In this paper I am taking issue with the literature modelling inter-institutional relations in the European Union either as principal-agent relations or in terms of transaction cost economics.
The intuition behind my critique is this: when a national minister sits in the Council, she does not try to discover either the best way to avoid meeting her peers too often (which would amount to economizing on transaction costs) nor the partisan profile of the relevant commissioner in order to delegate powers only to ideologically neighbouring commissioners (which would amount to solving a problem of adverse selection following the "ally principle"). In fact, the European supranational executive was never rationally designed to serve national ministers in any of these ways.
Through a careful re-examination of the history of the European Coal and Steel Community I show that the supranational executive was designed to be a complex web of checks and balances with a big core, thereby limiting problems associated with incomplete contracting and bureaucratic power.
