Lone mothers and welfare-to-work policies in Japan and the United States: towards an alternative perspective.

This paper compares recent efforts to reduce lone mothers' reliance on cash assistance and support their increased participation in the workforce and economic independence in Japan and the United States. Similar to reforms introduced in the U.S. in 1996, lone mother policies in Japan have been subject to a series of cuts leading to the introduction of time limits and work-related programs in 2002. In this paper, we examine the character of recent welfare reforms in both countries and their implications for lone mothers' welfare and economic independence. Based on Japan's experience and recent lessons from the U.S., we show the limitations of a focus on caseload reduction and work participation rates, and instead highlight the importance of addressing lone mothers' low wages in form of policies for the working poor.

Keywords: Japan, single mothers, employment, welfare reform, TANF

**********

Welfare support for lone mother families has become a major concern of policy makers in most advanced industrialized countries. Due to a significant increase in the number of divorced and never-married mothers, as well as their frequent reliance on public support, welfare expenditure on lone mothers has been subject to controversy and reform in a number of countries. Reforms introduced in the United States in 1996 have responded to criticisms of Aid to Families with Dependent Children (AFDC) with policies which emphasize independence through work rather than 'dependence on welfare,' dramatically reducing reliance on cash assistance. Similarly, in Japan, the major source of support for divorced and never-married mothers, the dependent children's allowance (jido fuyo teate), has recently been subject to restructuring. Responding to an increasing demand, policy makers have restricted its conditions of eligibility and significantly reduced the amount of support in a series of cuts in the 1980s and 1990s. Policy revisions in 2002, moreover, have introduced a five year time limit and an increased emphasis on income from work. Similar to welfare-to-work policies in the U.S., also in Japan, new programs and services for lone mothers are now aiming to reinforce their self-sufficiency through work (Fujiwara, 2003).

In this paper, we closely examine recent reforms of lone mother policies in Japan and the United States and assess their implications for lone mothers' welfare and independence. How can welfare-to-work policies enable lone mothers to become economically self-sufficient and independent from state support? Based on an analysis of lone mother policies and work patterns in Japan and the United States, we illuminate the challenges of facilitating a shift from 'welfare to work' as anticipated by welfare reform. We show that even though lone mothers in Japan have the highest workforce participation rate among advanced industrialized nations, engagement in paid work in itself does not necessarily move them out of poverty or beyond public support. Implementing welfare-to-work policies, therefore, not only involves moving lone mothers into the workforce but ensuring a living wage. In comparing the Japanese case with recent trends in post-reform United States, our paper explores key issues which need to be taken into consideration in facilitating lone mothers' welfare and independence after reform.

Welfare-to-work policies, by aiming for lone mothers' economic independence from the state through wage work, address issues that have long been at the center of discussions of the gendered character of welfare regimes. In general, lone mothers have been seen as a 'litmus test,' which illuminates the gendered character of a welfare state regime (Hobson, 1994; Kilkey & Bradshaw, 1999; Lewis, 1997). As comparative research has shown, lone mothers' living conditions vary considerably across welfare regimes. Whereas lone mothers tend to receive protection as mothers in some welfare regimes, they are primarily treated as workers in others (Duncan & Edwards, 1997). The degree to which welfare regimes reinforce a male breadwinner model of family, and women's dependence on husbands or support women's dual role as workers and caregivers has significant implications for the welfare of lone mothers (cf. Lewis, 1997)

Welfare policies have the most direct implications for lone mothers but they also shape the options and lifestyles of other women. Whether or not lone mothers have access to the labor market and a living wage affects the economic independence of other women as well. The general welfare of lone mothers, whether supported through paid work or government allowances, has also a significant impact on women's independence from marriage and family support. As Ann Orloff (1993) has argued, social citizenship (defined as basic right to a minimum level of welfare), viewed from the perspective of gender, not only depends on the extent of state compensations for failures of the market (as in form of unemployment insurance, disability or old age pensions), but also women's access to work and income and their ability to maintain an autonomous household outside of marriage and family support. Lone mothers' welfare after the introduction of welfare-to-work reforms therefore serves as an important indicator of women's economic autonomy and independence in the labor market and the welfare state.

Our analysis addresses welfare-to-work policies and their ability to foster economic independence from several angles. First, we introduce the general situation of lone mothers in Japan and major policies supporting their welfare. As most readers will be familiar with the U.S. context, the first two parts focus mostly on Japan. We then examine recent policy changes in the United States in light of Japan's experience. We conclude with observations on the consequences of welfare-to-work policies and areas that will require policy attention in order to foster lone mothers' welfare and independence.

1. Situating Lone Mothers in Japan

Before examining the character of lone mother policies in Japan, it is necessary to get a general sense of the population they aim to support. Even though policy rhetoric in Japan has used the United States as a model for reform, lone motherhood remains a comparatively marginal phenomenon in Japan. Whereas lone parent families accounted for 27.3% of family households with children under 18 in the U.S. in 1998 (U.S. Census Bureau, 1998), in Japan, they accounted for only 6.4% (Nihon rodo kenkyu kiko, 2003). Also, whereas public discourse in the United States centers on never-married teenage mothers, it was an increase of divorcees that placed lone mothers on the policy agenda in Japan. As indicated in figure 1, Japan's divorce rate increased significantly from a postwar low of 0.73 per thousand population in 1963 to 1.26 in 1988, and has experienced a steady rise in the 1990s reaching a high of 2.27 in 2000. Whereas widows predominated in early postwar Japan, in 1998, divorcees accounted for 68.4% of lone mothers in Japan.

