Introduction
Since World War II, preferential trading arrangements (PTAs) have become increasingly
pervasive features of the international economic system. Scores of these institutions have been
formed during this era and almost every country currently participates in at least one. PTAs are a
broad class of international commercial agreements that include common markets, customs
unions, free trade areas, as well as other arrangements. All of these arrangements require
members to mutually adjust their trade policies (usually through domestic legislation), granting
each member preferential access to the others’ market. A great deal of research has addressed
the economic consequences of preferential arrangements, but far less effort has been made to
identify the factors leading states to enter them, especially the domestic influences on PTA
formation. Recently, however, a number of studies have started filling this gap. Some have
emphasized the role played by interest groups (Maggi and Rodriguez-Clare 1998; Grossman and
Helpman 2002); others have focused on the effects of a country’s regime type (Mansfield,
Milner, and Rosendorff 2002).
We argue, however, that the number of “veto players” is an equally important influence
on whether states enter PTAs. Veto players are institutional or partisan actors whose consent is
needed to alter policies. Although the impact of these actors on PTA formation has not been
studied to date, existing research indicates that it is difficult to forge international agreements
when leaders confront an array of domestic groups with diverse preferences and the ability to
block policy initiatives (for example, Mo 1993; Milner and Rosendorff 1996; Milner 1997; Tarar
2001). Much of this existing research has focused on divided government, a situation where the
majority party in the legislature differs from that of the executive. Legislatures, however, are
2
Convention