Abstract
Over the last decade, students of the welfare state have produced an impressive body of research on retrenchment, the dominant thrust of which is that remarkably few welfare states have experienced fundamental shifts. This article questions this now-conventional wisdom by reconsidering the post-1970s trajectory of the American welfare state, long considered the quintessential case of social policy stability. I demonstrate that although most programs have indeed resisted retrenchment, U.S. social policy has also offered increasingly incomplete risk protection in an era of dramatic social change. Although some of this disjuncture is inadvertent—an unintended consequence of the very political stickiness that has stymied retrenchment—I argue that the declining scope of risk protection also reflects deliberate and theoretically explicable strategies of reform adopted by welfare state opponents in the face of popular and change-resistant policies, a finding that has significant implications for the study of institutional change more broadly.