We examine the effect of delaying motherhood on the transition to the second childbirth across European countries. There exist two opposite forces of delaying the first birth: biological and socio-cultural factors producing a postponement effect and career-related factors leading to a catch-up effect. Estimating a multistate duration model that addresses the endogeneity of age at first birth, we find a catch-up effect in countries where the career effect is large and a postponement effect in countries where the opportunity cost of childbearing is relatively high due to the lack of family friendly institutions and cultural influences, which may discourage late childbearing.
This is a preview of subscription content,to check access.
Access this article
Our focus on the first two parities is motivated by the observed decline in the TFR below two children per woman, which makes the transition to the second birth the relevant margin to study. Moreover, despite the declining trend in TFRs, survey data usually show that the modal desired number of children per woman is still two, and that many women fail even this relatively low fertility target (Bongaarts 2001).
Our work is also related to cross-country studies that observe a reversal of the negative correlation between female employment and fertility (Ahn and Mira 2002).
To avoid potential confusion, these social influences could push women to have their first birth earlier. This is something completely different from the postponement effect, which posits that women who have their first birth later are less likely to proceed to the second parity.
Theoretically, there is no reason why the income effect should prevail over the substitution effect. Empirically, evidence that late motherhood is positively associated with mothers’ wages can be found in Amuedo-Dorantes and Kimmel (2005) and Miller (2009), while Davies and Pierre (2005) use ECHP data to show that mothers under 24 are more likely to suffer from a family wage gap than older mothers. Del Bono et al. (2008) show that an unexpected career interruption, which lowers the opportunity cost of giving birth for working mothers, has a sizeable negative effect on women’s fertility. This effect could be explained by a reduction in expected life-time earnings due to the destruction of working mothers’ firm-specific human capital, which leads to a reduction in permanent income.
Estimation of country-specific regressions, which do not impose equality of coefficients across countries, is frequent in the literature using the ECHP, see Nicoletti and Tanturri (2008), Del Boca and Sauer (2009), and Del Boca et al. (2009), among others. However, we also provide evidence based on a pooled cross-country sample in Section 5.6, which allows for interactions between age at first birth and country specific institutional features.
The use of bivariate duration models based on the proportional hazard assumption is common in the analysis of labour market dynamics (see Lancaster 1990 and van den Berg 2001 for an overview), but it has also been used extensively in health economics in the analysis of the use of alcohol and tobacco (van Ours 2004) or drugs (van Ours 2003).
Miller (2009) in her study of age at motherhood on career path uses biological fertility shocks as instruments, in particular an indicator for first pregnancy ending in miscarriage or stillbirth, an indicator for “accidental” first pregnancy occurring while using contraception, and the lag in years from first attempt to conceive to first birth. Unfortunately, these variables are not available in the ECHP.
See the discussion on identification at the end of this section. Although we assume that the random effects are orthogonal with respect to the covariates in the model, the model allows for correlation between the unobserved effect and the main variable of interest (age at first birth).
Although our specification of the transition to both the first and the second birth is based on a proportional hazard model, Brinch (2007) shows that variation in covariates overtime combined with variation across observations is sufficient to ensure identification without the proportional hazard assumption.
Censored spells are contributed both by relatively “young” women, that is those who still have to complete their reproductive careers, and by women who drop from the panel due to panel attrition.
We have also considered three mass points in the empirical analysis, but either they did not improve the estimation results, or they could not be identified.
For the same reason, we do not include partners’ characteristics. Also in line with the previous empirical literature on tempo effects, we do not account for the potential endogeneity of educational variables. This means, for instance, that the coefficient on educational variables could also proxy for an individual’s preference towards childbearing. By including the higher education achieved in both the first and the second birth equations we aim to capture the effect of age at first birth on the timing of the second birth over and above the association between age at first birth and the level of completed education.
It should be noted that both definitions of career orientation do not necessarily imply that women must be continuously attached to the labour force. Indeed, a sufficient condition for observing career effects would be that working women who delayed their first birth have accumulated more resources to afford a further birth, even if they decided to drop the labour force after their first childbirth.
We do not describe in detail the estimates of the transition to the first parity, since they do not represent the main focus of our paper. Delaying of first birth in Europe has been recently investigated by Nicoletti and Tanturri (2008).
