“Post-Feminist” Era of Social Investment and Territorial Welfare? Exploring the Issue Salience and Policy Framing of Child Care in U.K. Elections 1983-2011

Earlier work has tended to overlook the formative origins of child care policy in liberal democracies. Accordingly, this study examines mandate-seeking and parties’ envisioning of child care with reference to issue salience and policy framing in party manifestos in U.K. Westminster and regional elections. It reveals a significant increase in issue salience following its emergence as a manifesto issue in the 1980s, thereby confirming it as part of the wider rise of “valence politics.” The framing data reveal that a “post-feminist” discourse of “social investment” has generally displaced the political framing of child care as a gender equality issue. It is argued that this is inherently problematic and reflects parties’ failure to address ongoing gender inequality in the labor market. Notably, the data also illustrate the way devolution is leading to the territorialization of child care in the United Kingdom—no longer solely mandated in Westminster elections, policy is now contingent on the discursive practices of regional party politics and shaped by local socio-economic factors.


Introduction
Effective child care policy is integral to overcoming gender segregation in the labor market and is an internationally held policy goal in governments' pursuit of economic growth (cf.Picchio, 1992;Seguino, 2000;Tzannatos, 1999).It can be defined as "a care service for children 0-5 years to cover working hours, provided by nursery nurses or unqualified care staff in a variety of settings including domestic settings" (Penn, 2000, p. 37).As a policy issue, it spans a number of key issues and debates including gender equality, the domestic division of labor, appropriate development and support for children, the contemporary role of the welfare state, the form and functioning of the labor market, and the wider health of the economy (cf.Saraceno, 2011).Although leading studies have charted the changing political context shaping child care policy development in the United Kingdom over recent years (Harker, 2006), the present study makes an original contribution by analyzing electoral discourse and the level of attention ("issue salience") and use of language ("policy framing") in relation to child care in party manifestos for Westminster and "regional" elections. 1 It is a transferable method suited to future comparative study of social policy developments in (quasi-)federal welfare regimes.Accordingly, the discourse-based process of mandate seeking in election programs matters to the development of child care, because it constitutes the link between the representative process and policy development in liberal democracies.In this, manifestos serve multitude of functions: (a) They provide substantive details of future government (and opposition) parties' policies, (b) they show how parties compare in the priority they attach to child care, (c) electoral discourse reveals areas of inter-party conflict and consensus, and (d) such a focus provides insights into how policy is shaped by party ideology and contingent on local socioeconomic and political factors.In this regard, it reveals the political use of language and discourse-based processes that underpin the development of public policy on child care, thereby providing a "discursive benchmark" to complement ex post analyses of policy delivery (Baslé, 2006;Meyers, Glaser, & Donald, 1998).In short, it is a focus that engages with Sumsion and Press's (2007) prescient call

SAGE Open
What there is to be learned from critical analyses of policy trends . . .and how we both deconstruct and reconstruct the policy landscape.(p. 96) The current focus acknowledges the role of electoral discourse as an important indicator of political agenda setting (Cobb & Ross, 1997).Moreover, as Rigby, Tarrant, and Neuman (2007) cogently observe, attention to discourse and policy design recognizes the way that both "institutionalize and legitimize particular forms of governmental involvement in children's lives-as well as give power and voice to some interests over others-resulting in a new political context for future policy debates" (p.98).Such an approach also provides insights into parties' attempts to appeal to particular audiences at the time of elections.In turn, this shapes wider voting patterns and determines which policies are endorsed by the electorate.
The latter is explained by mandate and accountability theory (Budge & Hofferbert, 1990;Fearon, 2003;Royed, 1996).The former asserts that when in government, parties should implement the policies that they pledged when running for office.In contrast, accountability theory asserts that elections are effectively "opinion polls" on the performance of the party (or parties) forming the previous administration-and whether they delivered their manifesto policy program (Przeworski, Stokes, & Manin, 1999).
Two non-discrete contemporary factors underline the importance of the foregoing theoretical underpinnings: devolution and the rise of coalition government.In the former case, study of child care politics needs to be cognizant of the discursive underpinnings of distinctive territorial approaches.This links to earlier pioneering study (Randall, 2000) that underlined the limits for child care policy convergence between (welfare) states in multi-level systems-as well as scholarship on how (quasi-)federal state architecture affects the design of welfare (Mahon & Brennan, 2013).Here, we complement this work by examining the impact of the pluralizing of electoral systems that accompanies state decentralization-such that single statewide ballots are supplemented by regional elections.With regard to the second factor, whereas the current Westminster coalition government is something of a rarity at a U.K. level, multi-party executives have become a routine aspect of devolved government in the United Kingdom.Electoral discourse thus plays a key role in constructing coalition agendas for government as the respective partners seek to integrate party-specific election pledges on child care into a single executive policy program (cf.Leitner, 2010).
In summary, the following discussion explores the contemporary development of child care policy by (a) exploring changes in the issue salience of child care since its emergence in the 1980s, (b) examining policy framing in manifesto discourse, and (c) analyzing the impact of state decentralization.Accordingly, the remainder of the article is structured thus: A discussion of the literature on the child care policy and electoral competition is followed by an outline of the research context and methodology.The findings in relation to the study aims are then presented in two principal sections: statewide and regional elections.In the latter case, the discussion is sub-divided, the first section of which deals with inter-polity contrasts in issue salience and framing, and the second with inter-party contrasts in issue salience and framing.The main findings and their implications are discussed in the conclusion.

