a regionalised Computable General Equilibrium model specifically calibrated for the Italian economy. Crop-specific direct and indirect impacts will be evaluated at different dry and exceptionally dry growing seasons: 2003, 2006 and 2011. This study will shed light on the country-wide characterisation of this hazard using spatially detailed agricultural and drought data and will be a valuable tool to develop local-level drought risk management plans.
Data and Methods
Regionalised Computable General Equilibrium (CGE) models helps us to perform economic assessments of environmental direct and indirect impacts with a finer spatial resolution compared to that offered by standard CGE models and, in doing so, to increase the comparability of and the possibility to exchange information across economic and physical impact models. Operationally the sub national development of ICES follows a stepwise procedure. The starting point is the GTAP 8 database (Narayanan et al., 2012). It comprises a collection of Social Accounting Matrices (SAMs) for 57 economic sectors and 134 countries (or groups of countries) in the world. The calibration reference year is 2007.
Statistical crop models
The fraction of Absorbed Photosynthetically Active Radiation (fAPAR) is recognised as an excellent indicator of the vegetation health status of the plant, reflecting faithfully variations in the water assimilation and irrigation changes derived from water scarcity periods. fAPAR-based statistical crop models at the Regione Agraria level based on linear econometric techniques will be built for each considered dry season. The estimated direct impacts will be later aggregated at the regional level and fed into the CGE model.
Integrated impacts of droughts
The integrated approach described in this paper proposes the combination of two modelling techniques to estimate the overall (direct and indirect) effects of droughts in the country-wide economy. It is conceptually divided into four parts: i) the local characterisation (duration and intensity) of the drought events; ii) the estimation of local-level, direct impacts on crop yields using statistical models calibrated for each crop; iii) the (area-weighted) aggregation of those local-level impacts to the regional NUTS2 level and their conversion of crop yield losses into factor productivity drought shocks for each production sector and NUTS2 administrative unit and; iv) the computation of the indirect impacts at the sectoral, regional and country-wide level with CGE simulations.