[FIGURE 1 OMITTED]

Never-married motherhood, while increasing, remains at only 7.6% (table 1). In the United States by contrast, never-married mothers predominate (42.2%) followed by divorcees (34.8%), and only 4.4% widows (U.S. Census Bureau, 2000). Since most lone mothers in Japan have become single parents only after marriage, childbirth and divorce, they are generally older than their U.S. counterparts and in their 30's and 40's (Koseirodosho, 2001). Lone motherhood in Japan is thus largely associated with divorced mothers in middle age rather than teenage mothers. Overall, even though demand for support has been increasing persistently in the past thirty years, policies in Japan address the needs of less than 800,000 households, which may seem negligible in comparison to a caseload of five million recipients of Temporary Assistance for Needy Families (TANF) targeted by reforms in the United States.

Public discourse, meanwhile, has mostly focused on economic consequences of becoming a lone mother rather than moral issues; they are neither associated with a discourse of dependency (Fraser & Gordon, 1994) or sexuality as in the United States (Luker, 1996). In the early postwar years, widows, who had lost their spouses during the war, were pitied for their misfortune of losing their breadwinner and living in poverty. Similarly, until the 1970's, divorce was often associated with desertion, and divorcees received sympathy for their economic struggles and failed marriage. Since the 1980's and 1990's, social movements have begun to step into the public arena and change this image (cf. Single Mothers' Forum, 1993). Their efforts have contributed to the emergence of a more cheerful image of lone motherhood, which has however also been met with criticism of women who abandon family values and selfishly pursue new lifestyles of their own. Rather than being portrayed as 'welfare queens,' however, in Japan, they have made headlines for failing to claim public support leading to several cases of starvation (Bokensha henshubu, 1999; Mizushima, 1990). In short, even today, lone motherhood continues to be thought of as a situation of harsh economic realities and social isolation.

One of the reasons why lone mothers in Japan have not been associated with welfare dependency is that their work participation rate is significantly higher than in other advanced industrialized nations. Uzuhashi's analysis of the early 1990s shows that whereas fewer lone mothers (60%) than married mothers (64%) worked in the United States, in Japan, they worked substantially more than married mothers (87% as opposed to 54%). Moreover, their work participation rate exceeds that of women in France and Sweden, who generally have among the highest work participation rates among industrialized countries (Uzuhashi, 1997, 138). Even more striking is the fact that their work participation rate has been over 80% for the entire postwar period (table 2). The tendency not to work among married mothers with small children, by contrast, remains remarkably persistent. As the well-known M-shaped labor force participation rate of women in Japan indicates, it is common for married women to retire from work with childbirth and return only as part-time workers in middle-age (Iwai & Manabe, 2000). Thus, in addition to being pitied for their economic struggles, lone mothers are seen as deprived of the capability of taking care of their children, as prescribed by the normative middle-class ideal of the full-time housewife and mother.

Despite such high work participation rate, poverty and low incomes are central problems facing lone mothers in Japan. Their income packaging shows that public support accounts only for a minor (yet important) share of their overall income. According to a survey of the Ministry of Health, Labor, and Welfare, in 2000, wages accounted for 78.8% of their income, 7.2% came from social security/allowances, 5.9% from pensions and only 2.6% from family contributions and child support (Koseirodosho, 2002). Even though almost all lone mothers are engaged in paid work, earn on average higher incomes than married mothers, and have a greater tendency to work in full-time jobs and as permanent employees, it remains difficult for them to make ends meet on their incomes from work (Fujiwara, 2005).

In the 1990s, lone mothers' average annual income was less than 40% of the average household income, a ratio that has fallen from 50% in the 1970's (Yuzawa, 1993). This increasing disparity can be related to an overall increase in the income of other households and an increasing number of dual-earner families. Lone mothers' low income also illustrates persisting differences in women's and men's average income: as of the year 2002, women's average wage remained at 64.9% of that of men (Koseirodosho, 2003). There are no data available in Japan that can assess the degree to which their incomes are below the poverty line. It can however be generally stated that 50% of lone mothers have an income that is lower than the amount they would receive if they were receiving public assistance (seikatsu hogo). Similar observations have been made about working lone mothers in post-reform Unites States. Even though an increasing number of lone mothers are working, the majority earn an income below the poverty line (Brauner & Loprest, 1999; Loprest, 2001). Although the demographic characteristics of lone mothers differ quite significantly in Japan and the United States, it seems that in both countries they share the difficulties of making ends meet between public support and income from wages.

2. Lone Mothers and Japan's Welfare-to- Work Regime

The high work participation rate of lone mothers makes Japan appear as an ideal scenario for policy makers in the United States, who have made a greater work participation rate a primary goal of reform. Lone mothers in the United States had long been criticized for their limited engagement in work, and it was argued that the provision of cash assistance reinforces 'dependency' (see Katz, 1993; Murray, 1984). To reduce reliance on AFDC, the Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA) in 1996 introduced (among others) a five year time limit to receiving cash assistance, as well as work requirements to increase participation in paid work. Cash assistance, now renamed TANF was reduced from a social right to a temporary measure aimed to assist mothers in leaving 'welfare' and entering the workforce. Viewed from this perspective, Japan's reform slogan to move lone mothers from welfare to work seems puzzling. If 87% of lone mothers are already working, what are the goals of Japanese welfare-to-work reforms?

