Journal of Population Sciences
Table 1 Studies on childcare services and fertility
From: Formal childcare services and fertility: the case of Italy
Author(s) | Outcome | ECEC indicator | Data | Method | Country (years) | Sample | Result |
|---|---|---|---|---|---|---|---|
Blau and Robins (1989) | Transition to a first birth or to a higher-order birth | Childcare Tax Credit | Retrospective survey data covering over a period of time ranging from 16 to 22 months | Discrete-time hazard models | US (1979–1980) | 15,110 married women | No significant effect of child-care subsidies on the rate of having a birth while employed |
Kravdal (1996) | Transition to a first birth or to a higher-order birth | ECEC coverage (1–3) at the regional level | Retrospective survey data combined with migration biographies and regional time-series data on ECEC coverage | Multilevel discrete-time logit models | Norway (1964–1991) | 4019 women | No significant effect of ECEC coverage on first or second birth transition; weak positive effect on third parity |
Del Boca (2002) | Transition to a first birth or to a higher-order birth | ECEC availability (0–2) at the regional level | Longitudinal survey data (Survey on Household Income and Wealth—SHIW) combined with regional time-series data on ECEC | Fixed effects conditional logit models | Italy (1991–1995) | 1708 married women (ages 21–45) | Modest positive effect of ECEC availability first birth or higher-order birth transition, not statistically significant for the common standard |
Hank and Kreyenfeld (2003) | Transition to a first birth or to a higher-order birth | ECEC availability (3–5 years) at the district/sub-regional level | Longitudinal survey data (German Socio-Economic Panel Study—SOEP) combined with district-level data on ECEC | Multilevel discrete-time logit models | West Germany (1984–1999) | 2890 women for first births; 1585 women for second births (age 20–35) | No significant effect of ECEC availability on first or higher-order birth transition |
Sleebos (2003) | Fertility | ECEC services | Available literature | Literature review on the topic | Europe and North America | 5 studies on formal childcare in Europe | Some studies find a positive but weak effect of ECEC on fertility; some studies do not find any effect |
Andersson et al. (2004) | Transition to a second birth or to a third birth | ECEC services (1–12 years), i.e. availability rate, the child-to-staff ratio, costs of care to parents | Longitudinal register data for the 1980s and 1990s combined with municipal-level data on ECEC | Multilevel discrete-time logit models | Sweden (1997–1998) | 500,000 couple-years | No significant effect of ECEC on second or third birth transition |
Rønsen (2004) | Transition to a first birth or to a higher-order birth | ECEC coverage (0–6) | Retrospective surveys (the 1988 Norwegian Family and Occupation Survey and the 1989 Finnish Population Survey) | Cox proportional regression | Norway and Finland (1970s and 1980s) | 3639 women in Finland and 3296 women in Norway born between 1943 and 1967 | Finland: positive effects on 1st and 3rd birth (no effect on second) Norway: positive effect only on 1st birth |
Rindfuss et al. (2007) | Transition to a first birth | ECEC usage (0–6) at the municipal level | Longitudinal census and register data for the 1970s combined with childcare coverage rates at the local level | Discrete-time hazard models | Norway (1973–1998) | 175,722 women (birth cohort 1957–1963, age 15–35) | Positive effect but mostly due to ECEC services for children aged 3–5 |
Gauthier (2007) | Fertility | ECEC services | Available literature | Literature review on the topic | Europe and North America | 11 studies on childcare in Europe | Some studies find a positive but weak effect of ECEC on fertility; some studies do not find any effect |
Thévenon (2009) | Fertility rate | Different types of family policies, including ECEC | Available literature | Light and not systematic literature review on the topic | France and Europe | Literature review on the topic | Some studies find a positive but weak effect of ECEC on fertility; some studies do not find any effect |