We only report in Table 2 the main variables of interest for the transition to the second birth. The full estimates from the jointly estimated model of specification 1 for both transitions are reported in Table 8 in the Appendix. Specification 1 of Table 8 is equivalent to the one of Table 7 based on the piecemeal approach.
For specification 2, the effect is positive but not significant for Belgium, Germany and the UK, while for specification 3 the effect is negative but again not significant for Belgium, Germany and Greece.
The effect of age at first birth for Denmark refers to those women who have been employed. The effect of having been employed cannot be identified due to low variation, as most of women have been employed at least once in their lifetime.
These results are available by the authors upon request.
A description of the variables used and the relative data sources are reported in the footnote of Table 5. For the series that do not cover the whole time period of fertility histories that we consider (1976–2001), we use the first observed value for all years of the unobserved period (that is the period not covered by the time-series). In case of missing values within the “observed” period, we use the mean between the first observed year preceding the missing year and the first observed following year.
Lalive and Zweimüller (2009) find that parental leave extensions affect positively fertility.
Ahn N, Mira P (2002) A note on the changing relationship between fertility and female employment rates in developed countries. J Popul Econ 15(4):667–682
Amuedo-Dorantes C, Kimmel J (2005) The motherhood wage gap for women in the United States: the importance of college and fertility delay. Rev Econ Househ 3(1):17–48
Apps P, Rees R (2004) Fertility, taxation and family policy. Scand J Econ 106(4):745–763
Bettio F, Villa P (1998) A Mediterranean perspective on the breakdown of the relationship between participation and fertility. Camb J Econ 22(2):137–171
Billari FC, Kohler H-P, Andersson G, Lundström H (2007) Approaching the limit: long-term trends in late and very late fertility. Popul Dev Rev 33(1):149–170
Blackburn ML, Bloom DE, Neumark D (1993) Fertility timing, wages and human capital. J Popul Econ 6(1):1–30
Bongaarts J (2001) Fertility and reproductive preferences in post-transitional societies. Popul Dev Rev 27(Supplement):260–281
Bratti M (2003) Labour force participation and marital fertility of Italian women: the role of education. J Popul Econ 16(3):525–554
Brinch CS (2007) Nonparametric identification of the mixed hazards model with time-varying covariates. Econom Theory 23(2):349–354
Cameron SV, Heckman JJ (1998) Life cycle schooling and dynamic selection bias: models and evidence for five cohorts of American males. J Polit Econ 106(2):262–333
Cigno A, Ermisch JJ (1989) A microeconomic analysis of the timing of births. Eur Econ Rev 33(4):737–760
Da Rocha JM, Fuster L (2006) Why are fertility rates and female employment ratios positively correlated across O.E.C.D. countries? Int Econ Rev 47(4):1187–1222
Davies R, Pierre G (2005) The family gap in pay in Europe: a cross-country study. Labour Econ 12(4):469–486
Del Boca D, Sauer RM (2009) Life cycle employment and fertility across institutional environments. Eur Econ Rev 53(3):274–292
Del Boca D, Pasqua S, Pronzato C (2009) Motherhood and market work decisions in institutional context: a European perspective. Oxf Econ Pap 61(suppl. 1):i147–i171
Del Bono E, Weber A, Winter-Ebmer R (2008) Clash of career and family: fertility decisions after job displacement. IZA Discussion Paper 3272, Institute for the Study of Labor (IZA), Bonn
Esping-Andersen G (1990) The three worlds of welfare capitalism. Polity, Cambridge
Fernandez R, Fogli A (2006) Culture: an empirical investigation of beliefs, work, and fertility. Am Econ J: Macroeconomics 1(1):146–177
Fernandez R, Fogli A (2009) Fertility: the role of culture and family experience. J Eur Econ Assoc 4(2–3):552–561
Francesconi M (2002) A joint dynamic model of fertility and work of married women. J Labor Econ 20(2):336–380
Gauthier AH (2002) Family policies in industrialized countries: is thereconvergence? Population 57(2):447–474