On the surface, there are several parallels between policy concerns in the United States and Japan. In both countries, an increase in the number of lone mothers and their reliance on cash assistance has caused concern with rising public expenditure on lone mothers' welfare. In the United States, the number of welfare recipients rose significantly in the early 1970s and reached a new peak of 14 million recipients in the mid 1990s. Although initially hailed as a progressive program (Mink, 1995), AFDC began to attract widespread disapproval with an increase of unmarried African American mothers among recipients (Heclo, 2001). Similarly, in Japan, an increase in the divorce rate and a shift from widowed to divorced lone mothers set the beginning of a series of cuts and the restructuring of support for lone mothers. The number of recipients of the dependent children's allowance--the main source of cash assistance for lone mothers--increased from 154,387 in 1962 (the year of the creation of the program) to 759,197 in 2001 (see Figure 1). Even though the overall numbers are comparatively small, Japan, like the United States has been faced with increasing demand for public support due to demographic changes, making divorced mothers' reliance on cash assistance a primary target of reform.

A primary difference between the two countries however is that whereas few lone mothers receiving AFDC were working, in Japan, reforms have targeted cash assistance for working lone mothers. In other words, rather than entry into the work force per se, policymakers have targeted income supplements that have added to income from wages. Even though Japanese lone mothers are working in 1993, approximately 80% of them qualified for the dependent children's allowance due to a low income. As their incomes tend not to increase significantly over time, many receive support until their children come of age. In 1998, only 7.1% of lone mothers reported having lost eligibility due to an increase in income (Nihon rodo kenkyu kiko, 2003). 'Welfare-to-work,' in the case of Japan is thus not associated with moving lone mothers into the workforce but rather to move them beyond cash assistance. To get a better sense of this scenario, let us review the main elements of lone mothers policies in Japan and how they support lone mothers' welfare and independence.

The dependent children's allowance (jido fuyo teate) has been the major source of support for lone mothers since its establishment in 1962. In early postwar years, the large presence of widowed mothers who had lost their spouses during the war became a public issue. Young in age, with small children, and little to no work experience or education, lone mothers were facing harsh living conditions at the time. Having nothing else to rely on, many received public assistance (seikatsu hogo), a general program established in 1946 to support anyone in need. As public assistance carried a high stigma and public funding was limited during the period of reconstruction, policymakers looked for alternative programs, which would ease the weight on public expenditure. In conjunction with the establishment of a public pension system, widowed mothers' pensions (boshi nenkin) were introduced in 1959, and the dependent children's allowance was introduced as a complementing measure for divorced and otherwise separated mothers. Rather than covering all needs of mothers and children, the program was aimed to contribute to their welfare by adding to mothers' income from work.

More specifically, the dependent children's allowance is an income-limited cash grant which supports mothers with dependent children below age 18 who do not have a husband or other sources of support such as widowed mothers' pensions. At the time of establishment, the allowance was paid at a flat rate to mothers with no or a low income. With an increase in the number of divorcees among lone mothers and the increasing demand for the dependent children's allowance in the 1970s, revisions narrowed conditions of eligibility considerably. In a major step, in 1985, the amount of the dependent children's allowance was significantly reduced, and the income ceiling lowered from [yen]3.6 million to [yen]3.0 million (which could be compared to an annual income of $30,000 in the United States). As the number of recipients continued to increase, the income ceiling for the dependent children's allowance was further lowered in 1998. In addition, the 1985 revisions divided eligibility for the allowance into a full amount for mothers with an income below [yen]1.7 million, and a partial amount for mothers with a moderate income below [yen]3.0 million (Fujiwara, 1997). In other words, only lone mothers with no or very little income now received the full amount, while the cash benefits for those with a moderate level of income were significantly reduced.

Although the number of lone mothers who received the allowance initially declined (figure 1), it began to increase again by the early 1990's. Faced with an burgeoning annual increase in expenditure on the dependent children's allowance, reforms introduced in 2002 set the beginning for a more radical set of changes. First of all, the amount of the dependent children's allowance was changed from a full and a partial amount to a sliding scale, ranging from [yen]42,360 to [yen]10,000 a month (approximately $400-$100). That is, the amount of the allowance now slowly declined with increase in income, whereas previously it was dispensed in form of two flat rates. In addition, mothers receiving the allowance after 2003 are now subject to a five-year time limit. Originally intended to 'end welfare' like in the U.S., the new scheme will reduce the allowance up to 50% for those who have received the allowance for more than five years. It should be added that lone mothers will only be able to rely on the dependent children's allowance in one consecutive time period up to five years, and will have no possibility to 'save' time for future emergencies as is the case in the United States.

In principle, the dependent children's allowance functions somewhat like the Earned Income Tax Credit (EITC), a refundable tax credit that adds to the income of low-income families in the United States. Like the EITC, the dependent children's allowance has an income ceiling of approximately $30,000 and in most cases adds to mothers' income from work. Of course, whereas benefits increase until a certain income in the case of the EITC, they decline with increase in income in the case of the dependent children's allowance. Unlike the EITC, the dependent children's allowance can however also be received by those without an income--in this case, it might be better compared to AFDC of TANF With the introduction of recent reforms, the dependent children's allowance is now also subject to a time limit, like TANF In essence, Japan has placed a time limit on the income-supplement approximating the EITC and TANF, leaving no alternative source of support such as Food Stamps or the EITC after reaching the time limit.

