Paper prepared for the First International Graduate Student Retreat for Comparative Research, UCLA, May 8-9, 1999.

¹ This is a slightly shortened version of my dissertation prospectus.

Introduction

Many welfare states that have historically supported a male breadwinner and female homemaker family model are reinterpreting the needs of single mothers as they are moving to increase women’s attachment to the labor market . Among male breadwinner/female homemaker welfare states, the Netherlands and the United States have made the most dramatic shift away from supporting stay-at-home mothering. In 1996, each state changed the goal of social assistance programs from providing income support to channeling single mothers into paid employment. Under the new programs, the Netherlands and the United States become intimately involved in directing the behavior of this particular group of citizens as they require women on social assistance to work or to participate in welfare-to-work programs. These changes in policy imply a new understanding of single mothers’ needs, away from an emphasis on enabling mothers’ direct child care, towards an emphasis on mothers’ material provision for their children. Needs interpretations, or the process by which needs are constructed by identifying and acting upon them, are an important tool in these attempts to change the way citizens live .

Policy intent, captured by social policy legislation, gives an indication of the needs the state is willing to recognize. Furthermore, the needs thus recognized form the bases of women’s citizenship rights. However, I argue that to understand how legislation constructs the needs of single mothers, we have to look beyond the intent of policy makers to the interactions between street-level policy implementers and the recipients of social assistance. It is at the level of interaction between representatives of the state and citizens that policy, and in some sense the state itself, ultimately takes shape. Arguing that processes of needs interpretation are influenced by differences in welfare state regimes and in bureaucratic implementation practices, the central question in this dissertation is how these regimes and practices interact with women’s own definitions of their needs to create a new form of female citizen. I will answer this question through observations of client-caseworker interactions and open-ended interviews with both clients and caseworkers in the Netherlands and the United States.

Choosing the Cases: The Contexts of Policy Change

The central issue in these restructurings of the welfare state is whether single mothers are better served by direct state support or whether their needs are best met through the market. In order to understand the impact of the shift towards the latter solution, we have to look at practices and ideologies surrounding motherhood, work, and care in states that are adopting these new approaches. The United States and the Netherlands are two of six countries which have historically provided single mothers assistance to enable stay-at-home mothering (See end note 1). Ireland, Britain, Australia, and Luxembourg are the other four (see endnote 2). This assistance is combined with very limited support for working mothers (i.e. absence of social services such as child care, tax policy, or family allowances that enable or encourage women to work outside the home). In all these countries, with the exception of the United States, this has lead to very low labor force participation rates for women in general and single mothers in particular. In contrast, in the United States, extremely low benefit levels combined with high levels of stigmatization resulted in very high rates of labor force participation among single mothers.

With the 1996 passage of PRWORA, the United States abandoned its already eroded commitment to supporting stay-at-home mothering by making work search activities mandatory and by removing the entitlement to income support for single mothers. The passage of the Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA) placed a lifetime cap of five years on welfare benefits for single parents and their children. PRWORA rescinded the entitlement provisions of Aid to Families with Dependent Children (AFDC) and turned AFDC into Temporary Aid to Needy Families (TANF). In that same year, the more extensive Dutch welfare state increased its emphasis on work over mothering by making work search activities a prerequisite for benefit receipt. Until recently, Dutch single mothers were exempt from the general requirement to look for work, but this exemption was revoked as of January 1, 1996 for women with children over the age of five. If women do not participate in work search activities their benefits will be reduced. In both countries, the goals of welfare programs have shifted from offering direct financial support to directing women to the labor market for their material needs. The other welfare states that had enabled stay-at-home mothering, Britain, Ireland, and Australia, have also chosen to try and increase lone mothers labor force participation . However, they do so through positive incentives rather than mandates. The Netherlands and the United States, then, are most closely aligned among welfare states that support male breadwinner/female homemaker family models in the similar ways in which they are reinterpreting the needs associated with motherhood, work, and care. Their similar policy changes attempt to alter women’s mothering practices as they designate work as the valued activity for single mothers and associate care with material provision rather than with face-to-face interaction between mothers and their children.

Despite these similarities, the policy changes in each country had different motivations, motivations which will impact how these policies are implemented and negotiated by women on welfare. In the Netherlands, the change came about as the Dutch "caring state," the label the Dutch give to their welfare state, defined the problems of poor citizens as social exclusion and lack of social participation, and attempted to remedy these problems by increasing the labor force participation of social assistance recipients . In the United States, shifts in social assistance policy resulted largely from concerns with the morality of supporting single, particularly never married, mothers. The passage of the Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA) came after a long public debate on the perceived dangers single motherhood posed to American society . In addition, married mothers’ high labor force participation rates made political support for public assistance programs that enable stay-at-home mothering untenable, an indication that needs reinterpretation in the United States might mean an attempt to reduce the needs recognized by the state. In sum, even as the approach to single mothers in both countries is more and more shaped by an emphasis on responsibilities rather than rights, the needs associated with fulfilling these responsibilities are much less stigmatized in the Netherlands than in the United States.