Electoral Politics and the Formative Phase of Child Care Policy Making
Over several decades, child care provision, including paid parental leave, has been an increasingly common feature of social policy in many European countries and beyond (Melhuish & Petrogiannis, 2006).In contrast, in the United Kingdom, child care policies have only developed since the late 1990s.Hitherto, provision was left to voluntary groups or the market; statutory maternity leave was limited and state involvement centered on vulnerable children and those in residential care.However, structural pressure on the "male breadwinner model" on which the welfare state was premised-not least by the increasing prevalence of dual-earner households-has, albeit belatedly, prompted government intervention (Land, 1999).This has also been driven by EU policy, notably the goal that by 2010, women's participation in the labor market should have reached 60%, and that child care services should be available to at least a third of children below 3 years of age and 90% of children aged between 3 and school-entry age (cf.Plantenga, Remmery, Seigel, & Sementini, 2008).Although existing accounts chart recent developments in government policy on child care (cf.Rahilly & Johnston, 2002;Scheiwe & Willekens, 2009;Smith, 2007), limited attention has been given to the formative phase of policy making and its electoral underpinnings.
When attention has focused on child care policy it has largely examined party divisions on child care (Leitner, 2010), internal party debates on the funding of child care programmes (Brennan, 1998), addressing opposition to parties' child care policy (Eisenstein, 1981), the interplay between structural and ideational factors in establishing party platforms (Sorensen, 2011), and sex discrimination and setting party election programmes (Howell, 2006).Little attention has been given to the longitudinal study of electoral discourse and the prioritization and framing of policy proposals on child care.
Accordingly, to address this lacuna, the following draws on the electoral theory of "issue salience" (RePass, 1971;Robertson, 1976).This is a conceptualization whereby pivotal importance lies not on party issue positions but on the prominence and attention afforded to different issues in their campaigns; ergo the more an issue is emphasized by a party (making it "salient"), the greater the likelihood it will attract voters who share similar concerns.Traditionally, quantitative analysis has been used to explore this (Libbrecht, Maddens, Swenden, & Fabre, 2009;Volkens, 2001).The present examination takes a more holistic approach by combining this with an exploration of policy framing.Frames here are "a necessary property of a text-where text is broadly conceived to include discourses, patterned behaviour, and systems of meaning, policy logics, constitutional principles, and deep cultural narratives" (Creed, Langstraat, & Scully, 2002, p. 37).
In electoral theory, child care is a valence issue (Stokes, 1992).In other words, it is a topic that generally unites voters (given its wider social benefits, few would argue against the regulated provision of child care).However, it is also a "position issue," meaning that parties differ in their views on what public policy should-and should not-aim to achieve, not least in relation to the amount of resources and support extended by government.Inter alia, parties' contrasting issue positions reflect their ideological standpoint on the appropriate balance between the market and state (Barnett & Barnett, 1997).Traditionally, those on the Right have embraced neoliberal, market-based solutions and eschewed state intervention; whereas those of the Left have advocated the harnessing of state power to promote social welfare (Navarro, 1998).In reality, party positions are rarely as polarized as this suggests, as evidenced by the literature on welfare pluralism (Beresford & Croft, 1983).Underlining the latter point, over the past two decades, U.K. politics has been marked by the rise of valence politics, as the main parties have converged toward the political center ground (Bara & Budge, 2001;Bromley & Curtice, 2002).The present findings showing the steep rise in issue salience over the past three-and-a-half decades are significant because they confirm child care's role in the rise of valence politics.
In turn, this increase in salience is a function of the increasing party politicization of child care.This refers to how issues rise and fall on the political agenda, as parties compete for votes on a given issue (Carter, 2006).It is allied to the concept of "issue ownership" (Petrocik, 1996)whereby parties prioritize certain policy issues, emphasize earlier policy successes, and attempt to highlight party competence on a given topic while dismissing rivals' records.The underlying motive is to be seen as the "owners" of an issue-thereby securing electoral and reputational advantage.
By focusing on statewide and regional elections, 2 this study provides a transferable methodology that gives further insights into the impact of multi-level governance on child care policy making.This approach is appropriate, because "devolution"-or move to quasi-federalism in the United Kingdom (Gamble, 2006)-is part of the wider international trend of state restructuring (Doornbos, 2006).Under the revised governance structures, the Scottish, Welsh, and Northern Irish governments have responsibility for many aspects of state child care policy in their territories 3 (for a discussion of the policy responsibilities of the devolved legislatures and governments, see, for example, Birrell, 2008).

