Social Science Research Council

 

CGP-SSRC SEMINAR SERIES

Fertility Decline, Women’s Choices in the Life Course, and Balancing Work and Family Life: Japan, the USA, and other OECD Countries
May 16-17, 2006
Conference Report

 

Naeyun Lee

This document reports on the tenth seminar held under the auspices of the Japan Foundation Center for Global Partnership (CGP) and the Social Science Research Council (SSRC). The seminar series was established in 1996 to assemble interdisciplinary groups of researchers and professionals to examine global and bilateral (US-Japan) issues. The workshop described herein, entitled “Fertility Decline, Women's Choices in the Life Course, and Balancing Work and Family Life: Japan, the USA, and other OECD Countries” was held at The Gleacher Center, The University of Chicago, May 16-17, 2006. Convened by Kazuo Yamaguchi of The University of Chicago, the meeting brought together 12 experts from Japan, the United Kingdom, and the United States. A major theme of the project of which this workshop was a part is to investigate the various causes of the fertility decline in modern society in relation to women's preference, the flexibility of work, and the compatibility of work and family life. Other topics include comparisons among the OECD countries in fertility and women's labor force participation and the growing economic inequality among households with children within marriage and after divorce. Naeyun Lee of The University of Chicago prepared this summary report of the discussion.

The SSRC administers the seminar series with funding provided by the Center for Global Partnership. The views expressed in this publication are summaries of the comments by participants and should not be considered a representation of the views of either the CGP or the SSRC.

Introduction

In the recent decades, extremely low fertility rates below the 2.1 replacement level have emerged as one of the most pressing issues in industrialized nations. In most OECD countries with rapidly aging populations, declining fertility is likely to lead to shrinkage of the working population, which will jeopardize the social security system and hinder economic growth. Thus, the focus of population policies has shifted from controlling fertility to raising fertility through introducing various family-friendly policies. Yet, it is difficult to explicate the real causes of fertility decline and assess the effects of current public policies.

In Japan, the decline of the total fertility rate began in the late 1980s. Scholars have mainly attributed the fertility decline to three factors: the delay in marriage, an increase in never-married people, and a decrease in childbirths among married couples. Because of the very low rate of out-of-wedlock births, marriage and fertility are closely related in Japan. Recently, more attention is being paid to the drastic fall in marital fertility. While the Japanese government has introduced measures to enable married women to better balance work and family life, the fertility rate has remained largely unchanged.

Therefore, this conference aims to locate the causes of the fertility decline and provide suggestions to help restore fertility rates to the replacement level or above. Participants strived to elucidate why the fertility rate persists to fall in some countries while other countries experience a gradual recovery in fertility. In order to better comprehend this discrepancy, they looked at women's varying preferences, labor market characteristics, balance between work and family life, and family characteristics.

Heterogeneity of Women's Preferences and Homecare Allowance Program

Catherine Hakim of London School of Economics opened the first session by proposing a new theoretical framework for understanding fertility behavior. Preference theory argues that a set of social and economic changes have created a new scenario for women, giving them genuine choices as to how they live. In this new scenario, women are revealed as heterogeneous in their life choices. On the basis of recent research evidence, she identified three groups of women with distinct value systems and lifestyle preferences: work-centered, adaptive, and home-centered. Hakim contended that this heterogeneity in women's preferences needs to be taken into account in designing family policies because lifestyle preferences better predict women's employment and fertility than 'social structural' variables such as education. Moreover, different groups of women respond differently to policies. The family-friendly policies, such as maternity leaves, have not been very successful in raising fertility, or in changing women’s return to work after childbirth, because they only focus on stereotypically standard, mostly careerist, women without considering the varying needs of women. In Hakim's view, economic incentives have proven to be most effective in raising fertility. In particular, the Finnish homecare allowance and French equivalents had serious success by supporting motherhood and parenting as central goals of life. Homecare allowance programs are vehemently criticized by orthodox feminist scholars for promoting sex-role differentiation and 'gender inequality.' However, Hakim maintained that they fail to accept the heterogeneity of women's and men's preferences and that family policies that target home-centered women should be introduced to counterbalance the current policy emphasis on working women.