Differences in mothering and work practices will also impact the process of needs interpretation. In the Netherlands, women’s low labor force participation rates mean that the works requirements of the new social assistance act put expectations on single mothers on welfare not experienced by mothers in general. In the United States, PRWORA can be seen as an attempt to bring single mothers on welfare in line with the new norm of the working mother. Dutch labor force participation is particularly low among single mothers, only sixteen percent of whom work full-time and 54% of whom receive some form of social assistance (See end note 3). The Dutch policy change then alters the lives of the majority of single mothers in the Netherlands as it goes against an evident preference for stay-at-home mothering and part-time work. In the United States, on the other hand, divorced mothers are even more likely to work than married mothers, while never married mothers’ labor force participation rates are high as well (See end note 4). Even women on welfare have often worked informally in order to make ends meet . This implies that in the United States, the transformation of single mothers on welfare into workers becomes an issue of balancing the requirements of home and work rather than one of exchanging home for work as it is in the Netherlands.

These differences in motivations for similar policy changes and differences in women’s expectations about their labor force participation might affect the extent to which they collaborate with the state’s new interpretation of their needs in various ways. Do women agree that they need to find (more) work? Do they articulate a desire to be with their children? Do they articulate a right to obtain an education that will enable them to find good jobs? Policy motivation and labor force participation rates will also play into women’s sense of self as mothers, workers, and caregivers. Particularly, differences in stigmatization — captured by the images of the helpless victim of circumstances or the immoral welfare queen — might affect how women see their mothering. Looking at direct implementation, in both countries, caseworkers are confronted with a dramatic shift in their job description. How do they use the discretion inherent in their position to accommodate some of the concerns single mothers might voice? The Dutch state’s self-image as a "caring state," sensitive to citizens’ needs, opens the possibility for clients’ resistance and policy alteration as a result of this resistance. The United States, on the other hand, has always had a more limited approach to care. It might well be that caseworkers see the goal of PRWORA simply as reducing the welfare rolls, as happened with the Work Pays Demonstration Project that preceded TANF in California, and not as a tool to alter the behavior of citizens .

Welfare State Theories, the Interpretation of Needs, and the Construction of Citizenship

General Theories of the Welfare State

My conceptual model of the welfare state follows some aspects of Esping-Andersen’s comparative analysis . Esping-Andersen’s framework centers around three dimensions of the welfare state: decommodification, stratification, and state-market relationships. Decommodification refers to programs which lessen citizens’ absolute dependence on the labor market. Class is the most important dimension of stratification identified by Esping-Andersen with gender and race playing a lesser role in his analysis. State-market relations capture the ways in which state welfare policies do and do not intervene in the market. Esping-Andersen argues that welfare states cluster around these three dimensions into three different regime types: liberal, corporatist, or social-democratic. These regimes came into being because of the absence or presence of (working) class mobilization, Catholicism, or a strong belief in individualism and/or equality. The United States is the prime example of a liberal welfare state regime, as it offers a limited safety net and relies heavily on market fulfillment of needs. The Dutch welfare state offers an example of a corporatist regime as it maintains citizens’ status and aims to uphold traditional family constructs through social insurance and assistance programs. At the same time, the Dutch welfare state also has some aspects of social democratic regimes. This regime type is marked by full employment policies, active pursuit of income equality, and extensive social services. Current policy changes, then, move the Dutch state away from its corporatist roots while the American approach is in sync with liberal regimes’ limited safety net provisions.

Various feminist scholars have pointed out that Esping-Andersen does not explain the welfare state’s gendered dimensions even when he mentions gender in his analysis. These scholars have shown that differences in welfare state regimes affect the division of labor within the household and among the household, the public sphere, and the market. Welfare state regimes also direct the gendered stratification of the labor market and they reinforce dominant gender ideologies . These authors argue that many of the insurance based policies of the welfare state, such as old age pensions, unemployment benefits, and disability benefits have historically been based on the assumption that men are primarily responsible for family income. The differing impacts of these foundational assumptions on various welfare states have led Jane Lewis (1992) to classify welfare states as gender regimes (see also Millar 1996; Williams 1995). Lewis argues that welfare states can be categorized as weak, moderate, or strong breadwinner states as they vary in the extent to which they do or do not support male family headship and female employment through social insurance, family allowance, full employment or social service policies. By articulating gendered dimensions on which welfare states vary, these scholars show that welfare states can alter the meanings and practices associated with motherhood, work, and care. However, how these meanings and practices are internalized and verbalized by caseworkers and clients, or whether clients and/or their caseworkers resist these messages cannot be explained within the current macro framework of these comparative analyses.

