Skip to main content
SpringerLink
Log in

Where can childcare expansion increase maternal labor supply? A comparison of quasi-experimental estimates from seven countries

  • Published:
Empirical Economics Aims and scope Submit manuscript

Abstract

The estimated effect of childcare availability on maternal labor supply varies highly in previous single-country estimates. We provide comparable quasi-experimental estimates of the childcare effect for seven countries, using harmonized data and a uniform method based on country-specific childcare eligibility cutoffs. We evaluate the estimates in light of key institutional factors to determine under what conditions childcare expansion is likely to be effective. We propose a measure that captures childcare scarcity and predicts the effectiveness of childcare expansion: the gap between the participation rate of mothers with older children (aged 6–14) and childcare coverage under the age of 3. In countries with a high gap, we find that childcare availability has a significant positive impact on maternal labor supply (Austria, Czech Republic, Hungary, Slovak Republic). No significant impact is found in countries where the gap is low due to either already high childcare coverage (France) or the low participation of mothers with older children (Greece, Italy). We discuss other policies that need to be addressed concurrently for childcare expansion to achieve its goal of increasing mothers’ participation in the labor market.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3

Similar content being viewed by others

Data availability

The research utilizes European Labour Force Survey (EU LFS) data available from Eurostat upon request, self-collected country-level information on birthdate-based childcare enrollment cutoffs (available at https://www.dropbox.com/s/u7rpf2z8xkk9pr2/Expert_Questionnaire_EU.xlsx?dl=0) or from the authors upon request, and country-specific institutional characteristics based on publicly available data sources (OECD Family Database, European Social Survey).

Notes

  1. A further important aspect, which we do not study here, is the impact on child development and inequities. Universal preschool has been shown to have a positive impact on noncognitive skills and individuals’ success in the long run (Gray-Lobe et al. 2021; García et al. 2020; McCoy et al. 2017).

  2. The European Union (EU) set specific targets for its countries in 2002 and renewed them in the Europe 2020 Strategy, prescribing a 33% coverage rate for children under 3, and a 90% coverage rate for those between 3 and the mandatory school age by 2010 (EC 2013, 2008).

  3. The former socialist countries were characterized by a very well-developed childcare system, with relatively high nursery school coverage under age 3, which was dismantled following the transition. Nursery schools, however, were not considered pedagogical institutions, but rather healthcare ones, and were not regarded positively by the population.

  4. We assigned the month of the interview as the imputed month of birth at which the age of the child increases. In fact, the actual month of birth can be any of the three months between the previous and the current interview. As a result, if a child is classified as having a May birth month they could have been born in March, April, or May.

  5. For example, in Hungary, the legislature states that children who turn three prior to September 1st must be accepted into childcare in the same calendar year, while those born after may be accepted if places remain available. In a previous study (Lovász and Szabó-Morvai 2019), more detailed enrollment data is used to show that the effective cutoff during the time period studied (1998–2009) was actually January 1st: children born up to that date were generally accepted into childcare, while those born after had to wait until next September.

  6. The categorization of birthdate groups based on birth quarter is not exactly the same as what we use in our analysis, and not all quarters are observed in every country, which limits the test. Further, the data is limited in terms of the number of observations, the availability of observations for each birth quarter and child age, and the lack of sufficient data in the case of France and Slovakia (see Appendix Table 13). However, the comparison of the available birth quarters does provide some evidence of the existence of discontinuities in childcare enrollment in the case of the other countries in our sample.

  7. The March 1st cutoff corresponds to having turned 2.5 years old by September 1st, which, in the case of Hungary, has been an increasingly common rule of thumb used by kindergartens in admissions, leading to a change in the law in 2010 specifically allowing it.

  8. Mothers with older children who are in the treatment group were able to enroll their children in childcare earlier than older mothers in the control group, which means that they may differ from older mothers in the control group due to the longer-run effects of the earlier treatment. As a result, the treatment effect estimates may be biased downwards. As a robustness check, we include mothers of 2-year-olds as the comparison group where no such issues should arise, with similar results (see Appendix Tables 19 and 20).

  9. It is important to note that in addition to removing seasonal effects that are common to the main sample and the comparison group, results from this specification may also differ because it imposes a restriction on the model that the coefficients of further characteristics (controls) are the same for mothers of children of different ages. It may well be that the coefficients are, in fact, different for the original and the comparison sample of mothers, so the seasonality-corrected estimates may differ from the baseline estimates due to either seasonality biases being removed or the restriction on the other coefficients in the equation.

  10. The Q2 results may represent longer-term effects, however, they may also be indicative of the flexibility of the September 1st enrollment date. For some countries, experts noted that enrollment is allowed year-round, depending on availability. It is therefore not possible to tell whether any significant childcare effects observed in Q2 are due to longer term effects of enrollment in September, or shorter-term effects due to enrollment later in the year.