Eiko Kenjoh of Asia University introduced the labor economics perspective on Hakim’s preference theory. Individuals, under budget and time constraints, endeavor to maximize their utility and in doing so determine their labor supply and fertility. She therefore concluded that women’s response to public policies could differ between individuals, even when they have the same preferences, because their constraints and, thus, economic incentives are different. Kenjoh asked why educational qualifications could have little explanatory power for women's employment and fertility behavior. She also pointed out that recently more and more women prefer the combination of work and family rather than being home-centered. Kenjoh further argued that not only homecare allowance, targeting mainly home-centered women as Hakim suggested, but also maternity and parental leave, depending on the structure of the leave and other measures of public policies related to work-life balance, could be effective to combat declining fertility. More importantly, she said, we need to first address what kind of society we want to create. Are we looking for comparative specialization in women by assigning full-time careers to work-centered women while entrusting home-centered women with reproduction?

Sawako Shirahase of University of Tokyo agreed with Hakim that women's preferences differ and that there cannot be one-size-fits-all policies for heterogeneous groups of women. However, she emphasized that people do not have the same preferences under the same circumstances. This is because preferences are shaped by social norms and expectations. She also argued that the total fertility rate cannot be used as a criterion for evaluating family-related policies because it is unrealistic to expect the total fertility rate to go up immediately. Concurring with Kenjoh's comment, Shirahase said we should instead focus on debating what kind of society we seek to build. Lastly, she was concerned with the high costs of providing different sets of policy menus to accommodate the varying preferences of men and women.

Hakim responded by stating that some of these policies are high-cost while others actually cost less. In the case of East Germany, the state-subsidized childcare was proven to be more expensive than the homecare allowance. She strongly opposed the economists' assumption that women who stay at home are less educated. She said that there was no IQ difference throughout the three preference groups. Many women, according to Hakim, go to college without the intention of pursuing a career, because college and university function as a marriage market as well as a place for training. Thus, college education is not a reliable indicator for career orientation. Moreover, some occupations, such as nurses, clerks, and nursery school teachers, do not necessarily require high education but tend to attract highly devoted group of people. Hence, career orientation is not necessarily related to high pay and education.

In the ensuing discussion, participants spoke about useful proxies for measuring the lifestyle preferences. Hakim explained that common questions on 'work ethic' and ideal family type were proven to be effective. Yamaguchi warned that since preferences are normative designs, they may be different from their actual behaviors. He also asserted that maternity leaves may influence fertility in countries with rigid labor markets where it is hard to leave work due to high opportunity costs. Hakim agreed that childcare leaves may be less important in societies where the labor market is sufficiently free, allowing women to return to work easily. Ono questioned whether preference theory is a classist argument since less privileged groups often have no choice but to work. Hakim replied that even within different minority groups three distinct preference groups emerged and that economically disadvantaged solo mothers often replied their priority is to be a full-time mother no matter how poor they are.

Work-Life Balance, Flexible Employment, and Fertility

Machiko Osawa of Japan Women's University suggested that the institutional structure of the labor market must be taken into account in our analysis on the fertility decline. She noted that while some countries experience a positive correlation between women's labor force participation and fertility rate, others face an inverse correlation between the two. She ascribed this difference to the way in which flexibility is introduced into the labor market. Osawa observed that countries with relatively high levels of fertility either have fewer temporary workers like the United States and the United Kingdom or have more regular part-time employment as in the Netherlands and Denmark. In contrast, she said, countries with a high proportion of temporary fixed contract workers, such as Spain, South Korea, and Portugal have low fertility. Thus, she concluded that countries with a dual labor market suffer most from the drastic decline in fertility.