Beyond the dependent children's allowance, lone mothers in Japan have had few alternative sources of support. In the year 2000, only 10.6% of lone mothers relied on means-tested public assistance (seikatsu hogo) in Japan, a ratio that has remained relatively constant throughout the postwar period (Seikatsu hogo no doko henshu iinkai, 2002). A rigorously means-tested program which allows few assets and savings, and which is associated with a high social stigma, few lone mothers consider public assistance as an option. Although public assistance is officially available for all Japanese in need, in practice, it resembles the Supplementary Security Income (SSI) in the United States. Although applicants do not need to have a disability, its rigorous screening process and strict conditions of eligibility make it primarily a source of support for those with injuries, disabilities or other barriers to work.

In addition, although most lone mothers in Japan are divorced, reliance on child support payments is limited. One particularity about Japan is that 90% of divorces in Japan are processed by 'mutual agreement' (kyogi rikon), that is, out of court. Since child support payments are required by the Japan's Civil Code but are not enforced, only those who go to court have the possibility of laying claims on child support payments. Even then, however, women have no effective means to enforce child support payments (Shimoebisu, 1993). In 1998, only 20.8% of divorced mothers reported receiving child support payments at the time of the survey; 60.1% answered that they had never received any child support payments (Koseirodosho, 2001).

In line with the main goals of reforms introduced in 2002, the government has recently begun a campaign to foster a greater enforcement of child support payments. Revisions have proposed regulations which make it possible for lone mothers to lay claim on past unpaid child support payments, and a new system where payments will be taken directly out of fathers' paychecks. Child support enforcement of this kind is however only possible if child support payments have been officially agreed on in the divorce settlements. It should also be added that the new regulations, 80% of child support payments are counted toward lone mothers' income. Since child support payments even now are not guaranteed, and a higher overall income will lower the amount of the dependent children's allowance, the new rules come with a significant disincentive to pursue child support payments.

Beyond cash assistance, policies dating back to the early postwar period have aimed to support lone mothers' economic self-sufficiency and welfare through various programs and services. One of the oldest programs for lone mothers, established in 1953, is a low or no-interest loan program (boshi fukushi shikin). These loans can be used for various purposes, such as to pay for children's education, vocational training, and the establishment of a small business such as a dry cleaning shop or tobacco store. Also, all lone mothers are covered by the national health insurance (generally referred to as kokumin kenko hoken) which covers all residents of Japan regardless of employment status. As insurance payments depend on income and size of household they can be quite substantial. Nevertheless, medical coverage does not constitute a major concern for lone mothers since all families are covered, fees are lower for low-income households and some localities waive co-payments for lone mothers with a low income.

In addition, a number of services not limited to lone mothers support their employment. In the late 1960s, lone mothers were defined as a group that is 'hard to employ' (shushoku kon' nansha, this also includes a number of other groups including the elderly, disabled, minority groups and ex-coal mine workers), and became eligible for a range of special work-related programs. They can also access job centers, which are available nation-wide and serve all types of job seekers as well as the unemployed. The centers provide information on job openings as well as consultation. Current policies place further emphasis on a program directed at employers, which subsidizes the wages of lone mothers under the condition that they will be employed as full-time workers after six months. Lone mothers can also attend job training programs run by local governments at no cost. For specific qualifications, such as nurses or elderly care providers, they can also receive a small monthly allowance during the period of training. Such programs are very similar to the job centers, training programs and wage subsidies recently introduced in the US but notably were established at a much earlier stage.

Finally, to allow mothers to enter the workforce, lone mothers are given preference in placing their children in public day care centers, whose fees are subsidized and calculated based on income. According to a 1998 survey with multiple answers, the majority of lone mothers used, among others, day care centers to care for their children (60.6%) or sent them to kindergartens (13.0%). In contrast to the U.S. where informal and in-home care is common, only 12.0% of lone mothers in Japan reported relying on family members or relatives (2.1%) at any time for child care (Koseirodosho, 2001). More generally, the presence of a strong network of subsidized public day care centers points to the fact that state policies have seen lone mothers (as well as working class mothers) as workers rather than mothers. Day care services and programs have recently been extended to accommodate the needs of working mothers. To ease the balance between work and family needs, day care services were expanded from subsidized daytime public day care to evening and overnight services to cover for parents' illness and overtime work.

Together, Japanese policies provide a quite comprehensive set of allowances and services, that possibly make Japan the 'oldest' welfare-to-work regime. Conceived at a time when public funding was scarce, the dependent children's allowance has been aimed to supplement rather than substitute lone mothers' wages, and has been backed by various work-supporting programs and services. Public assistance, meanwhile, has only supported those with major barriers to work, but has never attracted much attention in terms of caseloads. Yet, although Japan has achieved an astonishingly high work participation rate among lone mothers, the continuing demand for the dependent children's allowance shows that participation in the workforce in itself does not increase mothers' incomes beyond the poverty line. Consequently, policy makers in Japan face the challenge not merely to move lone mothers into the workforce, but allow them to earn an income that will make them independent from public support.

