Faculté des sciences économiques et sociales

Economics of war-torn countries : a politico-economic approach applied to Guatemala

Carbonnier, Gilles ; Pult, Guido (Dir.)

Thèse de doctorat : Université de Neuchâtel, 2001 ; 1567.

The overarching objective of this doctoral research is to integrate key political and conflict-related variables in the analysis of war-torn economies. For, as we conclude from a critical review of the literature on war, reconstruction and the economy, politics must be fully taken into account in economic analysis and planning. This is especially true during the fragile transition from war to... Plus

Ajouter à la liste personnelle

Autres fichiers

    Summary
    The overarching objective of this doctoral research is to integrate key political and conflict-related variables in the analysis of war-torn economies. For, as we conclude from a critical review of the literature on war, reconstruction and the economy, politics must be fully taken into account in economic analysis and planning. This is especially true during the fragile transition from war to peace to the extent that rebuilding efforts are vain if the conflict rekindles. In this dissertation, we elaborate on two complementary methodological approaches aiming at including distributional and political concerns in economic research on conflict-ridden countries. We test them in a case study on Guatemala and discuss their respective merits and limitations. The dissertation first focuses on the social accounting framework as a tool for distributional analysis. We construct a social accounting matrix (SAM) for Guatemala, based on which we calculate accounting multipliers. Multiplier effects are decomposed into transfer, open-loop and close-loop effects following the logic of the circular flow of income. The ensuing SAM-based model serves to simulate the distributional impact of several fiscal reform options on different socioeconomic groups. This in turn helps design politically-informed economic reform. We then present an analytical model that enables to capture the fundamental interactions between the economic and political spheres. We address in particular the dilemma posed by the diverging agendas of peace-building and economic reform, both of which are required in the aftermath of a protracted conflict to put the economy back on track and consolidate peace. The analytical framework highlights how the optimal policy mix from a government's perspective differs from the economic optimum advocated by the international financial institutions. Subsequently, a macroeconometric model is built, and tested in the context of our Guatemalan case study which requires the construction of ad hoc political and economic data sets. We then ask how coherent are the different responses devised by the international community to assist war-torn countries rebuild themselves. To this end, we run simulations of external interventions in the form of increased conditionality related either to economic reform or to peace-building. The outcome highlights stark contradictions between these two types of conditionality because of their induced effects in the economic and political spheres, thus calling for increased coherence in the global response of the international community. The dissertation is structured under three modules: critical literature review, distributional analysis, and politico-economic interactions. The originality of this doctoral research lies in the inclusion of conflict- and peace-related variables in distributional analysis and macroeconometric modelling. In the case of Guatemala, the model proves to have a stronger forecasting ability than traditional models limited to economic variables only. The politico-economic approach elaborated in this dissertation offers an avenue for further research on war-torn economies, all the more so that the major methodological obstacles are being removed: the lack of reliable data is being addressed by a more systematic collection of information on violence and conflicts while recent econometric software packages considerably ease the computational process.