Osawa admitted that Japan may seem to be an exception since it has a low fertility rate even with a high proportion of part-timers and a small proportion of temporary fixed contract workers. However, she emphasized that Japanese part-timers are, in practice, treated like temporary workers with no job security and low pay. Therefore, she maintained that such non-standard work arrangements in Japan depress fertility by not providing job protection for part-time workers and not adopting the principle of equal pay for equal work. Women are disproportionately represented in the non-standard work force, serving as the shock absorbers for the Japanese economy. This is partly because the government policies encourage them to seek part-time work, and also because such jobs are the only option for those who need flexible work schedules. She drew attention to the demand side of the picture by highlighting that flexible employment was implemented mainly to compensate for the high costs and rigidities of the existing industrial relations system. Since flexible employment is not introduced in a family-friendly manner, it contributes to lowering fertility rather than raising it. She argued that further decline in fertility is expected unless more equitable distribution of the fruits of flexibility enables working women to balance the demands of careers and families.

Hiromi Ono of Washington State University remarked that while it may be true that such work-family conflicts influence women's desire to have children, it is important to note that employment and fertility are both mediated by marriage in Japan. She pointed out that there may be an inherent conflict between being a wife and working in Japan that does not necessarily involve being a mother. She argued that rigid perceptions about wives' roles may deter women from working full-time. Ono also questioned whether we could actually impact the fertility behaviors by simply changing policies. She emphasized the role of inherent rules and practices in society that govern people's behavior. Lastly, she suggested that factors other than work flexibility, such as cooperation from co-resident family members, may have greater effect on fertility than work flexibility itself.

Kenjoh shared Osawa’s view on the importance of the institutional structure of the labor market on fertility behavior. Kenjoh, however, questioned Osawa’s analysis of the decomposition between demand and supply factors of the increase in non-standard work. While Osawa interpreted the part owing to changes in the proportion of non-regular workers within groups as the part caused by demand factors, Kenjoh warned that such changes could be driven by not only demand factors but also behavioral changes on the supply side. Osawa, however, contended that the number of temporary workers has grown while the young population has decreased in size. Kenjoh also found it problematic to conclude that full-time regular workers had more children than housewives or non-regular workers because the sample is only based on married women. She raised the possibility of selection into marriage since a number of female regular workers may have remained single and had no children.

Osawa replied to Ono's comment by stating that industrial relations are closely related to gender relations. She illuminated that the Japanese welfare system is heavily dependent on family for taking care of children and the elderly and that an implicit contract existed in society to induce women's full-time commitment to family. However, as marriage becomes unstable and less people marry, this model collapses in many directions. The introduction of flexible work practices has eroded the economic bases of Japanese families which largely depend on job security. Thus, she argued that it is necessary to increase job security by sharing the work, as is the practice in the Netherlands. She acknowledged that it is important to change the mindset of people first to value the balance between work and family.

In discussion, Hakim pointed out how the Japanese government provides tax benefits for full-time homemakers and job protection for careerist women, while leaving out the middle group who are the majority. Osawa confirmed this observation by stating that the Japanese society penalizes those who seek the middle ground.

Labor Force Transition of Married Women in Japan

In the second session, James Raymo of University of Wisconsin, Madison presented the findings of a paper co-authored by So-jung Lim. Their paper drew attention to how the likelihood of married women's labor force exit and entry vary by level of education in Japan. The analysis of panel data showed that university graduates were less likely to both exit and re-enter the labor force. Raymo argued that this result demonstrates the bifurcation of highly educated women into career-oriented group and home-centered group. He hypothesized that the low likelihood of re-entry for women with college education may be due to their human capital characteristics and their husband's relatively high income. He concluded that factors other than human capital, husband’s resources, and career orientation contribute to educational differences in labor force trajectories.

Yamaguchi remarked that there are some contingencies on the effects of high education for women. Strong family orientation, traditional gender roles, gender inequality in the labor market, and the nature of college education as a cultural capital for women may all offset the influence of college education on women's labor force participation. He noted that Japanese women still have strong family orientation although there is a significant change toward non-traditional gender roles and less gender discrimination in the workplace due to legal measures. He suggested that the lower likelihood of exit for college-educated women may be due to an increase in gender equality among the highly educated or a decrease in economic stability of highly educated men. Another possibility, he said, is that the weight of college education as economic capital has increased among women. He suggested various interaction effects must be taken into account as well.