3. U.S. Welfare Reform Reexamined

Our analysis of lone mother policies in Japan has important implications for our understanding of welfare-to-work policies in the United States. Even though Japanese policies have already achieved the major goals of U.S. reforms--a low reliance on public assistance and a high work participation rate--they have not eliminated lone mothers' need for public support because of their continued low incomes. Thus, in considering U.S. reforms, we need to examine not only the extent to which policies have moved mothers away from cash assistance and into the workforce, but rather how they have addressed the needs of the working poor.

The rapid decline in the TANF cash assistance load has been central to many discussions of the U.S. welfare reform. Between 1993 (when some states began to implement new rules under state waivers) and 2000, caseloads declined by an astonishing 56% (U.S. Department of Health and Human Services, 1999). While these figures are striking, they also raise concerns about the welfare of those who have left the welfare system. Although there is evidence that the majority of welfare leavers work, there is also reason to suggest that lone mothers often experience unstable employment, largely remain in low wage jobs, and continue to rely on some type of government assistance (Brauner & Loprest, 1999; Miller, 2002).

In Wisconsin, where policies have made a distinctive effort to move welfare recipients into the work force, job retention rates were strikingly low. According to one survey, the majority of surveyed employers had retained less than half of the welfare leavers they had hired two years earlier (Martin & Alfred, 2002). Similarly, a study of current and former recipients in Illinois indicates that 52.8% of those who left welfare currently held a different job than at the time of exit (University Consortium on Welfare Reform, 2002). Also studies of other states indicate that the average duration of welfare leavers' employment is short, from 3-4 months to 8 months (Martin & Alfred, 2002). Even more worrisome is the fact that despite employment, most mothers earn an income below the poverty line although they tend to work more than 30 hours a week (Brauner & Loprest, 1999). According to Loprest's study, 52% of recipients who left welfare recently had an income below the poverty line (Loprest, 2001). As a consequence, many lone mothers remain eligible and receive support as low-income families, although not in form of full cash grants. Similar to Japan, therefore, U.S. policy makers continue to face the challenge of fostering economic independence among lone mothers. How have U.S. policies tried to address the needs of lone mothers after welfare reform?

Even after reforms, a number of measures continue to support lone mothers' livelihood. Although TANF cash grants are no longer an option in the long term, Supplementary Security Income (SSI) is available to lone mothers who have mental or physical disabilities. As many TANF recipients have impairments (U.S. General Accounting Office, 2002b), the program provides a sanctuary for those with documented barriers to work, as it has (as yet) no time limits or work requirements. Recent research suggests that some recipients of AFDC may have shifted to the Supplemental Security Income (SSI) caseload after 1996 (Karoly, Klerman, & Rogowski, 2001; Schmidt & Sevak, 2000), as time limits and work requirements created incentives for families, which previously relied on AFDC, to apply for SSI. Particularly in states with more aggressive reform policies, it appears that women with particular disabilities are now more likely to receive SSI than TANF (Schmidt & Sevak, 2000). It can therefore be estimated that welfare reforms not merely shifted TANF recipients into the workforce, but possibly also to other welfare caseloads.

While fewer lone mothers are now relying exclusively on TANF cash assistance, an increasing number of them now qualify and receive support and services based on their low incomes. Medicaid, although tied to receiving AFDC in the past, has become available to low-income families. In some states, families can extend their coverage during their transition to work, up to one year (Greenstein & Guyer, 2001). Moreover, a significant proportion of welfare leavers continue to receive Medicaid, from 53% in Indiana to 83% of continuous leavers in Wisconsin (Brauner & Loprest, 1999). In short, although TANF caseloads have been declining, many families continue to rely on Medicaid to assure their well-being. Similarly, Food Stamps remain a valuable source of support for low-income families after reform. In fact, the ratio of households with an employed adult who receive Food Stamps has been increasing (Greenstein & Guyer, 2001). Brauner and Loprest's survey (Brauner & Loprest, 1999) further shows that 30-60% of families in the states surveyed continued to receive Food Stamps after leaving welfare. Similar to Japan, therefore, there are now an increasing number of families who do not receive a full cash grant but get Food Stamps in addition to their income from work.

Beyond welfare policies, the EITC has become an important source of support for the working poor, including lone mothers. Since its establishment in 1975, the EITC has expanded into a sizable refundable tax credit, which allows low-income mothers with dependents to almost double their income. Whereas a never-married non-working mother would receive $7,717 of welfare benefits in 1998, a never-married working mother with a salary of $10,000 could increase her disposable income to $14,593 with the credit (Ellwood, 2000). Under the condition of employment, therefore, lone mothers are eligible to substantial cash assistance in form of the EITC.

Adding to existing programs, many states have also invested in new childcare assistance programs and job centers to facilitate TANF recipients' employment. Although these programs do not dispense cash, they have added to the workload of social workers in charge of TANF programs. Needless to say, since these programs are not limited to recipients of TANF cash grants, the caseload welfare offices and job centers handle is larger than the TANF caseload alone. According to calculations of the General Accounting Office, at least 46% more families than were in the TANF caseload received services paid for by TANF dollars in 2002 (U.S. General Accounting Office, 2002a).