Baizán (2009) | Transition to a first birth or to a higher order birth | Different types of family polities, including ECEC usage (0–2) at the regional level | Longitudinal survey data for Spain (European community household panel—ECHP) combined with region-level data on ECEC | Event history models with regional fixed effects | Spain (1993–2000) | 4,303 women (ages 16–42) | a) significant effect of ECEC usage on birth; b) significant effect of degree of adaptation of social institutions to changes in gender roles only for second/more birth |
Del Boca et al. (2009) | a) working status; b) childbirth | ECEC usage (0–2) at the regional level, among other variables | Longitudinal survey data (European Community Household Panel—ECHP) combined with regional/country level data on ECEC (Eurostat REGIO database) | Bivariate probit model, estimating jointly probabilities of women’s decisions to participate in the labour market and to have children | Italy, Spain, France, Belgium, the Netherlands, Denmark, and the UK (1999) | 10,321 women (ages 21–45) | No significant or substantial effect of ECEC usage on childbirth |
Hilgeman and Butts (2009) | Realised fertility (total number of children ever born at the time of the interview) | ECEC usage (0–3) at the country level | Cross-sectional survey data (European Value Survey and World Value Survey) combined with country-level data on ECEC and female labour market participation | Hierarchical Bayesian model | 18 European countries, the US, and Australia (1997–2000) | 7,080 women (ages 18–45) | Positive effect of ECEC services on fertility, especially in countries starting from a very low level of coverage |
Rindfuss et al. (2010) | Transition to childbirth by parity (1–5) | ECEC usage (0–6) at the municipal level | Longitudinal census and register data for the 1970s combined with childcare coverage rates at the municipal level | Discrete-time hazard models | Norway (1973–1998) | 175,722 women (birth cohort 1957–1963, age 15–35) | Positive effect but mostly due to ECEC services for children aged 35 |
Van Bavel and Różańska-Putek (2010) | Having a second child | ECEC enrolment (0–2) at the country level | Cross-sectional data (third round of the European Social Survey—ESS) combined with childcare enrolment rates across Europe (OECD Family Database) | Discrete-time event history multilevel logistic regression models | 16 European countries (2006–2007) | 23,617 one-child mothers’ | Positive effect of ECEC availability on second birth transition, limited to highly educated mothers |
Lappegård (2010) | Transition to a second birth or to a third birth | ECEC coverage rate (1–2) at the municipal level | Registered data combined with administrative data on childcare services | Discrete-time hazard models (with municipal-level fixed effects) | Norway (1995–2002) | 159,430 one-child co-resident couples; 116,589 two-child couples | No significant effect of ECEC availability on second or higher-order birth transition |
Thévenon and Gauthier (2011) | Fertility rate | Different types of family policies, including ECEC enrolment rates | Available literature | Literature review on family policies effects on fertility | Europe | 2 studies on formal childcare | Positive effect of ECEC on fertility |
Fiori (2011) | Working women’s intention to have a second child | Proportion of children aged 0–2 enrolled in municipal kindergartens at the regional level (variable used in quartiles) | Data from the ISTAT Survey on Births | Multilevel logistic regressions | Italy (2005) | 5,145 women who had a child between 18 to 21 months prior to their interviews | No significant effect of ECEC on intention to have a second child |
Mörk et al. (2013) | Birth rate | Implementation of a fee maximum cap (lowering childcare costs for most households) in 2002 | Register data combined with childcare fees at the municipal level | Difference-in-Differences approach | Sweden (1996–2003) | 44,917 couples (woman’s age 20–45) | Positive effect of ECEC fee cap on first and higher-order births. Positive effect of ECEC fee cap only on timing of second births. Fertility increased mainly driven by low-income households |
Luci-Greulich and Thévenon (2013) | Fertility rate | Different types of family policies, including ECEC (both in terms of coverage and expenditure) | Macro panel data | Two-way Fixed Effects estimation model | 18 OECD countries (1982–2009) | 18 countries*27 years | Positive effect of ECEC on fertility rate (stronger than leave entitlements) |