Osawa raised the point that it is important to note that a variety of policies were implemented to help women, especially highly educated women, return to work. Among those with college education who continued to work, she said, more women were able to benefit from the childcare leave that replaces 40% of the income. She observed that education is an instrument to gain more opportunities and to control one's lives, which was confirmed by Raymo and Lim's finding. She criticized that current policies promote inequality further by concentrating on those with resources rather than assisting those who need most help. While agreeing with her point, Raymo cautioned that their focus was not on assessing social policies, but on describing differences in labor force attachment likely to impact the economic well-being of women in the high-divorce era.

Support by Extended Family, Husband’s Participation in Household Work, and Intention to Have Children

Akihide Inaba of Tokyo Metropolitan University investigated the factors that affect married women's intention to have another child. He observed that there were no significant effects of wives' educational years and husbands' participation in household work and in childrearing. He argued that this suggests that opportunity costs are not associated with education. He hypothesized that husbands` participation in such tasks does not have much effect, possibly because household work and childrearing are shared within the broader kin network. His analysis demonstrated that both traditional views of children and the availability of support by extended family had an impact on the intention to have a second child, but not for the third child. Thus, Inaba contended that we may inhibit the fertility decline by facilitating a close residence of parents with married couple. This will allow wives to reduce their participation in household work and childrearing, equalizing the gender relationship without having to change the husband's participation. However, he noted that this change may result in an intergenerational inequality. Despite such problematic aspects, he proposed this kin network model as an alternative to the dominant nuclear family model.

Shirahase questioned the peculiarity in the importance of kinship support in Asia. So-jung Lim of University of Wisconsin, Madison cautioned about making such generalizations since Asian societies have diversified greatly over time. Inaba responded that it may also be significant in other societies as well, depending on the gender division of labor and the extent of extended family. Shirahase wondered if husband's participation in housework and childrearing was measured by the relative share of these activities between the couple. Moreover, she remarked that the level of kinship support must be further differentiated when discussing its availability. Inaba responded that his analysis only focused on temporary, not daily, support. Shirahase highlighted that what we measure is the availability of support perceived by the respondent. Miyuki Vamadevan of Washington State University added that kinship help is possible only when parents are both not in the labor force and willing to help. Lastly, Shirahase thought that encouraging kinship support on the level of social policy was inappropriate. Concurring with this point, Raymo and Hakim suggested that homecare allowance could be an alternative policy.

Yamaguchi pointed out that birth intention is a combination of preference and the result of assessing the costs and benefits of alternative choices. Thus, he wondered how we could measure preference separately. He suggested that marital satisfaction must be taken into account in measuring family-role equity which includes husbands' economic performance. He said that co-residence may not be an alternative for career orientation if the wife was not concerned with gender equity. He thought that wives' individual income may have been a better measure of opportunity costs. Inaba replied that this was not possible because of the large number of home-makers whose personal income would be 0. Lastly, Yamaguchi raised the possibility of some selection bias since women would move from full-time work to part-time work after childbirth

Preferences for Wives' Employment Hours in Japan

Larry Bumpass of University of Wisconsin, Madison was concerned that Japanese wives' role overload that may affect fertility. His findings indicated that Japanese wives and husbands have strong preferences for wives' employment. However, he said, the persistent traditional gender role expectations already impose heavy domestic obligations for wives. Thus, strong preferences for wives to work long hours create major challenges in balancing work and family life. He noted that such preferences for wives' employment vary by family factors and resources. He suggested that the anticipation of such balancing formed during years spent unmarried may significantly affect one's behavior later. Especially for women, this may mean putting off marrying and having children as a way to avoid such struggle. Bumpass predicted that men's demand for wives' employment will further increase with the aging of population.

