Abstract
The proliferation of regional blocs has led to extensive research on regional integration arrangements (RIAs) in general and the European Union (EU) in particular. Yet, there is little work that analyses the integration process itself and its institutionalisation within a single analytical framework. In addition, existing theories tend to analyse integration and regional institution building as exogenous outcomes of government or societal preferences rather than as endogenous choices determined by governmentconstituent interaction. To address these shortcomings, we propose a political economy model that explains the causes, dynamics and policy-making procedures of RIAs as endogenous outcomes of government- constituent and government-government interaction. Extending Hotelling’s spatial competition model, we derive four hypotheses: (i) policy convergence precedes intergovernmental bargaining for regional institution building; (ii) governments establish regional institutions as commitment devices that would codify convergent policy choices; (iii) the relative incidence of supranationalism and intergovernmentalism is a function of constituent loyalty volatility and the barriers that governments can erect against loyalty shifts; and (iv) the deepening of integration is a function of endogenous and exogenous shocks that affect the way in which government-constituent interaction unfolds. The relevance of these hypotheses is discussed in the light of the existing evidence on the European Union experience.