Method
By applying mixed research methods, the current study responds to earlier calls for policy work to combine content and critical discourse analysis (Baker et al., 2008).Accordingly, issue salience is determined by content analysis of manifestos.This is applied by recording the number of incidences of key words, ideas, or meanings in party programs (Krippendorff & Bock, 2008) and is complemented by frame analysis (Schön & Rein, 1994).The latter is concerned with how, as key political texts, manifestos enable parties to construct (or "frame") policy proposals on child care and other matters.In electoral terms, as Nelson and Oxleya (1997) observe, "frames influence opinions by stressing specific values, facts and other considerations, endowing them with greater apparent relevance to the issue than they might appear to have under an alternative frame" (p.75).In this way, framing leads to political agenda setting (Cobb & Ross, 1997;Cohen, 1963) and, ultimately, the substantive policies that are mandated.
Comparative analysis of framing practices in different polities and tiers of government is a long-established methodological practice (De Vreese, Peter, & Semetko, 2001;Papacharissi, 2008). 4However, it is acknowledged that manifestos have limitations as a data source, not least because party policy proposals are also set out in debates, speeches, and other documents; yet, they constitute the principal political texts that reflect a party's priorities and issue positions, thereby allowing systematic analysis over time.Accordingly, as noted, electronic versions of the manifestos of the leading 5 parties in U.K. general elections 1983-2010 and regional elections 1998-2011 were analyzed using appropriate software. 6 Thus, in the preliminary stage of the research, the manifesto texts were divided into "quasi-sentences" (or "an argument which is the verbal expression of one political idea or issue," Volkens, 2001, p. 96). 7Dividing in this manner controlled for long sentences that contain several policy proposals.Individual quasi-sentences were subsequently coded using a deductive coding frame (Joffe & Yardley, 2003) based on key topics and themes derived from the academic literature on the child care policy (see Figure 4).Thus, this schema incorporates a range of frames including mixed economy provision, work/life balance, encouraging workplace provision, regulation/child care staff training, and gender equality.Divergent views on the coding emerged in less than 2% of quasi-sentences (N = 738 8 ; resolved by discussion between coders).Issue salience was then determined by logging the frequency of quasi-sentences in a database of party manifestos.As existing electoral studies reveal, over recent years, party programs have tended to become more detailed and have a greater word length.This has potential methodological implications for any claims of shifting salience over time, not least because any change might be regarded as a possible function of increased manifesto length rather than greater attention to child care policy by the respective parties.To control for this, the present analysis uses both "absolute" and "relative totals" methodologies.The former details the total number of quasi-sentences on child care; whereas the latter re-calculates them as a percentage of all quasisentences in each manifesto 9 (i.e., quasi-sentences on all topics and issues; see Figure 2 below).Because the impact of increased manifesto word length on saliency is complex and variable, 10 both approaches have advantages and shortcomings.Moreover, notwithstanding the overall trend toward greater manifesto length, there are major fluctuations in both manifesto word totals and the number of quasisentences (for example, in two of the seven election cycles studied here, the total number of quasi-sentences in the manifestos studies actually decreased compared with the preceding ballot).Nevertheless, to control for any discrepancies between the two methods, as noted, both were used in the following analysis.This dual approach affirmed that the "absolute totals" method (i.e., exploring the changing totals of child care quasi-sentences) produced findings consistent with those derived from the "relative proportion" method (thus, for example, they both reveal a significant increase in the issue salience of child care over time, see Figures 1  and 2).