A recent study of Wisconsin illustrates the magnitude of caseloads not counted towards the official TANF caseload that has been focus of so much attention in the aftermath of reform. Wisconsin is known for its rigorous welfare-to-work program, which introduced reforms at an early stage and drastically reduced caseloads with the introduction of TANF. Wisconsin's AFDC/TANF caseload fell by 93% within 13 years, from nearly 100,000 in 1987 to 6,500 in 2000. Yet, if we include cash assistance, child care, Food Stamps, Medicaid, assistance to care for a related or disabled child, and case management into the definition of 'caseload,' the total number of cases receiving any one of these services declined only by 3% between 1995 and 2000 (Swartz, 2001).

Rather than the size of the caseload, it is its composition, which has significantly changed with the introduction of reforms. Among 118,585 families who received Medicaid, Food Stamps, or AFDC in Wisconsin in 1995 for instance, 62% of families received AFDC (which includes those who also received Medicaid and Food Stamps), and 38% of cases received only Medicaid and Food Stamps. In 2000 however, of 111,830 cases, only 9.5% received TANF cash assistance and 90.5% received Medicaid and Food Stamps only (Swartz, 2001). Although the composition of cases administered has experienced dramatic changes, the number of cases receiving some form of support has not changed as dramatically as presented in discussions of the declining welfare caseload.

Viewed from this perspective, in the U.S. as in Japan, policy makers continue to face the question of how to facilitate working lone mothers' economic independence. Welfare reforms have enforced a higher work participation rate and eliminated cash assistance as an entitlement. In visibly reducing cash assistance caseloads, reforms have been able to respond to criticism, which sees cash assistance to non-working mothers as inadequate. Yet, with the introduction of new programs and the expansion of old ones to support the working poor, they have also added to other caseloads. In the U.S. as in Japan, it appears that reforms have curtailed women's access to cash assistance and ability to make claims as mothers, but have hardly eliminated lone mothers' need for public support. In recognition of this problem, U.S. policies have shifted their focus to programs and measures, which support the working poor. Such policies, in light of Japan's experience, will be crucial in order to facilitate not merely mothers' entrance into the labor market but help them to attain a living wage.

4. Conclusion

Our comparison of Japan and the United States provides a number of lessons for our understanding of welfare-to-work policies. First, whereas discussions of welfare reform in the United States have been preoccupied with moving lone mothers from cash assistance into the workforce, the Japanese experience shows that work in of itself does not necessarily facilitate independence from state assistance. Even though Japanese policies have lead to a work participation rate as high as 87%, they have, in the course of forty years, not been able to raise lone mothers' incomes much beyond the poverty line. Welfare leaver studies in the United States illustrate similar trends. Even though cash assistance loads have been declining, and work participation rates among lone mothers have been increasing, they have only increased demand for support for the working poor. Thus, even though rising work participation rates have been held up as a sign of success of welfare reforms, they have also generated new areas of concern which are likely to preoccupy policy makers for some time.

Furthermore, the situation of lone mothers in Japan, rather than an ideal scenario, is highly precarious. The dependent children's allowance has from the beginning, constituted an income supplement rather than full cash grant. Reforms introduced in 2002 have introduced time limits to the dependent children's allowance, which, in the absence of Food Stamps, sufficient enforcement of child support and policies equivalent to the EITC, constitutes a removal of the only source of income support beyond wage work. To be sure, also in Japan, lone mothers have public assistance (seikatsu hogo) as a safetynet to rely on. Yet in light of the significant disincentives associated with the program, it is unlikely that caseloads will shift toward public assistance in Japan. Given that engagement in full-time work has not allowed mothers to move beyond the income limit of the dependent children's allowance in the past, it also seems unlikely that removal of this income supplement will increase their income. Instead of removing cash assistance, it appears imperative to address the reasons for lone mothers' continued low income. The now almost yearly occurrence of deaths of mothers with children because of poverty only illustrates the desperation, isolation, and lack of a sense of entitlement to support among lone mothers, which is only reinforced by current reforms.

Our observations on lone mother policies in Japan provide important lessons for welfare-to-work policies in the United States. Welfare reforms have set rigid limits on lone mothers' entitlement, and those who do not or cannot work are likely to be in great difficulties. In comparison to Japan, however, it appears that at least a certain safetynet has been maintained with the expansion of the EITC, and the maintenance of Food Stamps and SSI. Whereas Japanese reforms have reduced access to income supplements, the EITC was expanded supplements for lone mothers with low wages, to support the idea that 'work pays.' Those who left welfare but have low incomes thus remain eligible and, as we have seen, continue to receive other forms of government support. In paying particular attention to policies, which support the working poor, U.S. policies have made important steps toward supporting lone mothers' economic independence despite low incomes. Maintenance of a strong safetynet and income supplements, in light of the discussion in this paper, will be crucial for the welfare of lone mothers and their children even when lone mothers are working.

Finally, our comparison provides important insights into the gender dynamics of welfare-to-work policies in Japan and the United States. As in the United States, also in Japan, work-centered policies have eliminated full-time care-giving as an option for poor mothers (cf. Abramowitz, 1988; Orloff, 2001). Yet, even though lone mothers are working, access to work alone does not seem to allow them to become economically self-sufficient. Viewed from this perspective, one of the major problems with welfare-to-work policies seems to be that they presume an idea of economic self-sufficiency that is modeled on a male worker: the problem of mothers' welfare reliance is to be solved by making them into full-time workers. Yet, as Orloff (1993) has shown, the dynamics of social citizenship are quite different when viewed from a gender perspective. Although welfare-to-work policies allow lone mothers to make claims as workers (as in the case of the EITC), policies have yet to address inequalities resulting from lone mothers' dual role as workers and as caregivers (cf. Orloff, 2001). As Japan's postwar experience has shown, the establishment of day care centers and mothers' participation in the labor market alone does not allow mothers to move beyond state support. Rather, state interventions will need to address gender inequalities in the labor market, and extend services and regulations, which allow women in general to balance work and motherhood and attain a living wage.

