The Impact of a Large Parental Leave Benefit Reform on the Timing of Birth around the Day of Implementation*
| DOI | http://doi.org/10.1111/j.1468-0084.2012.00707.x |
| Published date | 01 August 2013 |
| Date | 01 August 2013 |
585
©John Wiley & Sons Ltd and the Department of Economics, University of Oxford 2012.
OXFORD BULLETIN OF ECONOMICS AND STATISTICS, 75, 4 (2013) 0305-9049
doi: 10.1111/j.1468-0084.2012.00707.x
The Impact of a Large Parental Leave Benefit
Reform on the Timing of Birth around the Day
of Implementation*
Marcus Tamm
RWI Essen, Hohenzollernstr. 1-3, 45128 Essen, Germany
(e-mail: tamm@rwi-essen.de)
Abstract
The introduction of the German parental leave benefit (Elterngeld) applied to all children
born on 1 January, 2007 or later. The Elterngeld considerably changed the amount of
transfers to families during the first two years postpartum. We show that the incentives
created by using a cut-off date led more than 1,000 parents to postpone the delivery of their
children from December 2006 to January 2007. Concerning potential adverse impacts on
health outcomes of children we find a slight increase in average birth weight and the rate
of children with high birth weight (>4,000 g).
I. Introduction
Displaying one of the lowest fertility rates among European countries for several decades
and relatively long out of job periods of women after child birth, Germany experienced the
implementation of a new system of parental leave benefit in the year 2007. The preceding
benefit, which was means-tested and provided for up to two years after child birth, was
replaced by a wage-dependent benefit provided for up to 14 months which was applied to
all parents of children born on 1 January 2007 or later. Depending on prebirth labour mar-
ket participation (and earnings) and planned postbirth labour market activities households
might receive up to 25,200 Euro more or up to 3,600 Euro less under the new regulation than
under the preceding one. The main aims of this reform were to reconcile the work–family
balance, to increase labour market participation of mothers with young children, to increase
fathers’involvement in child care and to improve the financial situation of households with
young children. Several politicians also expressed their hope for an increase in fertility rates.
At the current point in time an evaluation of the long-term impact of the reform on
birth rates is not feasible. Instead this study is concerned with the impact on short-term
behaviour, specifically the timing of delivery of newborn children during December 2006
and January 2007. We analyse whether the use of a cut-off date (1 January) for allocating
ÅThe author is grateful to Thomas Bauer, David Card, Katja G¨orlitz and two anonymous referees as well as
seminar participants at UC Berkeley, RWI Essen, EALE (Tallinn)and ESPE (Sevilla) for helpful comments.
JEL Classification numbers: H31, J13.
586 Bulletin
parents to the old or to the new benefit system and the differences in transfer payments
between the two systems had an impact on the probability that a child was born in the last
days of December rather than in the first days of January.
Analysing these short-term effects is interesting for several reasons. First, the analysis
provides evidence on the power that parents have on the timing of birth. Overall there
is ample evidence that a certain number of deliveries are timed. For example, using our
data from birth registers we observe that the number of births taking place on weekends or
holidays is lower by around 21% than on regular working days. Yet, it is generally not clear
whether this timing is due to doctors and hospitals or due to parents. Any timing effects
around the cut-off date will be informative on the relative power of parents vis-`a-vis doc-
tors/hospitals concerning the choice or refusal of inducement and caesarean section and
the timing of such a delivery. Besides choosing the mode of delivery, mothers expected to
give birth at the margin might opt for more or less stressful, that is, labour-inducing behav-
iour. Second, timing effects represent revealed preferences and indicate whether parents
or sub-groups of parents are in favour of the reform (i.e. benefiting from the reform) or
not. This is interesting from a policy perspective as there has been uncertainty on the share
of parents that might be disadvantaged by the reform. Clearly, revealed preferences are
stronger indicators for preferences than results from opinion polls.
Third, the use of birth-related cut-offs to determine eligibility for policy programmes is
quite common and there is a small but growing literature analysing incentives created by
birth-related cut-off dates and their impact on the timing of delivery. In the literature very
different settings and institutions that rely on cut-offdates that were known to parents before
delivery have been analysed. Some cut-offs are associated with direct financial transfers,
for example with birth-related benefits or tax reductions. Other cut-offs are related to paren-
tal leave regulations which might have an indirect impact on income (or opportunity costs)
if they alter work incentives. Furthermore, there are cut-offs that are related to school or
kindergarten entry regulations which might have an impact if they reduce child care costs
or the opportunity cost to take care of the child by oneself. For some of these cut-offs birth
shifting was found (e.g. Dickert-Conlin and Chandra, 1999; Gans and Leigh, 2009) but not
for others (e.g. McEwan and Shapiro, 2008; Dickert-Conlin and Elder, 2010; Dustmann and
Sch¨onberg, 2011; Lalive and Zweim¨uller, 2009). Dickert-Conlin and Chandra (1999) show
that incentives provided by the US tax system1induce parents to shift delivery towards
December rather than January and that shifting is larger among high income parents. Gans
and Leigh (2009) analyse the Australian ‘Baby Bonus’, which was implemented in July
2004 and increased in July 2006 and gives one-time cash transfers to parents with babies
born after these dates. These bonuses heavily influenced parents’ behaviour and induced
postponements of deliveries by more than 14 days. In contrast, Lalive and Zweim¨uller
(2009) do not find that extending the leave duration from 12 to 24 months (including the
job-protection period and a monthly cash transfer of 340 Euro) in July 1990 affected birth
timing of Austrian women. Similarly, Dustmann and Sch¨onberg (2011) do not find that
an expansion of job-protected leave from 18 to 36 months implemented in January 1992
influenced birth timing in Germany. Concerning school entry-related cut-offs, McEwan
1Parents receive a full annual tax exemption for a given year if the child is born until December 31, while for
children born after December 31 parents receive the tax exemption for the following years only.
©John Wiley & Sons Ltd and the Department of Economics, University of Oxford 2012.
To continue reading
Request your trial