Child Care Policy and Westminster Elections 1983-2010
A survey of the election manifestos in the first decades after the Second World War provides a clear indication of the main political parties' role in reproducing a gender-unequal society.It also partly explains the United Kingdom's failure to put in place effective state child care provision seen elsewhere in Europe prior to the 1990s.Thus, the present manifesto analysis adds further empirical basis to Lewis's (2013) conclusion that civil servants and ministers "were determined to defend the status quo in the context of weak political will" (p.256).Against this backdrop, the history of child care during the immediate post-war decades was underpinned by a British welfare model oriented around the male breadwinner and largely concerned with poverty reduction (Crompton, 1999).This is evident in the manifesto discourse of the political Right; for example, during maternity, All proper arrangements, both voluntary and state-aided, must be made for the care of other young children in the family, in order that the energies of the male breadwinner or the kindness of neighbours and relations, which nevertheless must be the mainspring, should not be unduly burdened.(Conservative Party, 1945, p. 8) It is also apparent on the Left.For example, "Liberals oppose the bringing into industry of married women with young children, but would not discourage schemes of industrial outwork, to help the family budget by work done at home" (Liberal Party, 1950, p. 14).
Child care finally emerges as a manifesto policy issue in Westminster elections in 1983.Notably, it is articulated as a gender equality initiative.The left-of-center Labour Party (1983) promises, "[We will] introduce positive action programmes to promote women's rights and opportunities, and appoint a cabinet minister to promote equality between the sexes.[Furthermore] We will: Improve child care and other social services" (p.17). 11Yet, it continues to attract limited attention in party programs until the 1992 election (just 3.3% of all references were recorded before 1992).Subsequently, there is a significant increase in salience; 12.7% of all references were made in the 1992 election.There are a number of explanations for this step change.One is the growing domestic and international political attention to gender equality (Mancini & O'Leary, 1999;Randall, 1996).In turn, this was driven by the earlier round of activism associated with "second-wave" feminism, a process that highlighted the domestic division of labor as an enduring source of inequality between the sexes (cf.Keetley & Pettegrew, 2005).Further impetus was derived from legislative developments (including the U.K.'s ratification, in April 1986, of the United Nations [UN] Convention on the Elimination of Discrimination Against Women [CEDAW]).This obliged government "to encourage the provision of the necessary supporting social services to enable parents to combine family obligations with work responsibilities and participation in public life, in particular through promoting the establishment and development of a network of child-care facilities" ( UN, 1979, Article 11, 2c).
On the back of these developments, the data reveal how child care becomes an increasingly important policy issue in U.K. electoral politics from 1992 onward.Thus, there is a threefold increase in the number of references to child care when the 1992 and 2010 elections are compared.It is a political (re-)prioritization confirmed by the contrasting methodological techniques used here.Specifically, it is revealed when absolute totals of child care quasi-sentences are examined (Figure 1)-and, it is also apparent when all parties' child care references are plotted as a percentage of total quasi-sentences (i.e., on all topics and issues) in each election (Figure 2).Thus, 3.3% of references were made in the 1980s, 19.4% in the 1990s, 36.2% in 2000s, and 41.2% in 2010.The data also reveal the party dynamics behind the overall increase in issue salience.In this regard, there are statistically significant differences between parties (p ≤ .001). 12Over the post-1983 period, Labour can be seen as "issue owners" of child care; the party accounts for more than two thirds (68%) of manifesto references (this compares with 23.3% for the Conservatives and just 8.6% for the Liberal Democrats).It is only belatedly that the Conservative Party has begun to challenge Labour's position.For almost two thirds (61.3%) of the party's references to child care were made in the 2005-2010 election cycle.This finding supports earlier analysis highlighting how the party has attempted to (re-)position itself as being more socially progressive than in the past (cf.Kerr, Byrne, & Foster, 2011).Examples of the associated politicking in the manifesto discourse include the following: "the current government has shown itself wholly insensitive to the need to help develop family-friendly working practices" (Labour Party, 1997, p. 38), and "under Labour,