The policy implementation literature does focus on the micro level. This literature shows that street level implementers transform the policies they execute and that caseworkers’ discretion as well as the depth of caseworkers’ understanding of policy intent are important factors in determining policy success . For example, observations of caseworkers implementing the Job Opportunities and Basic Skills training program (JOBS), the largely unsuccessful welfare-to-work program that preceded TANF, showed how caseworkers used their discretion to meet caseload quotas set by the program without paying attention to clients needs or abilities (Brodkin 1997). Brodkin argues that while discretion in principle can have either negative or positive consequences, in this case, caseworker discretion had a negative impact on program success. However, I use the dimensions of caseworker discretion and knowledge, not to ascertain policy success or implementation success, but as a means to understand how a gendered citizenship emerges from interactions between clients and caseworkers, a citizenship that might differ from that intended by policy makers. In this framework, the caseworkers and clients are both gendered subjects. In addition, including the clients’ perspective and experience in the analysis shifts my research outside the frame of the literature on implementation in which clients mainly appear as passive objects.

As in the literature on policy implementation, my approach to the welfare state focuses on the arena in which states and citizens engage in direct interaction. The idea of gendered welfare state regimes captures differences in the structuring aspects of the state while implementation looks at differences in the practices of the state. Combining these two sides of the welfare state, I argue that while the state is a structuring force in citizens lives, the content of this structuration can only be understood through an examination of the practices by which these contents are transmitted. In addition, not all citizens are equal and single mothers on welfare inhabit a marginalized social space. Their experience of poverty interacts with experiences of race and gender to shape the interaction between them and the agents of the state who implement poverty policy. Seeing caseworkers as agents of multifaceted states and the interaction between caseworkers and clients as the site in which citizenship is constructed, I move from a focus on the state solely as structure to a focus on the state as both structure and practice.

Gender, Needs Articulation, and the Construction of Citizenship

A focus on caseworker-client interactions, takes as its point of departure feminist theories in which the welfare state is a cluster of at times contradictory arenas of identity formation, needs interpretation, and interest production . In this approach, women are not necessarily passive objects of state policy, as in earlier feminist theories, but have varied and complex relationships with the state. At the same time, the state itself is not a uniform entity and agents of the state such as social assistance caseworkers can and do develop practices not intended by policy makers. In this framework, needs are the currency of welfare state policies; once a need is identified as social, the state represented by its agents can choose to act on that need (Fraser 1989a, 1989b).

Looking at the United States welfare state, Fraser captures the state’s multifaceted nature by labeling it the "juridical-administrative-therapeutic state apparatus" or JAT (1989b pp. 153-4). Moving beyond the idea that welfare states simply satisfy pre-existing needs, Fraser argues that through the JAT the welfare state constructs and depoliticizes gendered identities by identifying and articulating gendered interpretations of citizens’ needs. Fraser argues that welfare agencies, as the administrative side of the JAT, privilege certain needs by forcing recipients’ varied problems into the agencies’ preexisting categories (Fraser 1989b pp. 154-5). For Fraser, the needs ultimately recognized by the state are articulated in a struggle between marginalized groups, the state, and other dominant but non-state actors (Fraser 1989a). In this reading, the welfare state controls citizens’ lives in multiple ways but citizens do not act against the state as individuals. I argue, on the other hand, that needs construction is not a one way process but takes place in the interaction between citizens and state representatives. A comparative analysis of two different welfare state regimes’ practices can show the impact of state level structural differences on these micro processes of needs construction.

The issue of needs also points to the social rights dimension of citizenship which covers the right to "economic welfare and security" (O’Connor 1993 p. 503). For Esping-Andersen, economic welfare and security depend, in part, on the level of decommodification or the extent to which states guarantee citizens ability to not be (fully) dependent on the labor market for their subsistence. However, this conceptualization of social rights has been critiqued by feminists who have argued that decommodification implies a preceding commodification, hence that these social rights are based on having (had) ties to the labor market (Langan and Ostner 1991; O’Connor 1993; Orloff 1993). In welfare states that have supported a male breadwinner/female homemaker model, women’s ties to the labor market have historically been weak and women have not built up individual rights to social security. Indeed, Ann Orloff (1993) has argued that access to paid work should be a dimension of citizenship rights. However, once we step outside the decommodification paradigm, it is also clear that these new reforms decrease the number of avenues open to women that guarantee the economic security necessary to "form and maintain and autonomous household," another social rights dimension identified by Orloff (1993, p. 303).

