Multi-country modelling at the commission of the European communities: Centralised and decentralised approaches

Since the early days of the European economic and monetary integration, European institutions have faced policy challenges requiring an assessment of the economic interdependencies across countries and the spillover effects of national policies. Some of these challenges have been addressed by developing multi-country models. By looking at the origins of multi-country modelling, we highlight the existence of a long-standing dilemma: the choice between the “centralized” approach, relying on homogeneous specifications of national economies and the “decentralized” approach, which “links” heterogeneous pre-existing national models. This article documents the debates over these two approaches, as the dilemma was confronted by macroeconomists working for the Directorate General of Economic and Financial Affairs (“DG II”) in the 1970s and 1980s. We explore the intellectual and institutional reasons that brought DG II to follow the decentralized approach for their first in-house multi-country model (“EUROLINK”), and later caused it to abandon this model in favour of a centralized approach. We also document the innovative solutions proposed by European economists to model trade linkages across countries, and their struggle with formalizing financial flows. Overall, this case study sheds light on the issues at stake when modelling economic interdependencies.