SAGE Open
Britain has the most expensive child care in Europe, and many working families receive no help" (Conservative Party, 2005, p. 29).
Analysis of the manifesto data also reveals statistically significant differences in way that the three major parties frame policy proposals on child care (p ≥ .05). 13 Reflecting recent emphasis on multi-agency and cross-sector working (Anning, 2005), the lead frame overall is "mixed economy of provision."The Labour Party accounts for just more than a half (52.6%) of such references, followed by the Conservatives (31.6%) and Liberal Democrats (15.8%).
The discourse under this frame is typified by statements such as "we will support the commitment of community and voluntary groups to build up a diverse range of child carefrom Early Excellence Centres to neighbourhood nurseries and informal care" (Labour Party, 2001, p. 34).In this way, the discourse the development of the "Sure Start" program of children's centers to provide child care, early education, health, and family support introduced by the party in government in 1998 (cf.Melhuish & Hall, 2007).It is a stark reminder of Labour's shift away from its earlier emphasis on statist solutions to welfare issues (Pugh, 2011).It is also reflective of the wider rise of welfare pluralism, whereby voluntary and private-sector organizations complement state welfare delivery (Beresford & Croft, 1983;Fyfe, Timbrell, & Smith, 2006;Kendall, 2000).
While noting Campbell-Barr's (2009) cogent observation that "it is neither so simple or accurate as to suggest that the private sector is more likely to focus on the business side of things and the voluntary sector on the care side" (p.85), there is a statistically significant difference in the parties' use of the frame (p≤ .05). 14 For the Conservatives make the majority (71%) of explicit references to private-sector/for-profit child care as part of a mixed economy of provision, whereas Labour and the Liberal Democrats predominate in advocating third-sector partnership with public-sector providers (68%).For example, Labour (2010) asserted, "We want to strengthen parental engagement with Sure Start Children's Centres.Some voluntary and third-sector organisations already run networks of Centres, and we will now pioneer mutual federations running groups of local Children's Centres in the community interest" (p.57), whereas the Conservative Party (2010) discourse is typified by the following: We support the provision of free nursery care for pre-school children and we want that support to be provided by a diverse range of providers-including the many child-minders and private . . .and independent nurseries which are currently being squeezed out of the system.(p.48) Notwithstanding the foregoing contrasts, there is evidence of Right-Left convergence in the second frame, "tax allowance/social security payments."This reflects the wider adoption of "social investment" approaches to welfare by parties across the political spectrum in Europe and elsewhere (cf.Vandenbroucke, 2011).It is a policy approach that aims to secure future societal well-being through investing in children and is predicated on breaking the inter-generational transmission of inequality and disadvantage.Labour accounts for 42.1% of "tax allowance/social security payments" quasisentences, the Conservatives 36.8%, and Liberal Democrats 21%.Examples include, We will extend the exemption from tax which applies to workplace nurseries to all forms of employer assistance with childcare.(Labour Party, 1992, p. 29) During the next Parliament, we will ensure that all working families who qualify for the working tax credit will receive up to £50 a week for each child under the age of five, irrespective of the type of childcare they choose.(Conservative Party, 2005, p. 38) As Thorpe, Millear, and Petriwskyj (2012) observe, Concern to ensure that all children have access to high-quality educational experiences in the early years of life has instigated policies to increase the qualifications of staff in the childcare workforce, and in particular, to increase the number of degree qualified teachers.(p.317) Thus, both main parties place emphasis on framing child care in the context of "vocational training/return to work support."In this, Labour accounts for 57.1% of references, with the remainder by the Conservatives.Earlier examples exhibit gendered language use that is less evident over later election cycles.For example, Expanded childcare will help women return to work and undertake training.A critical task is to upgrade the skills of people in work.Training and Enterprise Councils will be retained, reformed and made more broadly representative of their local communities and given stable budgets.(Labour Party, 1992, p. 31) Parents who have taken time out from their careers to care for their children will be able to apply for a scholarship to help them undertake vocational or professional training.(Conservative Party, 2001, p. 39) A striking aspect of the discourse is the limited number of quasi-sentences framed in the context of gender equality (1.3%).The majority of examples are from the three election cycles following the emergence of child care as a manifesto issue in 1983.This is significant.It suggests a "post-feminist" shift in framing whereby gender equality is displaced in the electoral discourse by a series of frames that emphasize child care as a mode of "social investment" (Hübenthal & Ifland, 2011, p. 114)-or policy intervention that "will produce pay-offs for the society's economy, as well as for the individual child" (pp. 114-115).In this, the state's role is defined more in terms of strategic coordination around wealth creation and child-safety regulation than the normative goal of gender equality and ending sex discrimination.
Examples of the earlier gender equality discourse include the following: we will "strengthen the rights of women.We will require public authorities and private contractors holding public contracts to be equal-opportunity employers and improve child-care support and facilities" (Conservative Party, 1997, p. 18), and "women and opportunity-we believe mothers should be treated equally by government, whether they work outside the home or not . . .we will act where a push by government is needed to stimulate the provision of childcare" (Conservative Party, 1992, p. 29).

Child Care Policy in Regional Elections 1998-2011
Following devolution in the United Kingdom, child care policy has been subject to a sharp increase in issue salience.There has been a fivefold increase when the number of quasisentences in the 1998/1999 elections is compared with that of 2011.A further indication that meso-elections have presented significant opportunities for child care policy development is the fact that in little more than a decade, almost twice as many references were made in the regional ballots (1998)(1999)(2000)(2001)(2002)(2003)(2004)(2005)(2006)(2007)(2008)(2009)(2010)(2011) than in all 18 post-war statewide elections (N = 528 compared with N = 210).

