Will Prime Minister Abe's Growth Strategy Work? Expanding nursery school capacity should be the top priority for utilizing the untapped potential of the female workforce

Fomer Faculty Fellow, RIETI
Associate Professor, Institute of Economic Research, Hitotsubashi University

As a centerpiece of his growth strategy, Prime Minister Shinzo Abe emphasizes the idea of "encouraging women to play a more active role in society." The first step to achieve this goal is a plan called "Zero Children on the Waiting Lists," designed to reduce dramatically the number of children on nursery school waiting lists by expanding their capacity. Based on my research conducted at RIETI, this article discusses the effectiveness of this plan.

Presenting my conclusion first, Prime Minister Abe's decision to put priority on expanding the capacity of nursery schools as part of his growth strategy deserves high recognition. Increasing the capacity of nursery schools is an effective means of enabling women to enjoy better "compatibility" between work and marriage/childbirth. However, focusing only on reducing the number of children on nursery school waiting lists to zero may lead to the distortion of relevant policies. In order to develop such policies into specific action and conduct follow-up evaluations, it is necessary to set more appropriate goals.

♦  ♦  ♦

Compatibility of marriage/childbirth with work is one of the most important indicators of the situation of women in Japan, a country with a rapidly aging population. Utilizing the untapped potential of women is essential to secure a sufficient labor force while encouraging more women to marry and have children is necessary to address the low birth rate. To achieve these two goals simultaneously, it is vitally important to enhance compatibility between work and marriage/childbirth.

So far, however, such compatibility has been somewhat sidelined from the center of policy discussions. On the one hand, the debates over women and work have centered primarily on correcting the inequality between gender in job opportunities and salaries. Countermeasures for the low birth rate, on the other hand, have aimed to address infertility and the lack of opportunities for women and men to meet the right partners. In either context, compatibility between work and marriage/childbirth has not drawn attention as a major policy issue. That is the very reason why the recently announced plan has been welcomed in that it regards compatibility as a central issue in the context of the growth strategy.

Another factor that has prevented such compatibility from becoming a primary issue in policy discussions is the technical challenge of devising an accurate means of measuring compatibility. As a result, it has been impossible to grasp accurately the status of compatibility, making it difficult to identify the need for policies designed to address this particular challenge. Measuring compatibility requires knowledge about the type and form of work engaged by women before and after marriage/childbirth and the follow-up survey using panel data on the situations of the same individuals over time. Japan, however, lags behind in terms of creating panel data.

To cope with this problem, I conducted an analysis by birth-year cohort using population census data. Comprising a group of individuals who were born in the same year, the birth-year cohort constitutes a "generation." The statuses of marriage and labor force participation of each generation can be identified from data by age group at each relevant point in time. Because individuals constituting a generation are identical at each point in time, birth-year cohort data can be regarded as quasi-panel data sets.

For example, women born in 1980 were 25 years old in 2005 and turned 30 years old in 2010. The percentage of unmarried women at each point in time was 74% and 41%, respectively. From the difference between the two figures, there was a decrease of 33 percentage points, indicating the number of women of this generation who got married during the 2005-2010 period. Similarly, the labor force participation rate of women of this generation dropped from 77% to 67%, a decrease of 10 percentage points, indicating those who exited the labor market.

If the percentages of women who got married and that of women who exited the labor market can be obtained, the "percentage of women who exited the labor market after marriage" can be estimated from these two figures. In Japan, 85% of married women have children within five years after getting married, and very few women have children without getting married. Therefore, it is statistically possible to regard the percentage of women who "got married" as those who "got married and had children." In other words, what is being measured here is compatibility of marriage/childbirth with work.

In the above example, 30% of women who got married left the labor market after marriage. Putting this in another way, 70% of women continued working (attained "compatibility"). The percentages of women of each generation who continued working after marriage are regarded as representing compatibility at each point in time.

♦  ♦  ♦

The trend of compatibility from 1980 to 2005 as represented by birth-year cohort analysis using population census data reveals some unexpected facts. Up until 2005, the degree of compatibility had remained almost unchanged. In fact, no improvement at all has been shown until recently. While the labor force participation rate of women increased substantially during this period, it was brought about not by improved compatibility but by a rise in the percentage of unmarried women.

Meanwhile, compatibility improved rapidly over the five years from 2005 to 2010 (See Fig. 1). This change, however, has not been widely recognized. Another trend that is virtually unthinkable from the range of change over the past 25 years is the dramatic increase in the percentage of women who continue working after marriage. This trend can also be confirmed from the Longitudinal Survey of Babies in the 21st Century conducted by the Ministry of Health, Labour and Welfare, which comprises individual-based panel data that have been available for public use since 2000. According to this longitudinal survey, while the percentage of mothers who quit jobs during the perinatal period was 67.4% for babies born in 2001, it dropped to 54.1% for babies born in 2010.