In general, then, citizenship is a gendered construct, something ignored in much of the general literature on citizenship in which citizenship is a universal category granting universal rights . In addition, as Fiona Williams points out, citizenship is not simply based on single social categories but arises out of a complex interlocking between race, gender, and class formations and nation-state development . Yet, in this area, again, too little attention is paid to how this citizenship is given meaning in the daily lives of citizens. Client-caseworker interactions are one context in which citizens learn the content of their citizenship. In addition, the citizenship learned in this context is not necessary the formal citizenship intended by the laws of the land. As citizens make claims against the state in the welfare office, they are confronted with the rights and responsibilities that caseworkers, as agents of the state, attach to these welfare programs. In addition, in interaction with caseworkers, women might articulate rights not (yet) formally recognized by the polity. The process of needs articulation that takes place in welfare offices in each country, can thus give insight into the actualization of single mothers’ citizenship rights.

In sum, a focus on state practices centered on Fraser’s ideas of needs articulation enables me to connect the gendered ideologies reflected in welfare state policies and the individual experiences of welfare recipients. In addition, it allows me to problematize the construction of gendered citizenships. The institutional, macro level focus of the power resource school represented by Esping-Andersen and feminist comparative welfare state scholars does not enable an understanding of the mutual meaning construction that takes place in the interaction between caseworker and client as the state attempts to instigate behavior change on the part of single mothers. Looking at caseworker client interactions, I will show how differences in welfare state regimes and bureaucratic practices within which these attempts at behavior change occur affect gendered understandings of the needs associated with mothering, work, and care, understandings which ultimately determine the content of citizenship.

Operationalizing Needs Construction

I suggest that knowledge, discretion, and ideology are the three dimensions of client-caseworker interaction that reflect the different bureaucratic practices and welfare state regimes within which each state articulates the needs of single mothers. Knowledge refers to caseworkers and clients’ understanding of policy goals, including the gendered dimensions of these goals. The dimension of knowledge thus captures caseworkers’ translation of legislative intent and the details of implementation described in the law. In order to understand caseworkers’ knowledge, we must also understand both their formal and informal training. For clients, knowledge includes knowing the rules and regulations as well as informal knowledge about the welfare system that might be passed on from client to client. Discretion refers both to the official discretion embedded in policy and to the discretion caseworkers have to circumvent policy intent. For example, there is room within the Dutch social assistance policy for caseworkers to classify women with children as unemployable and to exempt them from work requirements. However, while one caseworker might feel that women ought to mother and eagerly classifies women as unemployable, another might firmly believe that work is much better than welfare and rarely use the exemption. Clients’ discretion refers to what they do and do not reveal about themselves to their caseworkers. Finally, ideologies are experienced by clients and caseworkers as personal beliefs but they reflect larger society and culture, including meanings produced by government practices, which shaped these beliefs. Ideologies regarding family, work, and care are what ultimately filter caseworkers’ understanding of policy intent as well as the direction in which they will use discretion. These ideologies will also guide how women approach their caseworkers and how they will work the system.

The two cases will show two of the four possible ways in which the dimensions of discretion and knowledge on the side of caseworkers are filtered through caseworkers’ individually held ideologies. Caseworkers in the United States have historically had relatively low discretion and low knowledge (Brodkin 1997; Meyers, Glaser and Mac Donald 1998), while Dutch caseworkers have relatively high knowledge and high discretion . The low discretion and knowledge reflect the limited extent of the United States’ liberal welfare state while the high levels of training and the independence of Dutch caseworkers show the greater value attached to social assistance in social democratic and, to a lesser extent, corporatist regime types. In the United States, low discretion and knowledge might lead to indifferent policy implementation and confusion among clients . In addition, low knowledge and low discretion might not allow for personal ideologies to have a big impact on the interaction between caseworker and client simply because that interaction might be very limited. In contrast, in the Netherlands, high knowledge might lead to a more thorough understanding of the guidelines, including the loopholes, that allow the caseworker to put his or her own stamp on the interaction. When combined with high levels of discretion, high knowledge might result in highly unequal implementation of the policy because it leaves room for the personal ideals of the caseworker to affect the interaction with his or her clients. On the other hand, high discretion and high knowledge might also lead to greater consensus among caseworkers and even among caseworkers and clients exactly because it allows more room for caseworkers and clients to internalize the ideas represented by these policies.