Territorialization: Inter-Polity Contrasts in Issue Salience and Framing
In the wake of the United Kingdom's move to (quasi-)federalism, the territorialization of policy is shaped by statistically significant differences in issue salience as measured by the total number of child care quasi-sentences in the post-1998/1999 electoral discourse in each regional polity (p ≤ .001). 15Most references to child care were made in Wales (37.5%), followed by Scotland (35.8%) and Northern Ireland (7.2%; Figure 3).
Territorialization is also apparent in statistically significant inter-polity contrasts in policy framing (p ≤.05). 16In turn, this reflects ideational divergence at the regional level in parties' thinking about what future policy should achieve and the means by which it should be pursued.Figure 4 represents these differences graphically in the form of "framing profiles."For each of the three regional polities, a histogram displays the percentage of the manifesto discourse on child care falling under each frame.In turn, the contrasting shape, or "profile," of each underlines the differing ways in which policy is framed, or envisioned, in each territory.Key territorial differences revealed by this technique include the disproportionately strong emphasis placed on mixed economy approaches in Scotland, greater attention to tackling poverty and regulation of child care/staff training in Wales, and the relative emphasis on gender equality and workplace provision in Northern Ireland.
Overall, across polities, the lead frame is "mixed economy of provision" (24.3% of child care quasi-sentences).Manifestos in Scotland account for most references (46.6%), followed by Wales (40.5%) and Northern Ireland (13%; p ≤.001). 17Of particular significance here is the contrast with policy framing in statewide Westminster elections.Notably, significantly less attention is placed on the private sector's contribution to mixed economy provision at the regional level (just 28% of references were made in regional manifestos).

SAGE Open
This underlines the contingent nature of the formative phase of child care policy making and the way it is now directly driven by regional party politics in which the principal proponent of private "for-profit" provision, the Right-of-center Conservative Party, has significantly less electoral support.Thus, the historical basis for this contrast is Left-party strength in both Wales and Scotland where, in the post-war period, Labour and Nationalist Parties have tended to dominate.In these polities, there is correspondingly greater emphasis on third-sector provision.For example, "all pre-school, nursery and childcare provision will be offered under one roof.In a true public/private voluntary partnership councils will be encouraged to site centres in schools in rural areas or peripheral urban estates" (Scottish National Party [SNP], 1999, p. 14), and "through a National Early Years Strategy, local authorities will be encouraged to develop further partnerships, particularly with voluntary groups" (Plaid Cymru, 1999, p. 18)."Economic arguments/tackling poverty" is the secondranked frame (19.4% of all quasi-sentences).Compared with statewide elections (where it was the ninth-ranked frame, accounting for just 4.3% of all quasi-sentences), the greater propensity of parties to frame child care in this way reflects marked differences in regional economic performance and child poverty across the United Kingdom (Birch, MacKinnon, & Cumbers, 2010;Brewer, Browne, Joyce, & Payne, 2011).In part, its prominence in the regional discourse can be seen as a function of the fragility and comparative weakness of the economies of Wales and Northern Ireland, the greater contribution of the public sector to the regional economies (cf.Benneworth & Roberts, 2002;Dorsett, 2013), and the devolved administrations' prioritization of anti-poverty measures (Birrell & Heenan, 2010;Morelli & Seaman, 2009;Pedace, 2009).The territorialization of policy making is again evident, for there are statistically significant contrasts in the frame's use across polities (p ≤.05). 18Manifestos in Northern Ireland account for most references (41%), followed by Wales (36.2%) and Scotland (22.9%).Examples of the discourse include, "the SDLP views employment as a key route out of poverty which is why we must develop affordable and sustainable childcare policies that will allow many economically inactive people to return to the workplace" (SDLP, 2007, p. 33), and we will "build effective opportunities for individual and community wealth creation to reduce poverty and enhance personal fulfilment, including the expansion of childcare facilities, aiming to develop flexible childcare provision accessible to all" (Scottish Liberal Democrats, 2003, p. 17).
The regional manifesto pledges also neatly illustrate the wider issue of distributive or territorial justice in child care in the wake of devolution.This is a concept founded on two tenets: "who gets what where" (Smith, 1977) and "each according to the needs of the population of that area" (Davies, 1968, p. 16).As Kay (2005) explains, Territorial justice as conceived in the standard version refers to the principle that . . . the UK is a unitary state; it does not have sub-divisions of its territory that are subject to a different set of social rights.(pp.545-546) However, the discourse on child care shows how spatial variation in child care policy is driven not only by "devolved" policies but also by the regional administrations' adaptation of Westminster policies with extra resourcing and/or revised eligibility criteria.For example, . . .programmes such as the New Deal . . .We will support these UK programmes with additional resources [emphasis added] to provide childcare support in areas of high unemployment in order to help those in work, training or education make the most of the opportunities Labour has created.(Scottish Labour Party, 2003, p. 27)   "Regulation/training of childcare staff" was the fourthranked frame (10.7% of references overall).Most references were made in Wales (63.8% of the regional total), followed by Scotland (25.9%) and Northern Ireland (10.3%; p≤ .001). 19The geographical contrasts in the use of this frame again highlight the historically contingent way policy develops in the wake of the move to regional elections.The Welsh emphasis on regulation and training can be seen as the product of earlier official inquiries into failings in children's residential care, which has prompted a wider series of legislative and policy reforms encompassing child care (such as the creation of a children's commissioner's office, and Welsh Government adoption of the UN Convention on the Rights of the Child; cf.Pithouse, 2011).Examples of this discourse include, "an Early Years Taskforce will be established to develop . . .a national training framework for those working with young children, together with a simplified qualifications system" (Plaid Cymru, 1999, p. 14), and we will "recognise the need to develop the childcare workforce, provide more opportunities for training and update the information available on the sector" (Welsh Liberal Democrats, 2007, p. 29).Notwithstanding the territorialization of policy, the limited propensity of parties to frame proposals in terms of gender equality spanned in both tiers of governance; just 5.4% of all child care quasi-sentences came under this frame at the regional level.