References

Abramowitz, M. (1988). Regulating the Lives of Women. Boston: South End Press.

Bokensha henshubu (Ed.). (1999). Kurushii nikki: tengoku ni iru otto e [A diary of suffering: to my husband in heaven]. Tokyo: Bokensha.

Brauner, S. & Loprest, P. (1999). Where Are They Now? What States' Studies of People Who Left Welfare Tell US. New Federalism(A-32 (May)).

Duncan, S. & Edwards, R. (Eds.). (1997). Single Mothers in an International Context: Mothers or Workers? London: UCL Press.

Ellwood, D. T. (2000). The Impact of the Earned Income Tax Credit and Social Policy Reforms on Work, Marriage, and Living Arrangements. National Tax Journal, 53(4), 1063-1105.

Fraser, N. & Gordon, L. (1994). A Genealogy of Dependency: Tracing a Keyword of the U.S. Welfare State. Signs, 19(2), 309-336.

Fujiwara, C. (1997). Boshisetai no shotokuhosho to jidofuyoteate [Lone mother households, income security and the dependent children's allowance]. Josei to rodo 21, 6(23), 6-28.

Fujiwara, C. (2003). Jidofuyoteate to shugyo shiensaku no kadai [The dependent children's allowance and work support policies: current issues]. Josei rodo kenkyu, 44.

Fujiwara, C. (2005). Hitorioya no shugyo to kaisosei [Class aspects of lone parents' work patterns]. Shakai seisaku gakkai shi 13.

Greenstein, R. & Guyer, J. (2001). Supporting Work through Medicaid and Food Stamps. In R. Haskins (Ed.), The New World of Welfare (pp. 335-368). Washington, D.C.: Brookings Institution.

Heclo, H. (2001). The Politics of Welfare Reform. In R. Haskins (Ed.), The New World of Welfare (pp. 169-200). Washington, D.C.: Brookings Institution.

Hobson, B. (1994). Solo Mothers, Social Policy Regimes, and the Logics of Gender. In D. Sainsbury (Ed.), Gendering Welfare States (pp. 170-187). London: Sage.

Iwai, H. & Manabe, R. (2000). M-jigata shugyo pataan no teichaku to sono imi [The persistence of the M-shaped work pattern and its meaning]. In K. Seiyama (Ed.), Nihon no kaiso shisutemu 4: jendaa/shijo/kazoku (pp. 67-91). Tokyo: University of Tokyo Press.

Karoly, L., Klerman, J. A., & Rogowski, J. A. (2001). Effects of the 1996 Welfare Reform Changes on the SSI Program. In R. Haskins (Ed.), The New World of Welfare (pp. 482-499). Washington, D.C.: Brookings Institution.

Katz, M. (1993). The Urban "Underclass" as a Metaphor of Social Transformation. In M. Katz (Ed.), The "Underclass" Debate (pp. 3-23). Princeton, NJ: Princeton University Press.

Kilkey, M. & Bradshaw, J. (1999). Lone Mothers, Economic Well-Being, and Policies. In D. Sainsbury (Ed.), Gender and Welfare State Regimes (pp. 147-84). Oxford: Oxford University Press.

Koseirodosho. (2001). Zenkoku boshi setai to chosa kekka no gaiyo [National survey on lone mother and other households]. Tokyo: Koseirodosho.

Koseirodosho. (2002. Kokumin seikatsu kiso chosa [Basic Survey on National Life]. Tokyo: Koseitokeikyokai.

Koseirodosho. (2003). Josei rodo hakusho [White paper on women's work]. Tokyo: 21 seiki shokugyo zaidan.

Lewis, J. (Ed.). (1997). Lone Mothers in European Welfare Regimes. London: Jessica Kingsley Publishers.

Loprest, P. (2001). How are Families That Left Welfare Doing? A Comparison of Early and Recent Welfare Leavers. New Federalism, Series B(No B-36).

Luker, K. (1996). Dubious Conceptions: The Politics of Teenage Pregnancy. Cambridge, MA: Harvard University Press.

Martin, L. G. & Alfred, M. (2002). What Employers and W-2 Job Experts Think About Retention and Advancement Barriers and Services. Madison: Wisconsin Department of Workforce Development.

Miller, C. (2002). Leavers, Stayers, and Cyclers: An Analysis of the Welfare Caseload: Manpower Demonstration Research Corporation.

Mink, G. (1995). Wages of Motherhood. New York: Sage.

Mizushima, H. (1990). Okasan ga shinda [Mother has died]. Tokyo: Hitonaru shobo.

Murray, C. (1984). Losing Ground: Basic Books.

Nihon rodo kenkyu kiko. (2003). Boshi setai no shuro shien ni kansuru kenkyu [Research on work support for lone mothers]. Nihon rodo kenkyu kiko.

