In Gosta Esping-Andersen's paradigmatic analysis of welfare capitalism (1990), the French welfare state is classified in the conservative corporatist regime. (1) France is generally recognized as belonging to the Bismarckian family of welfare states because of the centrality of social insurance. In fact, most of benefits are earnings related and entitlement is conditional upon a contribution record. Social security contributions are paid by both employers and employees.
However, in France, there is also a non-contributory element, a Beveridgean component. It refers to the tax-financed system and includes social assistance programs, minimum incomes and programs aimed at vocational training, and at integrating the young or long-term unemployed. These programs benefit those who do not have access to the social insurance system, such as the long-term unemployed or the young unemployed who have not contributed enough or have only a very short contribution period. Since the late 1970s, several kinds of minimum incomes or social minima have been set up to compensate for the failures of the Bismarckian sector (the social insurance system). There are at least eight different minimum incomes: minimum income for old people over sixty-five years, allowance for disabled adults, integration allowance, specific solidarity allowance, lone parent allowance, minimum income for disabled people (for those who have small invalidity insurance benefits), widowhood allowance, and minimum guaranteed income for adults over twenty-five years. These minimum incomes are granted with means tests.
In the last three decades, the Beveridgean component has grown in a significant way. The economic crisis and the economic changes (more flexibility in the labor market and job insecurity) have brought about an increase in the number of people not covered by the social insurance system.
The family benefits scheme is one of the most important elements of the French social protection system and exemplifies the fact that the Beveridgean and Bismarckian logics are deeply interlinked. It is traditionally acknowledged that French Family policy has three main aims: to increase the birth rate, to help families financially so as to offset the cost of raising children (horizontal redistribution), and to provide priority assistance to poor families in order to offset social inequalities (vertical redistribution). In recent years, another goal has been assigned to French family policy: to promote equal opportunities for men and women by measures designed to reconcile family obligations arising from the care and upbringing of children and parents' employment, education, and training. Historically, family benefits were first introduced on some employers' initiative as a social insurance scheme and were financed by the latter. They were granted to workers with children and were part of the wage system. It was both a question of keeping the labor force and favoring the traditional Family model. This initiative was gradually generalized and became compulsory from 1932 onward, eventually encompassing the whole of the population in 1978.
Until the end of 1990, employers and independent workers exclusively financed the family benefits scheme, but their cost is gradually being included into the state budget. Family benefits are now financed partly through taxation. Consequently, the family benefits scheme is a mixture of Bismarckian and Beveridgean characteristics.
Nowadays, there are six types of benefits delivered by the Family Allowances Funds (the CAFs): benefits associated with birth and infancy, such as allowance for young children or parental education allowance; income maintenance benefits, such as family allowances granted for second and subsequent children; targeted benefits such as allowance for school children; lone parent allowance; housing benefits; and childcare support or allowance for mother's help or nannies. (2) Most of these allowances are means-tested, except for parental education allowance, which is contributory (see appendix). But the targeting of these income-related benefits does not necessarily lead to reserving them for the poorest families. The ceiling of incomes retained to allocate benefits can be rather high, so average-income families are also entitled to social welfare provision. Over the long period from 1973 to 2001, the means-tested benefits have increased more than the benefits without the means test. But it is not the case for the recent period (1995-2001), where their share has remained stable.
Concerning tax credits, French legislation does not include specific tax credits associated with family policy. However, income tax takes into account the family situation and the number of persons considered as fiscally dependent (spouse, children, etc.) through the "family quotient" system used to calculate tax allowances. It is the outcome of dividing the incomes of the nuclear family by a given number of shares, which correspond to its size and financial position, taking into account the various benefits that the family is entitled to. (3) Tax reductions are generally granted for the childcare cost paid by families for children under the age of seven years.
It is noteworthy that France is taken as an example of "modified male-breadwinner" countries, as opposed to the "strong male-breadwinner model" in Britain, inasmuch as "state policy has recognized the reality of women's roles as both mothers and paid workers" (Lewis 1992, 167-8). Furthermore, other authors have underlined the high labor-force participation rates among lone parents in France (Bradshaw et al. 1996). It is related to the high level of public childcare provision, to the quality of nursery and elementary schools, and to the priority access given to lone parents. It is also an effective strategy to fight against child poverty by developing social and collective services. According to Barbara Bergmann (1996), it shows that France has chosen to invest in the human capital represented by children in contrast to the United States. The "dual welfare state" and gender bias in U.S. income support policy are responsible for the high rate of childhood poverty in comparison with other Western nations. (4)
The main features of the French system being set out, the purpose of the present text is to highlight the principles of justice that explicitly or implicitly underlie France's family-oriented redistribution efforts. The field of study is defined by the family benefits and direct income tax systems. More specifically, we will be looking at non-contributory cash benefits, thus excluding benefits in kind, such as equipment and services of use to families. We will also focus on how household taxation varies for different types of family configuration. Income tax, precisely because of mechanisms such as the "family quotient" and "family-oriented" tax cuts, constitutes a redistributive instrument that impacts family policies. What are the dynamics underlying the criteria of social justice that have been at work in the French family aid and transfer system that has been in place since 1945?
To deal with the issue of the normative principles that are embodied in different institutional arrangements that make up the French welfare state, one can study variations in welfare state provisions through the lens of typologies. Our method is quite different and more qualitative. We start with some theories of justice drawn up by the "post-welfarist" authors (Richard Arneson, Gerald Cohen, Ronald Dworkin, Amartya Sen, for example). This choice needs to be justified. In fact, such theories deal, in a framework that is distinct from utilitarianism and welfarism, (5) with the difficult issue of knowing how to allocate rare resources fairly among members of a population whose initial incomes, physical capacities, talents, and preferences vary. More specifically, post-welfarist authors focus their analysis on the fair compensation of certain types of inequalities, even as they seek to avoid rewarding a lack of effort or a predilection for expensive tastes. They offer a theoretical framework that opposes compensation to individual responsibility. What should State authorities compensate, and, inversely, what should they not compensate? The general principle of post-welfarist theories is to distinguish the individual characteristics that lie outside individual control from those that involve personal responsibility. As such, a fair distribution will compensate for the phenomena for which individuals are not responsible and will not compensate for those for which they are responsible. Furthermore, studies in experimental economics show that the moral intuitions which ordinary individuals follow when facing concrete problems of redistribution are relatively similar to the lessons that can be drawn from the post-welfarist theories of justice (see Schokkaert 1999). Moreover, such theories are useful at two different levels: they exemplify the interrelationship between economic and politico-philosophical elements, and they highlight the tension that can exist between principles of freedom and principles of equality. In addition, they raise the crucial issue of determining what type of equalization should take place. In short, they seem to be attractive and relevant ways of ordering thinking about the criteria of justice underlying the French system of direct and indirect family-oriented transfers, especially since the preconditions for awarding some of these benefits are affected by certain individual characteristics and unaffected by others.
For all reasons mentioned above, we utilize the post-welfarist literature to assess the French benefits system. It is true that communitarian framework also seems to be able to account for some of the peculiarities of the French system, especially the beginning of the French family policies. But we need a unified ethical framework to evaluate the French family transfers...