Party Politicization: Inter-Party Contrasts in Issue Salience and Framing
In each polity, the data reveal inter-party contrasts in issue salience and the framing of child care policy.This matters because policy is now set by contrasting arrays of parties in each regional polity, each of which not only has its own vision of contemporary child care policy but also has held government office since 1998.Notably, the issue-salience data show that, across meso-polities, nationalist parties pay greatest attention to child care.Thus, in Scotland, the SNP accounted for most references to child care (38.6%), followed by Scottish Liberal Democrats, Scottish Labour, and Scottish Conservatives (21.2%, 20.6%, and 19.6%, respectively; p ≤.05). 20For the SNP and Scottish Liberal Democrats, the lead frame is "mixed economy provision" (35.9% and 30% of each party's child care references, respectively; for example, "the piloting of Children's Centres across Scotland . . . in a true public/private voluntary partnership councils will be encouraged to site centres in schools in rural areas or peripheral urban estates," SNP, 1999, p. 22).In contrast, the lead frame for Scottish Labour was "the economy/tackling poverty" (23%).
In Wales, Plaid Cymru accounted for most of the references (40.9%) to child care, followed by Welsh Labour, the Welsh Liberal Democrats, and Conservatives (29.5%,15.7%,and 13.8%,respectively).Reflecting its emphasis on community engagement/development, Plaid's lead frame is "mixed economy of provision" (36% of party quasi-sentences; for example, "A Plaid Government will prioritise universal, affordable, and high quality child care, provided by a range of deliverers, for every family in Wales," Plaid Cymru, 2007, p. 38).In contrast, rival Welsh Labour (2007) places greatest emphasis on "the economy/tackling poverty" (27.4%; for example, "High quality childcare for all 2 year olds in our most disadvantaged communities . . .This will assist those parents who are currently economically inactive to find employment and, thereby, to tackle one of the roots of child poverty," p. 39).The latter reflects its strategic focus on ending poverty (p ≤.001). 21 In Northern Ireland, the nationalist parties again predominate; they account for more than two thirds of all references to child care (SDLP and Sinn Fein, each accounting for 37.5% of the regional total). 22For both, the lead frame is "tackling poverty" (40% and 38.2% of each party's references to child care; for example, "the SDLP views employment as a key route out of poverty which is why we must develop affordable and sustainable childcare policies that will allow many economically inactive people to return to the workplace," SDLP, 2007, p. 29).
In the case of statewide parties in Scotland and Wales, the territorialization of child care policy is also driven by intraparty differences in both issue salience and framing (p ≤ .05), in other words, contrasting policy programs advanced by the United Kingdom and "regional" divisions of the same party (e.g., between U.K. Labour, Scottish Labour, and Welsh Labour). 23As Laffin, Shaw, and Taylor (2007) observe, such divergence has emerged from rapid and far-reaching institutional change whereby they "have shifted significantly from being traditional, centralized parties" (p.88).For example, Welsh Labour (2007) pledged, "We will provide targeted support to help lone parents to find and stay in work, including support for cooperative enterprises which provide emergency childcare for women in work" (p.28).Scottish Labour made no such commitment.Instead, it promised, "by 2011 we will ensure that every early years and childcare facility will be led by an Early Years professional" (Scottish Labour Party, 2007, p. 18).In like fashion, the Scottish Conservatives (2003) pledged, "we will encourage all employers to make salary-sacrifice childcare vouchers available to their employees, and we will ensure that the public sector leads the way in this provision" (p.21), a commitment unmatched by their Welsh colleagues who instead stated they would pass primary legislation in Wales whereby "small and medium sized businesses would be rewarded for . . .providing childcare" (Welsh Conservative Party, 2011, p. 36).
In the case of minority nationalist parties (MNPs), child care was integral to the parties' vision of each nation.For example, "now is the time for that new direction-a time to build the kind of Wales we want-universal childcare . .." (Plaid Cymru, 2007, p. 23).child care was also expressed in the context of their ideological goal of independence.For example, "We will be able to address the priorities of people in Scotland, from better state pensions to universal free 10 SAGE Open childcare . . .Independence will allow us to make Scotland a better place to live" (SNP, 2011, p. 37).In Wales, the discourse also emphasized Welsh medium child care (e.g., we will put in place "targets for Welsh-medium educare in every locality," Plaid Cymru, 1999, p. 18).