Orloff, A. S. (1993). Gender and the Social Rights of Citizenship: The Comparative Analysis of Gender Relations and Welfare States. American Sociological Review, 58, 303-328.

Orloff, A. S. (2001). Farewell to Maternalism: Welfare Reform, Ending Entitlement for Poor Single Mothers, and Expanding the Claims of Poor Employed Parents. San Francisco: Paper presented at the Annual Meeting of the American Political Science Association.

Schmidt, L. & Sevak, P. (2000). AFDC, SSI and Welfare Reform Aggressiveness: Caseload Reductions vs. Caseload Shifting: Population Studies Center, Institute for Social Research, University of Michigan.

Seikatsu hogo no doko henshu iinkai. (2002). Seikatsu hogo no doko [Trends in public assistance]. Tokyo: Chuo hoki.

Shimoebisu, M. (1993). Boshikatei e no shakaiteki shien [Social support for lone mothers]. In Shakaihoshokenkyujo (Ed.), Josei to shakai hosho (pp. 247-266). Tokyo: Tokyodaigakushuppankai.

Single Mothers' Forum. (1993). Boshikatei ni kanpai! [Cheers to the single mothers]. Tokyo: Gendai shokan.

Swartz, R. (2001). What is a "Case" in Post-Reform Wisconsin ? Reconciling Caseload with Workload: Hudson Institute, Welfare Policy Center.

U.S. Census Bureau. (1998). Household and Family Characteristics: March 1998. Current Population Reports P20-515.

U.S. Census Bureau. (2000). America's Families and Living Arrangements. Current Population Reports P20-537.

U.S. Department of Health and Human Services. (1999). Characteristics and Financial Circumstances of TANF Recipients, Fiscal Year 1999. Washington, D.C.

U.S. General Accounting Office (2002a). States Provide TANF-funded Work Support Services to Many Low-Income Families Who Do Not Receive Cash Assistance.

U.S. General Accounting Office. (2002b). Welfare Reform: Outcomes for TANF Recipients with Impairments. Washington, D.C.: U.S. General Accounting Office.

University Consortium on Welfare Reform. (2002). Welfare Reform in Illinois: Is the Moderate Approach Working?: Illinois Family Study Second Annual Report.

Uzuhashi, T. (1997). Gendai fukushikokka no kokusai hikaku [An international comparison of contemporary welfare states]. Tokyo: Nihonhyoronsha.

Yuzawa, N. (1993). Josei to hinkon [Women and poverty]. In C. Hayashi and fujin fukushi kenkyukai (Eds.), Gendai no baibaishun to josei (pp. 15-22). Tokyo: Domesu Shuppan.

AYA EZAWA

Swarthmore College

Department of Sociology and Anthropology

CHISA FUJIWARA

Iwate University

Faculty of Humanities and Social Sciences

Table 1
Trends in lone motherhood in postwar Japan and composition of lone
mothers in the United States (number and percentage of households)

              Estimated number and percentage of households
Survey year
                Total      Widows    Divorced

1952             694,700   590,900      52,400
                   100.0      85.1         7.5
1961           1,029,000   793,000     173,000
                   100.0      77.1        16.8
1973             626,200   387,300     165,100
                   100.0      61.9        26.4
1978             633,700   316,100     240,100
                   100.0      49.9        37.9
1983             718,100   259,300     352,500
                   100.0      36.1        49.1
1988             849,200   252,300     529,100
                   100.0      29.7        62.3
1993             789,900   194,500     507,600
                   100.0      24.6        64.3
1998             954,900   178,800     653,600
                   100.0      18.7        68.4
U.S. 1998      9,828,000   431,000   3,416,000
%                  100.0       4.4        34.8

              Estimated number and percentage of
                         households

Survey year   Never-married     Other

1952                 11,200      40,100
                        1.6         5.8
1961                 20,000      43,000
                        1.9         4.2
1973                 15,300      58,500
                        2.4         9.4
1978                 30,300      47,100
                        4.8         7.4
1983                 38,300      67,900
                        5.3         9.5
1988                 30,400      37,300
                        3.6         4.4
1993                 37,500      33,400
                        4.7         4.2
1998                 69,300      40,200
                        7.6         4.2
U.S. 1998         4,148,000   1,833,000
%                      42.2        18.7

Source: Created from Koseirodosho (Ministry of Health, Labor, and
Welfare), "Zenkoku boshisetai to chosa kekka no gaiyo," (National
survey on lone mother and other households), various years, and U.S.
Census Bureau (various years), "Households and Family
Characteristics: March 1998," Current Population Reports P20-515.

Note: Japanese data on lone mothers include family groups with their
own children under age 20. U.S. data include family groups with their
own children under age 18.

Table 2
Work participation rate of lone mothers in Japan, 1949-1998 (%)

Year   Total   Widows   Otherwise
                        separated

1998   84.9     78.3      86.8
1993   87.0     82.1      89.1
1988   86.8     84.4      87.8
1983   84.2     85.0      83.8
1978   85.2     N/A        N/A
1973   83.9     N/A        N/A
1961   85.6     N/A        N/A
1956   91.9     N/A        N/A
1952   89.8     N/A        N/A
1949   82.2     N/A        N/A

Source: Koseirodosho (Ministry of Health, Labor and Welfare),
"Zenkoku boshi setai to chosa" (National survey of lone mother
and other households), various years.