The dimensions of knowledge, discretion, and ideology then guide the way the state, through caseworker-client interaction, constructs gendered understandings of the needs associated with work, family, and care. In the process of policy construction, legislatures and high level bureaucrats cannot fully control the ways in which caseworkers ultimately implement policy. In addition, caseworkers’ discretion, knowledge, and ideologies are not the only factors that determine which needs are recognized in caseworker-client interaction. Finally, the client’s construction of her needs, her knowledge of the system, and her ideologies and practices form the parameters within which caseworkers attempt to change clients’ behavior and clients’ perceptions of their needs.

The Research

In order to understand how the three research dimensions I have identified interact, I will observe meetings between clients and caseworkers and conduct interviews with both caseworkers and clients. During my stay in the Netherlands, I will observe client-caseworker interactions at welfare offices in two different sites: a large central city and a (smaller) city on the periphery. Under the new social assistance act, municipalities have to develop individual plans for implementation. I chose these two sites because offices in larger cities have more extensive bureaucracies with potentially higher levels of specialization and greater resources for both policy development and the education of caseworkers than smaller cities. Larger cities also tend to have more ethnic and racial diversity and higher concentrations of poverty. I will spend approximately three months in each site. My observations will predominantly focus on caseworker-client interactions but I will also spend time with caseworkers on their breaks, and attend meetings and training sessions to gain a fuller understanding of what motivates caseworkers’ actions. During these times I will conduct informal interviews with the caseworkers I observe. In these interviews, I will ask caseworkers about their experiences with the new policies and their views on the possibilities of instigating changes in clients’ behavior through positive and negative incentives. In addition, I will ask them about their views on the ideal policy for this category of citizen.

I will also conduct semi-structured interviews with a total of fifteen clients in each municipality. Interviews with clients will focus on their experiences with the new policies, including experiences with caseworkers. I will ask clients for their life histories in order to understand how they came to use social assistance and to understand the meanings they attach to mothering, work, and care. In terms of the clients’ interview sample, the Dutch welfare population, like the one in the United States, is mixed based on racial, ethnic, and socio-economic background. Approximately twenty-five percent of single mothers who use social assistance are of foreign descent (the Dutch government classifies people by descent rather than race). To reflect this diversity, I will interview women of Dutch as well as of Caribbean and Surinamese descent. These categories of women likely hold different cultural ideas about work, mothering, and care. As in the United States, black and white women, on average, differ in their likelihood of being a single mother, in levels of educational attainment, and in kinship structures.

In the United States, each individual state can give its own interpretation to PRWORA. I will focus on the state of California, which implemented its version of PRWORA, CalWORKS, in January 1998 being the last state in the nation to do so. Why California? First, while its hard to identify an "average" state when it comes to PRWORA, California seems relatively middle of the road compared to states that put great emphasis on cutting down the welfare rolls or compared to states that have created extensive programs to help get women into the labor force such as Oregon. Secondly, California is unusual in that it is one of four states that will continue to provide a grant to the children of unemployed single mothers even after the five year federal time limit is up. (New York, Rhode Island, and Maryland are the other three.) This means that California more closely resembles the Netherlands where there is a constitutional right to a minimum income which enables me to hold one important variable constant in the comparison. Thirdly, the timing and extent of the impact of reform in California makes it a good choice. California differs from states like Wisconsin in its limited experimentation with welfare-to-work programs preceding the 1996 federal Act. This means that in California PRWORA entails a more profound shift than in other states in which many of the aspects of PRWORA were in place prior to 1996. In addition, perhaps because California was late passing CalWORKS, the state differs from other states in that its reduction of the welfare rolls lags significantly behind the national average. As reduction of the welfare rolls is one of the goals of PRWORA, California will have to play catch up which again will heighten the magnitude of the policy change. In sum, California seems a good site for this research project because the shift from AFDC to TANF is likely to be more pronounced than in other states because of the late passage of a state TANF plan and the limited experimentation preceding TANF.

I will do my research in two sites similar to the ones in the Netherlands. One county will be in a more affluent, suburban area, while the second county will be urban with a largely poor population. Again, I will interview the caseworkers in whose offices I do observations and I will conduct semi-structured interviews with fifteen clients in each site. With respect to client interviews, I will match clients in the Netherlands and the United States, making sure that the samples are similar with respect to age, marital status, race, education, and work experience. As teenagers constitute only a very small percentage of Dutch single mothers on welfare, I will exclude them from my sample in both countries.

I will start the California part of the research in the Spring 1999. I will conduct the Dutch side of this study between September 1999 and May 2000. During that period I will be affiliated with the Faculty of Social Sciences, University of Utrecht, the Netherlands, under sponsorship of Professor Knijn. I will write my dissertation upon my return to the United States, graduating in May 2001.