Discussion
The present findings highlight the discursive underpinnings of the late-20th-century rise in child care policy in the United Kingdom.They show a significant growth in the issue salience of child care in the electoral discourse of the main statewide parties in Westminster elections, thereby revealing it to be part of the rise of valence politics.The data also show that, across parties, child care policy is latterly framed more in terms of social investment than as a gender equality issue.One reading of this is that it augurs well; it signals the reconciliation of the Wollstonecraft dilemma 24 and the ushering-in of a "post-feminist" era in which policy-makers have progressed from the male-breadwinner, hetero-normative assumptions of earlier decades.On this reading, it suggests a policy framework that supports parents and guardians equally, regardless of sex, when seeking to balance child raising and engagement with the labor market.
However, it is argued that the foregoing is eclipsed by another, more worrying interpretation.For although child care issues affect both male and female parents, collectively, women remain the principal carers of young children (Windebanke, 1996).Deep-set and enduring patterns and processes of sex inequality in the labor market (Charles, 2003) mean that parties' failure to frame child care in the context of gender equality can be seen as a key failing.Not least, it is a derogation from the obligations of international agreements such as CEDAW.This was highlighted in the latest UN Convention Progress Report on the United Kingdom (2013), which concluded that contemporary issues facing women include, The continued discrimination against women in the labour market in terms of opportunities and equal pay; the reduction in legal support and access to justice for women suffering discrimination; [and] welfare reforms which are pushing more women into poverty and insecurity.(p.56) Moreover, it singled out the fact that "there was no universal childcare provision," in turn, questioning, "was the Government prepared to increase infrastructure to make affordable childcare better available?"(p.56).
Thus, the current manifesto data provide further empirical backing for Lister's (2006) insightful work on New Labour policy and disquiet related to "the construction of children as 'becomings' rather than 'beings'; the paid-work-focused and future-oriented model of citizenship; and the eclipse of parents ', and in particular, mothers' welfare . . . [and] principles of (gendered) social justice" (p.315).Moreover, the manifesto discourse reveals issues of inter-generational equity and how, as Jensen (2009) cogently observes, contemporary emphasis on social-investment "child-centred policy ideas, [means that] the equality claims of adult women and attention to their needs are side-lined in favour of those of children" (p.446).
A further issue that emerges from the foregoing analysis is that of territorial justice.This is the principle that unitary states do not have sub-divisions of territory that are subject to a different set of social rights.However, as the manifesto discourse on child care illustrates, this principle is being revised in the wake of devolution.This confers a unique status on the discursive process of framing policy proposals in party manifestos.On one hand, it is the origin of differential child care regimes-as regional parties set out programs (and territorial rights) that are contingent on local socio-historical factors and reflective of regional party politics.Yet, on the other, it is also the answer to concerns over territorial (in-) justice, because, in democratic terms at least, it is inherently just-owing to democratic process of elections and mandate seeking.
Allied to the foregoing, existing work explains the resilience of welfare regimes in terms of the "stickiness" of existing policy arrangements and political constituencies associated with welfare institutions and policies (cf.Pierson, 1994).According to this view, earlier rounds of policy choices, such as in relation to child care, become institutionalized and tend to "crowd-out" policy alternatives.Furthermore, over time, welfare institutions and policies generate their own political constituencies composed of groups such as service users, employees, and professionals.These combine with party politics to constrain politicians' policy choices and political options.However, the present analysis suggests that this can be challenged by devolution and the pluralizing of electoral systems.This is because (quasi-)federalism introduces a new spatial policy dynamic whereby regime resilience is also shaped by electoral discourse, issue salience, and policy framing.As the present analysis shows, in multi-level systems, earlier statewide mandated child care policies are revised over successive election cycles as part of a discursive process, as parties compete in placing policy proposals before voters.

Figure 1 .
Figure 1.The issue salience of child care policy in the three main parties' general election manifestos 1983-2010: All-party absolute totals of quasi-sentences in each election (N = 210).

Figure 2 .
Figure 2. The issue salience of child care policy in the three main parties' general election manifestos 1983-2010: Child care policy as a percentage of all quasi-sentences in each election (N = 20,022).

Figure 3 .
Figure 3.The issue salience of child care policy in UK meso-elections 1998-2011: Combined all-party total of quasi-sentences referring to child care in each polity (N = 528).