In the ensuing discussion, Inaba asked how we should assess or interpret these preferences. Yamaguchi added that preference for work is a murky indicator in that it may either show wives' desire to seek social rewards of work itself or income needs. He asked if preferences for certain hours of work could be associated with other things such as benefits from the regular worker status. He contended that it is important to include the distinction between regular and temporary worker status because of the rigidity of the market. Inaba suggested that the high rate of marginal employment for women who live with parents may involve intergenerational division of labor and conflicts between the mother-in-law and the wife. Hakim found that Bumpass' finding confirmed the polarization of three patterns: non-working wives, dual-earner couples, and middle ground families. She thought that, due to the rapid transformation from agricultural to non-agricultural economy, work overload may be taken for granted since the workload was greater in the past.

Comparison among the OECD countries

Kenjoh opened the third session by presenting her cross-national analysis on women's employment behavior after birth of the first child in connection with family policy and part-time employment. She observed that employment patterns of first-time mothers in the 1980s and 1990s were in line with the actual family policies in each country. She argued that family-friendly public policies, that design flexible parental leave with sufficient income compensation and provide affordable good-quality childcare, could effectively increase labor force participation of new mothers. She also argued that the availability and stimulation of part-time employment could provide additional strong incentives for new mothers to enter or regain active employment. In the case of Japan, she noted that the probability of being in part-time employment was low, particularly for highly educated mothers. She argued that this is attributable to the lack of good-quality part-time work together with the limited availability of day care for children under age 3 and the insufficiently generous parental leave system.

Raymo questioned the appropriateness of proxying public policies by country and time. He argued that country and time may be indicators for other factors such as normative environment or husband's participation in housework. Thus, he said, it would be desirable to develop comparable policy measures over time. Moreover, he thought that focusing on the actual transition of women in and out of the labor force would yield more informative results. Kenjoh responded that only a small proportion of women were actually making multiple in-and-out transitions shortly after first childbirth in the observation periods. Raymo also wondered if the educational differentials in the labor force participation could be proxies for human capital characteristics such as opportunity costs and attitudes. Kenjoh argued that higher education is associated with higher wages, implying higher opportunity costs of not working in paid labor as well as higher affordability of childcare in the situation without public subsidies. Lastly, Raymo emphasized that it is important to consider that there are both careerist and home-centered women within the group of highly educated women.

Hakim pointed out that Kenjoh's analysis has failed to include a substantial proportion of women who chose not to have children and carry on full-time careers. She addressed the methodological problem of relying on women's self-report which may not always give accurate, objective accounts and cautioned against the difficulty of assessing the effect of public policies. She found it interesting that the percentage of women who go (back) to work full-time immediately after first childbirth are consistent at the 10 to 20% level in each of the 5 countries.

In the second part of the third session, Yamaguchi presented findings that showed that the weakening of negative relationship between women’s labor force participation and fertility is largely due to the promotion of work-family balance. He observed that while women's labor force participation was associated with a decline in fertility in the past, the relationship has recently reversed among some of the European countries. He argued that the negative relationship between female labor force participation and the total fertility rate was reduced by the interaction effect of the flexibility of work and the indirect effect of the compatibility between work and childrearing. He urged that societies must facilitate an environment with a greater flexibility of working which includes high-quality part-time employment, flexible work places and a system of re-employment in firms for those who leave work for childrearing.

Raymo maintained that Yamaguchi's analysis provided an empirical support for the effect of work flexibility. He questioned the role of work-family balance for marital fertility rate. Bumpass remarked that it influences fertility through its relationship to marriage. Hakim opposed this view by stating that work-family balance policies focus on promoting the combination between parenthood and work life, which may not necessarily be related to marriage. Raymo pointed out that in some countries, such as Japan, marriage may be an important factor. He said that a lack of work-family balance may suppress fertility, but raise labor force participation since women may not choose to marry if they believe that it is likely to be difficult to be both a mother and continue work. He also remarked that the reactions to changes in policies may have been discrete rather than linear and encouraged the consideration of alterative specifications of time in the models. He wondered why Japan is not faring so badly in OECD measures of fertility despite the fact it is hard for women to return to work after childbirth.