Figure 1: Increase in the capacity of nursery schools and the percentage of women who continue working after marriage/childbirthFigure 1: Increase in the capacity of nursery schools and the percentage of women 	who continue working after marriage/childbirth

The birth-year cohort approach is also applicable to data by prefecture and age group. Calculation of the percentage of women who continue working after marriage/childbirth by region revealed that figures varied greatly. Lower compatibility was found in such metropolitan areas as Tokyo and Osaka, while it was higher in prefectures along the Sea of Japan coast such as Yamagata, Toyama, Ishikawa, and Fukui. Meanwhile, all prefectures shared the same trend with respect to compatibility—it remained nearly unchanged until 2005 and improved by 2010.

Once compatibility is measured, it is possible to speculate on its determinants. The key is statistical properties such as longitudinal changes and regional differences, because if there is a factor that affects compatibility, it should have the same statistical properties.

By analyzing these statistical properties in reverse order, the capacity of nursery schools is identified as the largest and almost single factor that determines compatibility of marriage/childbirth with work. The capacity of nursery schools is measured by what I call the "potential nursery school capacity ratio," which represents the capacity of nursery schools compared to the population of women aged 20-44, and this remained unchanged until 2005 and dramatically improved after that. There are also regional differences in this ratio—it is higher in the regions along the Sea of Japan coast and lower in bigger cities. In other words, this ratio has the same statistical property as compatibility.

Meanwhile, the significance of most of the factors that have thus far been believed to affect compatibility can be discounted. For instance, the childcare leave system introduced in 1992 has rapidly taken hold. The prevalence of three-generation families has consistently been on the decline. If these are determining factors, there must have been a longitudinal change in compatibility, contradicting the results of my data analysis. More generally, judging from the fact that compatibility remained unchanged during the period from 1980 to 2005, policies that were implemented or societal changes that occurred during the period were not major determining factors.

♦  ♦  ♦

The observation that nursery school capacity had not increased by 2005 might sound implausible. This sense of incongruity may come from the fact that the capacity of nursery schools, which is usually measured by the nursery school capacity ratio (the ratio of nursery school capacity compared to the population of children aged 0-6), has expanded steadily over the past 30 years. In fact, the increase in the ratio resulted from a decrease in the number of children aged 0-6, who comprise the denominator, instead of an increase in nursery school capacity. In other words, the nursery school capacity ratio increased not because of successful countermeasures against the low birth rate, but as a result of the low birth rate itself.

The same problem can be seen in the number of children waiting to enroll in a nursery school. If the availability of nursery schools is so limited that many parents give up enrolling their children, the number of children on waiting lists will not increase substantially. On the other hand, as many municipalities have already realized, the larger the capacity of nursery schools becomes, the more there are of children who will be added to their waiting lists, because the sign of availability encourages more parents to apply for enrollment for their children. So "zero children on the waiting lists" could mean either a "good zero" situation, where anyone wishing to enroll can be admitted to a nursery school, or a "bad zero" situation, where nursery school capacity is so limited that no one can even think of applying for enrollment. And it has nothing to do with whether nursery schools are actually in short supply. In this sense, seeking to achieve "zero children on the waiting lists" as a policy target may hinder appropriate resource allocation.

To prevent this kind of development, I measured nursery school capacity by the potential nursery school capacity ratio, which uses the population of women aged 20-44 as the denominator. By determining the nursery school capacity for each woman who can potentially become a mother, I circumvented the effect of women who, due to the shortage of nursery schools, gave up applying for enrollment for their children or women who decided against getting married at all. Although this indicator may not provide a policy target that is as clear-cut as the number of children on waiting lists, it is an appropriate tool for measuring the situation more objectively.

Just recently, the city of Yokohama was reported to have succeeded in reducing the number of children waiting to enroll in nursery schools to zero. It is hoped that the national government actively incorporates into its growth strategy the steps taken by Yokohama. What is worth even more recognition, however, is Yokohama's sincere attitude toward the improvement of the capacity of nursery schools. It is hoped that the Japanese government will develop appropriate indicators and proceed with policy implementation, with the awareness that "zero children on the waiting lists" per se should not be the ultimate goal.

>> Original text in Japanese

* Translated by RIETI.

June 11, 2013 Nihon Keizai Shimbun

July 29, 2013