Preliminary Research and Access to the Field

A pre-dissertation fellowship from the Center for German and European Studies at the University of California, Berkeley enabled me to spend the summer of 1998 in the Netherlands. While there, I reviewed the current policy and its implementation. This work allowed me to see that the policy change that affects single mothers not only brings the issue of the state’s involvement in the gendered division of work and care to the center of the debate but that the whole construction of the state as a caring agent is being redefined. While the state is no longer committed to enabling single mothers to be full-time parents, the state has not withdrawn from actively structuring the lives of citizens. Instead, the Dutch state is redefining what it means to care for its citizens and care now becomes making sure citizens have ties to the labor market. At the same time, the Dutch state is undermining older notions of motherhood which rest on the ideal of full-time private care provision. Preliminary reports show that caseworkers struggle to convince both themselves and single mothers of the merits of the shift from at-home-mothering to work (Knijn 1998).

During this time, I also made certain that I would have access to the field. In the Spring of 1998, Professor Trudie Knijn conducted a survey of 1,000 single mothers on welfare in five Dutch municipalities. Of these 1,000 women, 700 agreed to be contacted for further research. Professor Knijn has offered me access to this database and I will recruit women who live in the municipalities I study from this list, contacting them by mail and following this up with a phone call. I will also use Professor Knijn’s contacts at various municipal social service agencies to gain access to caseworkers. In short, I should not have any trouble locating women on welfare or caseworkers for this study. In the United States, I will use various faculty and community contacts at welfare agencies in the two counties in which I will do observations. I will recruit single mothers on welfare through these contacts as well.

Expected Findings and the Significance of the Research

The process of welfare reform raises fundamental issues about work, mothering, and the importance of care. Whereas previous policy allowed these issues to stay in the background by apparently enabling single mothers to stay home, current policy makes the very real tension between work, family, and care central to states’ effort to reinterpret clients’ needs. The question is how caseworkers and clients deal with this tension. Particularly in the Netherlands, I expect resistance to the policy shift, from caseworkers as well as clients. Here, I expect caseworkers to struggle with having to adopt a new view of their clients that might conflict with their own beliefs about family, mothering, care, work, and the role of the state in constructing these. Furthermore, caseworkers might struggle with the transition from providing income support and general counseling to altering clients’ behavior and increasing their ties to the labor market. In the United States, TANF recipients are more likely to accept the idea that they are better off as workers and I expect less resistance to policy shifts on the part of clients. Under AFDC, American caseworkers operated with limited resources and focused predominantly on getting everybody’s checks out on time. The transition to TANF implies a stronger social work approach to clients which caseworkers might feel ill-equipped to deal with. In both countries, the question remains what clients want themselves. Not having been consulted in the policy making process, single mothers are now required to change important aspects of their lives. However, it is entirely possible that for some women, this is a long awaited opportunity to enter the labor market, while for others, it means giving up an important aspect of their identity as mothers.

This comparative project is unique in analyzing the construction of gendered citizenships at the subnational level. I look at the welfare office as one of the locations in which citizens meet the state face-to-face and suggest that the state is not simply a uniform entity that can easily enforce certain gendered constructs. Using the concept of needs interpretation I will show how the state, represented by those who implement policy, is implicated in the construction of gendered needs associated with mothering, work, and care. Comparing these two states will show how policies that appear to approach the problems of single mothers with very similar solutions, can have different results in the type of gendered citizenship ultimately constructed. I expect that these differences result not only from differences in welfare state regimes but also from differences in the levels of knowledge, the amount of discretion, and kinds of ideologies held by both clients and caseworkers. While each state is actively trying to reinterpret women’s needs in such a way as to give primacy to mothers’ material care provision over other forms of care, only an analysis of the interactions between clients and caseworkers can show how each state constructs a new type of female citizen.

End Notes

1. Policies geared to single mothers form three clusters. In the first cluster we find countries such as France, Austria, Finland and to a lesser degree Germany, which have historically emphasized material support through work supplemented by family allowances, except in the case of very young children (up to age 3). A second cluster consists of Scandinavian nations, which combine labor market and family policies so as to enable full employment by all citizens regardless of gender and parental status. The third cluster consists of the countries discussed in this section. .

2. I have no further information on Luxembourg and will leave it out of the remaining discussion.

3. Forty-five percent of mothers in two parent families work an average of 21 hours a week; 37 percent of single mothers work an average of 28 hours (Niphuis-Nell 1997, p. 96). For the last 15 years, single mothers have headed around ten percent of all families with children under 18 (Bussemaker et al. 1997).

