Abstract Data and method of analysis

1 Introduction

The fertility swings that Sweden has experienced in recent years are stronger than in any other Western country [5]. Over the last fifteen years or so, Swedish Total Fertility rose from what was then a lowest-ever level of about 1.6 (in 1983) to around 2.1 in 1990-92 and then fell swiftly to a new lowest-ever of about 1.5 in 1997 [Figure 1]. The figure for 1998 is even somewhat lower. Tempo shifts may cause the development of the period TFR to exaggerate real fertility trends, but we avoid such problems by working with longitudinal (register) data on the individual level and by holding apart childbearing behavior at the various birth orders. In general terms, it then turns out that the picture has been the same for all the usual birth orders [Figure 2]. In the 1990s, entry into motherhood has been postponed, second-birth rates have dropped by some 25 percent, and third-birth rates have been cut almost in half.

The purpose of this paper is to give a more complete description of the trends in first births and to relate them to developments in the labor market. We hypothesize that childbearing among childless women is dominated by their own economic fortunes and by their perception of general economic developments in their communities. This may be particularly important in a country like Sweden, where women hold a job almost as often as men do, even when they have small children. Under such circumstances, a woman needs to be established in the labor market before she enters motherhood. This means that we may see a strong reduction in first childbearing at times when it is difficult to land a job. We believe that there is some additional effect of public policies directed to the family; in particular the generosity of parental-leave regulations may be important. Economic developments and changes in public policies may influence childbearing both direct and via their impact on people's impressions of what the near future will bring.

Economic trends seem to influence childbearing in several ways. First there is a direct income effect. Since 1990, employment has dropped dramatically, particularly among young entrants into the labor market [Figure 3]. Many young people have found it difficult to get a regular job and have turned to taking more education instead. At the same time, cutbacks in family benefits have hit all families, particularly toward the end of our period of observation. In 1995, parental-leave benefits were cut from 90 to 80 percent of your income, and in 1996 they were reduced further to 75 percent. In 1996, child benefits were cut too for the first time ever, namely by some fifteen percent (to SEK 640 or some US$ 80 per month), and child-care costs have risen in addition. Even though the resulting levels were still generous by international standards, families with children experienced a considerable reduction in their standard of living.

Such cutbacks in public contributions are likely to be regarded as authoritative signals about hard times to come, and they will add to adverse economic trends in creating a pessimistic climate of opinion conducive to the postponement of childbearing, particularly at young ages, where people feel they have time to wait. Young people are also induced to pursue life strategies that do not readily include early family formation and entry into parenthood. When times are bad, the new generations seek education to a much larger extent than their predecessors, because education is seen as a means to become more competitive [12]. This effect may be particularly strong in a country like Sweden, where education is given with no tuition and where students are offered public grants and loans. In response to general developments, the authorities expanded the educational system greatly and there was a massive influx of young Swedes into educational activities as we passed through the 1990s.

In the present report, we concentrate on first births to women born in Sweden and study how they were influenced by the dramatic swings in the labor market. Our concentration on first births is motivated in part by the greater social significance that first births have over second or third births, and in part by our data limitations. Changes in first-birth fertility have dominated general fertility trends, and a decrease in or postponement of entry into motherhood is of prime importance for fertility developments in general. Furthermore, in our analysis we can use the income earned by a woman herself but not the income of the man she has children with. This limitation is bound to be more crucial for a study of childbearing rates after than before a first birth. By the construction of the income-compensation system for parental leave, a woman is strongly motivated to have a firm footing on the labor market herself before she starts having children, and this is what we focus on here. We should be able to allow ourselves to rely on her own income situation as an indicator of economic circumstances before first birth, therefore, not least because of the extent of assortative mating even in Sweden. After a first birth, many women reduce their working hours, and it is not unusual that they spend some time without a job or on leave without pay. In such a case, the man's income could be the main determinant of their economic situation. The lack of information about his income is a much greater handicap for a fertility analysis at this stage. Nevertheless, we have made some analysis of second and third births and have reported it elsewhere along with some information about births to immigrant women [4]. Third-birth rates to Swedish women turn out to be at least as sensitive to economic fluctuations as their first-birth rates are, but second-birth rates are much less easily affected. Immigrant women have been less influenced by economic trends than Swedish women have.

To improve our understanding of the determinants of entry into parenthood, we have constructed a special longitudinal data set based on registers available in Statistics Sweden. We now turn to a description of this data set.


Abstract Data and method of analysis

logo70.gif (2450 bytes)

Entry into motherhood in Sweden:
the influence of economic factors on the rise and fall in fertility, 1986-1997

Britta Hoem
© 2000 Max-Planck-Gesellschaft ISSN 1435-9871