Gustafsson S (2001) Optimal age at motherhood. Theoretical and empirical considerations on postponement of maternity in Europe. J Popul Econ 14(2):225–247
Gustafsson S, Kenjoh E, Wetzels C (2002) The role of education on postponement of maternity in Britain, Germany, the Netherlands, Sweden. In: Ruspini E, Dale A (eds) The gender dimension of social change: the contribution of dynamic research to the study of women’s life courses. The Policy Press, Bristol, pp 55–79
Heckman JJ, Singer B (1984) The identifiability of the proportional hazard model. Rev Econ Stud 51(2):231–241
Heckman JJ, Walker J (1990) The third birth in Sweden. J Popul Econ 3(4):235–275
Heckman JJ, Hotz VJ, Walker J (1985) New evidence on the timing and spacing of births. Am Econ Rev 75(2):179–184
Hotz VJ, Miller RA (1988) An empirical analysis of life cycle fertility and female labor supply. Econometrica 56(1):91–118
Kohler HP, Skytthe A, Christensen K (2001) The age at first birth and completed fertility reconsidered: findings from a sample of identical twins. MPIDR Working Paper 2001-006. Max Planck Institute for Demographic Research, Rostock
Kohler H, Billari FC, Ortega JA (2002) The emergence of lowest-low fertility in Europe during the 1990s. Popul Dev Rev 28(4):641–680
Lalive R, Zweimüller J (2009) Does parental leave affect fertility and return-to-work? Evidence from two natural experiments. Q J Econ 58(6):1363–1402
Lancaster T (1990) The econometric analysis of transition data. Cambridge University Press, Cambridge
Langdridge D, Blyth E (2001) Regulation of assisted conception services in Europe: implications of the new reproductive technologies for “the family”. J Soc Welf Fam Law 23(1):45–64
Lutz W, Skirbekk V (2005) Policies addressing the tempo effect in low-fertility countries. Popul Dev Rev 31(4):699–720
Martínez DF, Iza A (2004) Skill premium effects on fertility and female labor force supply. J Popul Econ 17(1):1–16
Miller A (2009) The effects of motherhood timing on career path. J Pop Econ. doi:10.1007/s00148-009-0296-x (forthcoming)
Moffitt R (1984) Profiles of fertility, labour supply and wages of married women: a complete life-cycle model. Rev Econ Stud 51(2):263–278
Morgan SP, Rindfuss RR (1999) Reexamining the link of early childbearing to marriage and to subsequent fertility. Demography 36(1):59–75
Nicoletti C, Tanturri ML (2008) Differences in delaying motherhood across European countries: empirical evidence from the ECHP. Eur J Popul 24(2):157–183
van den Berg GJ (2001) Duration models: specification, identification and multiple durations. In: Heckman JJ, Leamer E (eds) Handbook of econometrics, vol V. North-Holland, Amsterdam (Chapter 55)
van Ours JC (2003) Is Cannabis a stepping-stone for cocaine? J Health Econ 22(4):539–554
van Ours JC (2004) A pint a day raises a man’s pay; but smoking blows that gain away. J Health Econ 23(5):863–886
Walker JR (1995) The effect of public policies on recent Swedish fertility behavior. J Popul Econ 8(3):223–251
We thank three anonymous referees, Deborah Cobb-Clark, John Ermisch and Daniel Hamermesh for their very helpful comments and discussions. Earlier versions of this paper (circulated under the title “Explaining How Delayed Motherhood Affects Fertility Dynamics in Europe”, IZA DP No 3907) have also benefited from comments received at seminars at IZA, RWI-Essen, MPI in Rostock and Keele University, the BHPS Conference in Colchester, the III Workshop on ‘Economics of the Family’ in Zaragoza, the ESPE Conference in London, the EALE Conference in Amsterdam and the EEA Conference in Barcelona. The paper is partly based on work carried out by Massimiliano Bratti during his visits to the European Centre for Analysis in the Social Sciences (ECASS) at the Institute for Social and Economic Research (University of Essex) supported by the Access to Research Infrastructure action under the European Community’s ‘Improving Human Potential Programme’, and at IZA in Bonn, which are both acknowledged for their financial assistance. The usual disclaimer applies.
Responsible editor: Junsen Zhang
About this article
Cite this article
Bratti, M., Tatsiramos, K. The effect of delaying motherhood on the second childbirth in Europe. J Popul Econ 25, 291–321 (2012). https://doi.org/10.1007/s00148-010-0341-9