4. Between 1970 and 1996, the labor force participation rate of married women with children under six years old rose from 30.3 to 62.7 percent (Statistical Abstracts of the United States 1997, table 631). In 1996, the labor force participation rate for divorced, separated, and widowed mothers of children under six, was 69.2 percent. In 1996, single women with children, the target of PRWORA, headed 27 percent, or 11.7 million, of all households with children under 18. In January 1996, 4.6 million families received AFDC.

Bane, Mary Jo, and David T. Ellwood. 1994. Welfare Realities: from Rhetoric to Reform. Cambridge, MA.: Harvard University Press.

Boorsma, S., B. van den Brande, P. Renooy, and R. van der Veen. 1994. Een activerende bijstand? (Activating social assistance?). Den Haag: Uitgeverij Vuga.

Brodkin. 1990. “Implementation as Policy Politics.” Pp. 107-118 in Implementation and the Policy Process:  Opening Up the Black Box, edited by Dennis J. Palumbo and Donald J. Calista. New York: Greenwood Press.

Brodkin, Evelyn Z. 1997. “Inside the Welfare Contract:  Discretion and Accountability in State Welfare Administration.” Social Service Review :1-33.

Bussemaker, Jet, Annemieke van Drenth, Trudie Knijn, and Janneke Plantenga. 1997. “Lone Mothers in the Netherlands.” Pp. 96-120 in Lone mothers in European welfare regimes:  shifting policy logics, edited by Jane Lewis. London; Philadelphia Pa.: J. Kingsley Publishers.

de Bruin, Paul, Manon Goddijn, Eiko Smid, and Cornelis van der Werf. 1997. “Het gemeentelijk armoedebeleid:  een inventarisatie.  (Municipal poverty policy:  an inventory).” : Ministerie van Social Zaken en Werkgelegenheid.

Duncan, Simon, and Rosalind Edwards. 1997a. “Single Mothers in Britain:  Unsupported Workers or Mothers?” Pp. 45-79 in Single mothers in international context: mothers or workers?, edited by Simon Duncan and Rosalind Edwards. London; Bristol, Pa.: University College London.

Duncan, Simon, and Rosalind Edwards (Eds.). 1997b. Single mothers in international context: mothers or workers? London; Bristol, Pa.: University College London.

Edin, Kathryn. 1991. “Surviving the Welfare System:  How AFDC Recipients Make Ends Meet in Chicago.” Social Problems 38:462-474.

Edin, Kathryn. 1998. “Ears to the Ground:  Assessing the Impact of Welfare Reform through Community-Based Ethnographic Research.” in American Sociological Association Annual Meetings. San Francisco.

Edin, Kathryn, and Laura Lein. 1997. Making Ends Meet:  How  Single Mothers Survive Welfare and Low-Wage Work. New York: Russell Sage Foundation.

Esping-Andersen, Gøsta. 1990. The Three Worlds of Welfare Capitalism. Cambridge, UK: Polity Press.

Fraser, Nancy. 1989a. “Struggle over Needs:  Outline of a Socialist-Feminist Critical Theory of Late Capitalist Political Culture.” Pp. 161-187 in Unruly Practices:  Power, Discourse and Gender in Contemporary Social Theory. Minneapolis: University of Minnesota Press.

Fraser, Nancy. 1989b. “Women, Welfare and the Politics of Needs Interpretation.” Pp. 144-160 in Unruly Practices:  Power, Discourse and Gender in Contemporary Social Theory. Minneapolis: University of Minnesota Press.

Fraser, Nancy, and Linda Gordon. 1994. “A Genealogy of Dependency:  Tracing a Keyword of the U.S. Welfare State.” Signs 19:309-336.

Gordon, Linda. 1990. “The New Feminist Scholarship on the Welfare State.” Pp. 9-35 in Women, the State, and Welfare, edited by Linda

Gordon. Madison: University of Wisconsin Press.

Haney, Lynne. 1996. “Homeboys, Babies, Men in Suits:  The State and the Reproduction of Male Dominance.” American Sociological Review 61:759-778.

Heidenheimer, Arnold J., Hugh Heclo, and Carolyn Teich Adams. 1983. Comparative Public Policy:  the Politics of Social Change in Europe and America. New York: St. Martin's Press.

Kamerman, Sheila B. 1995. “Gender Role and Family Structure Changes in the Advanced Industrialized West:  Implications for Social Policy.” Pp. 231-256 in Poverty, inequality, and the future of social policy:  Western states and the new world order, edited by Katherine

McFate, Roger Lawson, and William Julius Wilson. New York: Russel Sage Foundation.
Knijn, Trudie. 1998. “Sociale Diensten en Alleenstaande Moeders:  Een gedeelde cultuuromslag?” Pp. 55-64 in De andere k(l)ant:  Speurtocht naar een klantgerichte sociale dienst,, edited by Paul Lemmen and Yvonne Wijnands. Utrecht: Divosa.

