Stephen Gardbaum has posted this draft on SSRN (forthcoming, American Journal of Comparative Law). Here is the abstract:

This article aims to show that whatever the formal arrangements on the separation or “fusion” of executive and legislative powers — whether presidential, parliamentary or semi-presidential — the way any constitution operates in terms of concentrating or dispersing power is significantly a function of both the political party and electoral systems in place. They can not only fuse what a constitution’s executive-legislative relations provisions separate, but also separate what they fuse. As a result, the same set of institutional relations can operate quite differently in separation of powers terms depending on party and electoral system contexts.

In so doing, the article broadens and deepens the insight that the original Madisonian framework of institutional competition between the President and Congress has been rewritten by the subsequent, unanticipated development of the modern political party system, so that concentration or dispersal of political power — unified or divided government — depends mostly on electoral outcomes. It broadens the insight by showing this is true of all forms of government and not only the U.S. presidential system. It deepens it by drilling down one layer further and taking into account how party systems and electoral outcomes are themselves affected by the method of voting employed.

The article seeks to counter the tendency of constitutional lawyers to focus on inter-branch relations alone and to overlook other important institutional variables in thinking about separation of powers and constitutional design more generally. It also aspires, through the use of comparative and historical examples, to enhance our understanding of the U.S. system of “separation of parties, not powers.”