For the last few decades, social policy practitioners and economists have waged a continual, open debate about the amount of responsibility a state can and should have with regard to welfare provision. This debate was largely provoked by the global economic crisis of the late 1970s, which challenged the idea of an active welfare state. High levels of government involvement in social welfare provision were tied to Keynesian support for full employment, which was accepted (more or less) throughout most of the industrial West between 1950 and 1975. After this, however, rampant inflation, unemployment, and slow growth challenged the viability and indeed the utility of government-sponsored social welfare. The policy responses of different states were determined not only by their particular political and economic circumstances, but also by their historical, cultural and ideological backgrounds. Some countries opted for social welfare maintenance while others chose some type of retrenchment (Mishra 1990, 2).
Although social welfare is a commonly discussed topic, it can have many different meanings depending on the focus of analysis. In general, though, welfare refers to socioeconomic wellbeing. This is correlated with 'the basic level of economic development, of course, but focuses more specifically on a variety of goods and services believed to be essential for individual and social happiness and security (such as health care, housing, social insurance, other employment-related benefits, and additional forms of social assistance). Together, the institutions and policies supporting these types of goods and services form a welfare state regime. The welfare state defends and supports the development of social rights, aiming "to make civil rights actually work ... removing the barriers that blocked the full and equal exercise of civil and political rights" (Bussemaker and van Kersbergen 1999, 10).
Standard welfare state analysis classifies advanced Western countries according to regime type. Past analyses have tended to treat the Southern European countries of Spain, Portugal, Italy, and Greece in one of two ways: either bundling them together with continental European nations such as France and Germany, or excluding them altogether. However, the countries of Southern Europe exhibit very different characteristics from those in continental Europe (Del Boca, Pasqua and Pronzato 2003; Rydell 2003; Flaquer 2000). In contrast to continental Europe, Southern European nations witness a combination of low fertility levels and low levels of female employment; the labor market structures are more rigid and unemployment is higher; part-time work options are limited (for both men and women); the welfare system is heavily based on family networks rather than state provision; and it remains very difficult for women to combine market work and family responsibilities. These considerable differences suggest that standard welfare regime typologies cannot accurately portray welfare systems in Southern Europe, especially as they relate to women's employment. An important reason for this is that standard typologies do not sufficiently incorporate a gender component of analysis. In order to explain why Spanish female labor force participation is so low, it is necessary to take a multidimensional view of decommodification (the degree to which individuals, or families, can uphold a socially acceptable standard of living independently of market participation). In so doing, it becomes clear that the eligibility requirements for various social benefits are tied to women's roles as wife, mother or worker. The Spanish case study reveals that social policies in Spain tend to privilege the wife and mother roles for women--not the worker role--and this can help to explain the comparatively low labor force participation rate for women in that country. As such, this case study provides an interesting and important contribution to the literature of welfare state analysis.
Labor Force Participation Across Europe
As Table 1 shows, Southern European labor force participation rates for women are consistently lower than those in continental and Northern Europe. At the same time, male labor force participation rates in Southern Europe generally exceed (or at least, approximate) corresponding rates in other European regions. In 2006, Spain had one of the highest male labor force participation rates in Europe (80.7%). These patterns of difference in labor force participation cannot be explained by traditional regime typology.
In order to understand the Spanish difference, it is important to analyze the way that social, family and tax policies affect labor force participation in the country--particularly for women. Welfare state analyses examine how market, state and family interrelate to provide social welfare. However, it is not sufficient to examine these institutions "without explicitly theorizing the gender structure that in reality underlies and determines the relationships between [them]" (Bussemaker and van Kersbergen 1999, 20). Given that there are specific eligibility requirements for each type of social assistance, it is important to ask how women gain access to the benefits and privileges afforded by the welfare state. More specifically, how are the social rights of women attached to their role(s) as wives, mothers, workers and citizens? Although these roles all have links to the market, state and family, we can hypothesize that the roles of wife and mother are most firmly attached to the institution of the family, while the role of worker is more firmly attached to the market, and the role of citizen, to the state. The issue, then, is how the market-state-family configuration in a welfare regime may support a particular set of gender roles and family structure, which may either encourage or derail gender equity. In particular, welfare eligibility requirements support and reinforce certain gender roles--affecting the distribution of resources among men and women, influencing the economic dependence of women on men, and shaping women's labor force participation decisions. Accordingly, this Spanish case study analyzes several social, family, and tax policies--parental leave, child benefits, childcare provision, and income tax structure--showing how they can support particular roles for women: wives, mothers, or workers. This adds a new dimension to welfare regime analyses, both expanding knowledge of the "Spanish difference" as well as highlighting the importance of gender for labor market outcomes.
Background: Welfare Regime Analysis
There are many different levels and forms of welfare provision around the world, which has long stimulated comparative research in sociology (and, more recently, in economics and political science). Nonetheless, until Gosta Esping-Andersen's seminal book, The Three Worlds of Welfare Capitalism (1990), the progressivity and/or similarity of different welfare regimes were commonly determined by a simple expenditure measure (percentage of GDP allocated to social assistance programs). To Esping-Andersen, this was a faulty indicator. (1) Countries with similar levels of welfare expenditure often had very different emphases for social assistance and, as a result, they experienced different socioeconomic stratification patterns. Esping-Andersen embraced a three-tiered approach to welfare regime analysis: (1) exploring the variations in social rights resulting from the type, quality, and quantity of welfare expenditures; (2) examining the ensuing class-based divisions; and (3) gauging how the state, market and family interrelate in the provision of welfare. In so doing, Esping-Andersen devised a more solid, multifaceted foundation for comparative analysis.
Although Esping-Andersen's methodology represented a great leap forward in welfare regime scholarship, his approach has often been criticized from a gender-based perspective (see, for example, Orloff 1993; 1996; Sainsbury 1996; Korpi 2000). Even though he mentions the role of the family in welfare provision, his theory is strongly biased toward the dichotomy of public (states) versus private (markets) (Bussemaker and van Kersbergen 1999, 13). This has two important effects. First, class-based divisions are accentuated while other types of societal divisions--including gender- or ethnic-based divisions--are deemphasized. Second, it is unclear how families (and especially the unpaid work of women in families) fit into the analysis. This is strange, given that the family is the most significant provider of welfare and care in all regime types (Daly 1994, 105).
It is common knowledge that family structure and intra-family relationships are important determinants of women's employment patterns. However, the Esping-Andersen regime typology does not adequately explain how welfare policies affect different types of families (and the relationships within those families). This means that even though Esping-Andersen's regime types can be linked to different levels of women's labor force participation, a crucial piece of the puzzle is missing. We do not have a solid theory that links family structure, social assistance, and women's employment patterns.
Above all, it is crucial to understand why a focus on labor force participation is so important. Not all women want to enter the marketplace. There are still many women who would rather stay home and raise their children as their full-time job (or pursue some other alternative), so homogeneity does not exist among all women. (2) Nonetheless, since most poor people are women, employment opportunities for women are critical for food and housing security (Grown, Gupta and Kes 2005). A great deal of empirical evidence demonstrates that increasing the share of women's income in the household improves family and social welfare, given women's likelihood to spend more time than men on their children (for examples, see Collier 1990...