Langan, Mary, and Ilona Ostner. 1991. “Gender and Welfare.” Pp. 127-150 in Towards a European Welfare State?, edited by Graham Room. Bristol: SAUS Publications.

Lewis, Jane. 1992. “Gender and the Development of Welfare Regimes.” Journal of European Social Policy :159-73.

Lewis, Jane. 1997. Lone mothers in European welfare regimes: shifting policy logics. London; Philadelphia, Pa.: J. Kingsley Publishers.

Lipsky, Michael. 1980. Street-level bureaucracy: dilemmas of the individual in public services. New York: Russell Sage Foundation.

Lister, Ruth. 1997. Citizenship:  Feminist Perspectives. New York: New York University Press.

MacLaughlin, Eithne, and Paula Rodgers. 1997. “Single Mothers in the Republic of Ireland:  Mothers not Workers.” Pp. 9-43 in Single Mothers in an International Context:  Mothers of Workers?, edited by Simon Duncan and Rosalind Edwards. London; Bristol, Pa.: University College London.

McHugh, Marilyn, and Jane Millar. 1997. “Single Mothers in Australia:  Supporting Mothers to Seek Work.” in Single mothers in international context:   mothers or workers?, edited by Simon Duncan and Rosalind Edwards. London; Bristol Pa.: University College London.

Meyers, Marcia K., Bonnie Glaser, and Karen Mac Donald. 1998. “On the Front Lines of Welfare Delivery:  Are Workers Implementing Policy Reform?” Journal of Policy Analysis and Management 17:1-22.

Millar, Jane. 1996. “Mothers, workers, wives:  comparing policy approaches to supporting lone mothers.” Pp. 97-113 in Good Enough Mothering?  Feminist Perspectives on Lone Motherhood, edited by Elizabeth Bortolaio-Silva. London; New York: Routledge.

Morgen, Sandra. 1990. “Two Faces of the State:  Women, Social Control and Empowerment.” in Uncertain Terms, edited by Faye Ginsberg and Anna Lowenhaupt Tsing. Boston: Beacon Press.

Murray. 1993. “The Coming White Underclass.” Pp. 1-4 . Washington: American Enterprise Insitute for Public Policy Research.

Niphuis-Nell, Marry. 1997. “Eenoudergezinnen, stiefgezinnen, en niet-verzorgende ouders.” Pp. 85-114 in Sociale atlas van de vrouw; deel 4, edited by Marry Niphuis-Nell. Rijswijk: Sociaal en Cultureel Planbureau.

O'Connor, Julia. 1993. “Gender, class and citizenship in the comparative analysis of welfare state regimes:  theoretical and methodological issues.” British Journal of Sociology 44:501-518.

Orloff, Ann Shola. 1993. “Gender and the Social Rights of Citizenship:  The Comparative Analysis of Gender Relations and Welfare States.” American Sociological Review 58:303-328.

Orloff, Ann Shola. 1996. “Gender in the Welfare State.” Annual Review of Sociology 22:51-78.
Palumbo, Dennis James, and Donald J. Calista. 1990. Implementation and the Policy Process:  Opening up the Black Box. New York: Greenwood Press.

Phillips, Anne. 1992. “Universal Pretensions in Political Thought.” Pp. 10-30 in Destabilizing Theory:  Contemporary Feminist Debates, edited by Michele Barrett and Anne Phillips. Stanford: Stanford University Press.

Pressman, Jeffrey L., and Aaron B. Wildavsky. 1984. Implementation : how great expectations in Washington are dashed in Oakland : or, why it's amazing that federal programs work at all, this being a saga of the Economic Development Administration as told by two sympathetic observers who seek to build morals on a foundation of ruined hopes. Berkeley: University of California Press.

Pringle, Rosemary, and Sophie Watson. 1992. “Women’s Interests’ and the Post-Structuralist State.” Pp. 53-73 in Destabilizing Theory:  Contemporary Feminist Debates, edited by Michele Barrett and Anne Phillips. Stanford: Stanford University Press.

Sainsbury, Diane (Ed.). 1994. Gendering Welfare States. London; Thousand Oaks, Calif.: Sage Publications.

Sainsbury, Diane. 1996. Gender, Equality, and Welfare States. Cambridge England; New York, NY: Cambridge University Press.

Scott, Joan Wallach. 1988. Gender and the Politics of History. New York: Columbia University Press.

Williams, Fiona. 1995. “Race/Ethnicity, Gender, and Class in Welfare States:  A Framework for Comparative Analysis.” Social Politics 1995:127-59.