References

Download references

Acknowledgements

We are grateful to Gabor Kezdi, John Earle, Erdal Tekin, Agota Scharle, Joris Ghysels, and participants of the GDN Summer Workshop, the Szirák Labor Market Conference, IZA Young Scholars in DC, the Education Economics workshop in Leuven, the YEM 2017 conference in Brno, EEA-ESEM in Lisbon, seminars at CERS HAS, and the Virtual Research Collaboration on Gender and Family in the Labor Market for valuable comments. All remaining errors are ours.

Funding

The authors gratefully acknowledge financial support from the Hungarian National Scientific Research Program (OTKA), Grant no. FK131422 and the Lendület programme of the Hungarian Academy of Sciences (grant number: LP2018-2/2018). This research was supported by a grant from the CERGE-EI Foundation under a program of the Global Development Network. All opinions expressed are those of the authors. The project leading to this application received funding from the European Union's Horizon 2020 research and innovation programme under grant agreement No. 691676.

Author information

Authors and Affiliations

  1. HUN-REN KRTK KTI, Tóth Kálmán u. 4., Budapest, 1097, Hungary

    Ágnes Szabó-Morvai & Anna Lovász

  2. Economics Department, University of Debrecen, Böszörményi út 132, Debrecen, 4032, Hungary

    Ágnes Szabó-Morvai

  3. University of Washington Tacoma, 1900 Commerce Street, Tacoma, WA, 98402-3100, USA

    Anna Lovász

Authors
  1. Ágnes Szabó-Morvai
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Anna Lovász
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Ágnes Szabó-Morvai.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Appendices

Appendix A: Probability linking of households across quarters of the EU LFS dataset

Linking is based on exact matches (or logical increases/decreases, such as for age variables) of 56 variables describing household-level characteristics, household composition, and individual characteristics of certain members of the household, such as the year parents completed their highest level of education. We apply a very conservative strategy in the linking process, leaving a low chance for incorrect linking. We consider all observations with multiple links as unlinked. All codes for the linking process can be found at: https://github.com/szabomorvai/childcare_EU.

We checked the accuracy of our linking process using Hungarian LFS data, where it is possible to directly compare the stochastic panel resulting from the linking process based on quarterly EU LFS data to the original panel LFS data that is the basis of the EU LFS data but includes a household ID. Within households that include at least one person under 7, the process linked 62.5% of the households correctly, 1.7% incorrectly and could not link 35.8% of the households.

In the tables below, we provide descriptive statistics of the accuracy of the linking process, broken down by certain characteristics. These show that the linkage probability is very similar in different household types, thus the resulting sample selection is minimal (Table 6, 7, 8, 9, 10, 11).

Table 6 Number of persons in the household
Full size table
Table 7 Region
Full size table
Table 8 Level of education of person 1 in the household
Full size table
Table 9 Age of person #3 in household
Full size table
Table 10 Age of person #4 in household
Full size table
Table 11 Marital status of person #1 in household
Full size table

Appendix B

See Tables 12, 13, 14, 15, 16, 17, 18, 19, 20, 21 and 22.

Table 12 Descriptive statistics of the sample by country (2005–2012)
Full size table
Table 13 Enrollment rates by child age (quarter) and birth quarter, EU-SILC
Full size table
Table 14 CEE countries
Full size table
Table 15 Western EU countries
Full size table
Table 16 Southern EU countries
Full size table
Table 17 Robustness checks: placebo cutoff regressions—1
Full size table
Table 18 Robustness checks: placebo cutoff regressions—2
Full size table
Table 19 Robustness checks: estimated impact of cutoff on sample of mothers with children of age 2–1
Full size table
Table 20 Robustness Checks: Estimated impact of cutoff on sample of mothers with children of age 2–2
Full size table
Table 21 Robustness checks: seasonality-corrected equations with alternative comparison groups (2 and 5-year-olds)—1
Full size table
Table 22 Robustness Checks: Seasonality-corrected equations with alternative comparison groups (2 and 5-year-olds)—2
Full size table

Appendix C: Childcare eligibility cutoff information

To collect country-specific cutoff information, we reached out to subject area experts from every EU country. Those who agreed to participate in the survey were compensated for providing detailed data on childcare enrolment regulations and practices.

The survey form and information collected for each country is available at: https://www.dropbox.com/s/u7rpf2z8xkk9pr2/Expert_Questionnaire_EU.xlsx?dl=0.

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Szabó-Morvai, Á., Lovász, A. Where can childcare expansion increase maternal labor supply? A comparison of quasi-experimental estimates from seven countries. Empir Econ (2023). https://doi.org/10.1007/s00181-023-02531-6

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1007/s00181-023-02531-6

Keywords

JEL Classification

Search

Navigation