Jacob Hacker's notion of 'drift' is an important addition to the conceptual toolkit of institutionalists and welfare state analysts concerned with change as well as continuity. Drift occurs when policy fails to adapt to changing patterns of social risk either because of the 'stickiness' of welfare institutions, or as a result of 'deliberate efforts by political actors to prevent the recalibration of social programs' (Hacker 2005: 46). While many institutionalists concerned with policy change have focused particular attention on ideas (inter alia Blyth, Cox, Hay, Schmidt), they feature less explicitly or prominently within Hacker's framework. This paper brings together 'ideational' perspectives with the analysis of policy drift through a comparative analysis of Early Childhood Education and Care (ECEC) in three 'liberal welfare regimes'. Although existing accounts of drift are predominantly concerned with the (non) adaptation of existing policies to new circumstances, social-structural change can also create needs that could only be met by wholly new policies. Indeed, Hacker himself notes that 'US social policy could have adapted to changing social realties
some nations
have dramatically expanded public protections that help women enter the labour force and balance work and child rearing
however, the US clearly did not follow this path.' (2005: 54). But this argument is too sweeping. Here I excavate and analyse attempts to follow this sort of path in the US and two other 'liberal' states.First, the paper insists that purportedly 'new' risks are rooted in social (gendered and socio-demographic) trends that date back to the 1950s and 60s. Second, drawing on Barrington Moore's classic, but neglected concept of 'historically suppressed alternatives' (1978) it retrieves US, Australian and UK political and policy initiatives that sought or seek to address the new risks associated with change gender roles. Despite expectations that liberal regimes would treat ECEC as largely a private/market matter, the paper shows that each of these countries has each played the role of ECEC 'leader' during different decades since the 1950s (first the US in the 50s and 60s, then Australia in the 70s, and particularly 80s and early 90s, and the UK since the late 90s). This pattern - both the achievement of and (at least in two cases) decline from leadership - requires explanation. The paper focuses particularly on two aspect of the role of ideas. First their international currency - the manner in which ideas developed in one state can help to shape debate and policy elsewhere - especially the early - and enduring - influence of US policies and policy ideas (especially those surrounding the Head Start programme) in Australia and the UK. Second, it argues that the existence or non-existence of ideas is rarely crucial - apparently influential policy ideas have been present in and since the 1950s and 60s in all three states. Indeed, elements of policy that attempt to 'counter' drift induced by changing gender roles have also been present in them (some, like Head Start apparently well entrenched). As Crouch and Keune (2005) note, states' policy configurations can contain diverse - even contradictory - elements (or as TH Marshall classically argued they can 'make a square meal of a stew of paradox'). Much more than the existence of particular ideas or even individual policies, drawing on work by Blyth and Hay I argue that to address changing social risks requires the skilful articulation of particular ideas and policies into a broader political project. ..PAT.-Unpublished Manuscript [ABSTRACT FROM AUTHOR]