The individual woman's labor-market situation Acknowledgements

8 The effect of public policies

As we have seen, the natality of childless women is influenced both by their own individual circumstances and by economic developments in their local community. They may also be affected by changes in the level of public benefits. Families surely see such changes as signals about authorities' assessments concerning the country's economic prospects, and these signals are likely to amplify the direct consequences of purely economic factors because they affect the expectations that families have about their own immediate future. Such expectations can be gauged by data from a monthly survey of household purchase plans (Hushållens inköpsplaner) run by Statistics Sweden for many years [15]. Among the information collected in this survey is how respondents expect their own household's economic situation to develop over the coming twelve months. In [Figure 11] we give a picture of some of the survey outcome between early 1985 and mid-1998. We see that the respondent households were quite optimistic during the second half of the 1980s, precisely when there was a strong increase in fertility. Expectations changed in 1990, after which we got a period of pessimism during which Swedish natality dropped in the manner described above. Positive signals about a pick-up in the economy produced some brief optimism during 1994, but this was reversed in parallel with cuts in family benefits in 1995 and 1996, and it took until the end of 1997 for optimistic attitudes to return. This late upturn came too late for us to see any effect in our childbearing data.

Our inclusion above of the individual woman's income and activity situation and the employment trend in her home municipality allowed us to "explain" most of the period effect except the rise in first births during the late 1980s and a residual decline after the mid-1990s. It is possible that the remaining period effects are related to changes in the generosity of Swedish family benefits, for they occur in years where such benefits were changed.

To produce an index for the development of family benefits is not straightforward, for some parts of the system are income dependent and others are at a flat rate, and they change often. To see some pattern nevertheless, we have focused on a woman who worked full time all year at the mean income for adult women, and who did not have any unearned income. We have multiplied her earned income (at fixed prices) by the benefit level each year, i.e. by 90 percent in 1985-1994, by 80 percent in 1995 and by 75 percent in 1996. This gives the amount that she would get in maternity benefit (at fixed prices). To this amount we have added the child allowance paid in the same year (also at fixed prices). The sum would give the major part of her cash benefits during the first year after entry into motherhood if she took out all her maternity leave continuously for the statutory number of months just after her first birth. (The system actually allows her much greater flexibility than this [5].) We have then computed an index for the years 1985 through 1989 by dividing the annual sum by the corresponding sum for 1989. For 1991 through 1996 we have divided by the sum for 1991. This two-tier procedure avoids problems that would have arisen if we had produced an index for all years in the whole period, because Sweden had a major tax reform in 1990. We have plotted the indexes for the two sub-periods in [Figure 12] and are struck by the close resemblance with the curve for Model 2 in [Figure 9]. We see this as evidence of an influence of directed public policies on first-birth behavior in the Swedish population.

It is important to note that these ups and downs in Swedish first-birth rates are genuine and not artificial consequences of tempo changes produced by reforms in family legislation, say, the way some of the fluctuations in period Total Fertility are [Figure 1]. Some changes in social policies in the 1980s induced a shorter spacing of births. This caused second and third births to occur at a quickened pace ([1], [7], [9]), but it did not affect first births. Childbearing among childless women has been affected by public policies and economic fluctuations (see also [5]), but not by changes in the tempo of childbearing.


The individual woman's labor-market situation Acknowledgements

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