Ono emphasized that the measures are "contextual effects" from a multilevel perspective. Thus, she argued that it is important to consider what happens at the individual level. She also warned that the causal relationships between independent variables may run in other directions. Lastly, she asked if the measurement distinction between compatibility and flexibility at work was substantively meaningful.

Concurring with Raymo's point on marital fertility rates, Yamaguchi admitted that he should reinclude marital rates in his analysis because implications for becoming a mother and a wife are different. On Ono's remark, he agreed that the causal link may be reversed and that it is important to return to the panel data for micro level analysis.

In the ensuing discussion, Bumpass suggested fertility in different contexts such as single-parent family or non-marital union, must be taken into account since they require different approaches to solve work-family conflicts. Osawa raised the point that we need to look at the labor market and the organization of industrial relations in order to understand how flexibility at work is introduced. She argued that these institutional contexts shape the traditional division of labor and family model that are the bases of each society. She emphasized that unless the whole attitude toward work itself changes, work and family balance is hard to achieve.

Gender, Children, and Inequality

In the last session, Shirahase drew attention to the way in which socio-economic inequality is associated with individual behaviors regarding marriage and childbearing. She observed that both young men and women in low-income households with low level of education were increasingly likely to remain single. Single rate was high for highly educated females but not for males. She argued that it is crucial to investigate how men and women in the low socio-economic status are disadvantaged not only in the labor market but also in the marriage market. She noted that married women still drop out of the labor force in Japan, and later return as secondary earners to supplement the family income. She found that when wives' economic contributions to the household income are high, they tend to have a fewer number of children. Thus, in her view, it is critical to find ways to assist women so they can find better jobs and stay in the labor force in order to lower the opportunity costs of marriage and childbirth. Lastly, she was concerned that the income inequality has been aggravated by the aging of population. Recently, she said, poverty has spread to young nuclear families with small children, especially single mother families.

Bumpass pointed out that the rise in single rates may indicate that the search for a partner has simply become longer due to the economic uncertainties that force people to delay marriage. He noted that, with the increase in contract works, it has become more common to find unstable jobs. Shirahase agreed that the increase in single rate has to be viewed from a broader social context. Bumpass further suggested that since co-residence has a significant impact on mother's employment, it has to be included in the analysis. Shirahase replied that her data was constrained to nuclear families in order to conduct comparative analyses with other countries. Bumpass also remarked that the sample was limited to those who are employed among married women.

Inaba raised the possibility of selection bias resulting from low response rates of unmarried people. He argued that this may seriously affect the composition of single people in the sample Shirahase admitted this problem and said that she will weight the data set in the future.

In the following discussion, Yamaguchi contended that both the high single rate of poor men and the rise of poverty rate among young nuclear families are linked to the fall in fertility. He argued that the former indicates the increase in the number of poor men who are not marketable in the marriage market. Thus, it is not only women's choice to remain single, but it is harder to find eligible men. The rise in poverty rate implies that poor married men refrain from having children, further contributing to the fertility decline. Raymo added that, in his own findings, the transition to marriage was found to be associated with the economic prospects of the husband. He also warned that using junior high school graduates as a reference category was inappropriate not only because they are a tiny portion of the population but also because they lead rather different lives. Osawa asked what the driving force is behind this increase in economic inequality for young people. She maintained that it was the change in the labor market structure that produced an increased number of young people who have irregular employments.

