14(38)
Regarding the first assumption, there are other changes in the social
security system introduced at the same time as the first and second
reserved months as well as the gender equality bonus but they generally
affect treatment and control groups equally.
There is one exception. On January 1, 1995, the replacement rate was
decreased from 90 to 80 percent of previous earnings. Although this
affected all parents equally, parents with children born before January 1,
1995 could keep their higher replacement rate until the end of 1996 (the
entire period of study in this paper). It should be noted, however, that the
reserved days were excluded from this change and was still replaced at 90
percent of previous earnings for children born after January 1, 1995.
2
The first reserved months may indicate incentives to have a child before
the reform for some parents, both because of the regulations introduced in
the parental leave scheme, but also because of the lower replacement rate.
The second reserved months and the gender equality bonus indicate
incentives to have a child after the introduction of the reforms as parental
leave became more generous.
This brings us to the question whether the potential parents could
anticipate reforms in sufficient time to plan a child according to the reform
changes. Governmental propositions on the second reserved month and the
gender equality bonus were proposed more than nine months before the
reforms were introduced, and it was also clear that these would be adopted
by Parliament. Thus, parents were able to plan for a child to be born after
the introduction of these reforms.
For the first reserved month, the situation is more complicated. Although
the Governmental proposition of a reserved month was presented in
February 1994, it was uncertain that it would pass in Parliament until after
the children born at the turn of 1994/95 were already conceived. However,
it was clear that the proposed reduction in reimbursement would be
introduced before the children were born at the turn of 1994/95 were
conceived. Thus, potential parents had incentives to plan a child before the
introduction of the reforms, but probably more so due to the lower
replacement rate than the introduction of the reserved month.
3
In order to obtain indications about endogenous sorting, we examine the
number of births at the time of the reforms and compare them with the
number of births at the same time the surrounding years. We also
investigate whether there are any differences between the samples in
observed characteristics.
4
When comparing the number of births around the reform cutoffs with
surrounding years the results are not clear cut. Figure A1 in Appendix A
presents the number of children born 18-31 of December and 1-14 of
January in 1991/92 - 2004/05. Although the number of children born in
January is higher than the number of children born in December all years,
the difference in number of born children varies over time. The difference in
2
Also, for children born after December 31, 2001, replacement for flat rate days was
increased from 60 SEK to 120 SEK (approx. 6 and 12 Euro). As the analysis in this
paper only covers earnings-related days this change is considered to be of minor
importance.
3
For a more detailed discussion, see Johansson (2010).
4
Of course, there could be endogenous sorting that does not show up in terms of
observables, but that is not possible to investigate in this study.