As the last presenter of the conference, Ono reported the findings from her paper co-authored by Miyuki Vamadevan. She noted that, despite the rapid increase in divorce rates in Japan, not much attention was paid to the gendered economic consequences of divorce. She hypothesized that this may be because Japanese parents step in to mitigate the problem. She observed that divorced men have substantially higher levels of labor income and household income compared to divorced women in Japan, whether or not co-resident children are present. This socioeconomic gender gap among the divorced was greater in Japan than in the United States. She argued that one of the reasons for this gender gap in household income is that divorced Japanese men are much more likely to co-reside with parents than women, regardless of the presence of co-resident children. In Japan, married couples tend to live with husbands' parents rather than wives' parents. Thus, she suggested that Japanese women are forced to exit the husbands' parental household upon a divorce, leaving behind any capital accumulated while married. Lastly, she pointed out the presence of children does not seem to directly disadvantage divorced Japanese women's earnings, household income, and the likelihood of living with parents, compared with those of men. She thought that this may be due to the fact that women are categorically disadvantaged in the rigid labor market regardless of whether they have children or not. In the final analysis, she concluded that while gendered economic inequality among the divorced exist, families are not stepping in to help out, which may result in major negative economic consequences for divorced women in Japan.

Bumpass raised the possibility that the increase in divorce may be concentrated among those with lower level of income in Japan. He warned that co-residence with parents is very rare in the United States which may impose a great constraint on the data. He remarked that 40% of all divorces in Japan do not involve children, which suggests that many married couples put off having children in the prospect of an anticipated divorce. He pointed out that it is crucial to take into account that those who are divorced without children disappear from the sample because they are back on the marriage market as "singles." This may make the economic well-being of divorced persons with children seem relatively worse off.

Osawa thought it would be more informative to include the demographic characteristics of people who are divorced with or without children in comparison with the general population. She contended that Ono's analysis demonstrates that divorced mothers cannot get help from their parents, calling for a drastic change in Japanese social policies that rely heavily on the traditional kinship system. She illuminated on the current social welfare payment system that excludes people who can depend on family.

In discussion, Raymo highlighted the fact that remarriage is the common consequence of divorce. In the United States, he said, remarriage offsets the economic differentials generated by divorce although it does not improve the well-being of children. Yamaguchi added that gender gap will be greater if remarriage is considered in the analysis. He remarked that divorced men with high income are more likely to remarry while divorced women with high income tend to remain single. Ono responded by saying that marital data is no longer collected in the survey since 2003 in Japan. Bumpass argued that remarriage may serve as a measure for the outcome of children in various aspects as well.

Conclusion

The main goal of the conference was to incorporate the wide-ranging views on fertility decline to reach a more complete understanding of the phenomenon. Participants first struggled to locate the driving force behind the fertility decline in industrialized societies. Some participants maintained that different groups have inherently different lifestyle preferences that invariably affect their fertility behaviors. Others drew attention to the structural features of the labor market such as the industrial relations system, flexible work arrangements, and compatibility of work and family life. Participants also focused on the role of social norms regarding gender roles, family factors, and socioeconomic resources in depressing fertility.

One of the major areas of contention was in assessing the effect of family-friendly policies. Some participants were skeptical about the efficacy of such policies since they only focus on a specific group of work-centered women. Moreover, it was pointed out that changes in policies do not necessarily induce behavioral changes in people unless it is accompanied by a shift in social norms and attitudes. Others highlighted that family-friendly policies have proven to be effective in raising fertility, especially in societies with rigid labor markets.

Another area of debate was in policy suggestions for improving the fertility rate. While participants agreed on the seriousness of fertility decline, they had different approaches to solve the problem. Some participants asserted that economic incentives must be introduced to accommodate the heterogeneous needs of women in childcare. Others argued that high-quality part-time employment should be readily available in order to assist both women and men with balancing work and family life. Securing kinship support through co-residence with parents was also suggested as a way to mitigate the burden on women.

Lastly, participants projected the future of Japanese society by discussing how fertility decline is interrelated with the growing inequality in society. Current family-friendly policies in Japan were criticized for only benefiting those who need least help and transferring the burden to families. Thus, under the current system, low-income families, including single mother families, are struck hardest due to lack of kinship support and resources to afford childcare.

The second part of the conference is scheduled for Tokyo, Japan in May 2006.