Lee and Lee (2014) | Total fertility rate | Total capacity of childcare centres in terms of number of children | Aggregate level time-series data | Granger causality method | Japan (1971–2009) | Aggregated data on Japan for every year between 1971 and 2009 (n° 38 observations) | No significant effect of ECEC on TFR |
Nakajima and Tanaka (2014) | Transition to a first birth or to a higher-order birth | Daycare Services Expenses per child by municipality | Retrospective survey data covering over a 36 months period in four major metropolitan areas combined with municipal level data | Two-stage estimation models (household’s location choice model in the first stage; the fertility decision model in the second stage) | Japan (2001–2004) | 5,697 households with a wife of childbearing age, i.e. between 16 and 50 | No significant effect of ECEC on fertility |
Baizán et al. (2016) | Total number of own children living in the same household as the mother at the time of interview | Different types of family policies, including ECEC coverage and usage (0–2) at the country level | Longitudinal data (European Statistics on Income and Living Conditions—EU-SILC) combined with country-level data on ECEC services | Multilevel (random effects) Poisson’s regression models | 16 Western and Southern European countries (2004–2009) | 69,213 women (age 36–44) | Positive effect of ECEC coverage or usage on the total number of children in household, even when controlling for gender norms |
Bauernschuster et al. (2016) | Age-specific birth rates | Public ECEC coverage (0–3 years) at the county level | Register data combined with administrative data on ECEC coverage | Difference-in-Differences approach | West Germany (1998–2008) | 325 West German counties Women (age 15–44) | Positive effect of ECEC coverage on the number of births, with stronger effect at higher parities |
Wood et al. (2016) | Transition to a second-birth | ECEC enrolment (0–2) at the country level | Longitudinal microdata (Harmonized Histories) combined with contextual data on family policy (Comparative Family Policy Database) | Discrete-time hazard models | Belgium (2008–2010), Germany (2005), France (2005), Norway (2007–2008), Netherlands (2003), Spain (2006), Great Britain (2005–2006) | 22,298 one-child mothers (age 15–49) | Childcare enrolment has significant positive effect on second births hazards in the first 3 years after the first birth |
D’Albis et al. (2017) | Transition to a second-birth | ECEC coverage (0–2) at the country level | Longitudinal microdata (EU-SILC) combined with country’s childcare coverage for children aged 0 to 2 | Multilevel logistic regressions | 26 European countries (2003–2011) | 22,143 observations (person-years) of one-child mothers (age 15–45) | Moderating role of ECEC services on the relationship between mother’s education level and second child births: in countries with low childcare coverage, the relationship is U-shaped, while in countries with high childcare coverage, the probability of second child birth is increasing with education |
Fukai (2017) | Birth rate as the number of births in a given year from among the total population of women of childbearing age | Ratio of childcare centre capacity to the number of children aged 0–5 | Aggregate level time-series data at the municipal level | OLS and IV regressions | Japan (2000–2010) | 1749 municipalities * 3 year (2000, 2005, 2010) | Positive effect of childcare availability on fertility only for women living in regions where the propensity for women to work is high; no significant effect in other regions |
Schaffnit and Sear (2017) | Transition to a second-birth | Use of paid childcare | Longitudinal microdata (Millennium Cohort Study) | Model averaging for logistic regressions | UK (2000–2008) | 3,893 one-child mothers | Negative effect of use of paid childcare on transition to second-birth |
Wesolowski and Ferrarini (2018) | Total fertility rate | a) earner–carer support policies (ECEC 0–2 only in sensitivity analysis); b) traditional–family support policies | Data at the country level (Social Policy Indicator database—SPIN) | Pooled time-series regressions with country fixed effects and stepwise control for female labour force participation, unemployment rates and GDP | 33 industrialised countries (1995–2011) | Country*year (33 countries*16 years) | Positive effect of ECEC availability on fertility |
Wood and Neels (2019) | Transition to a first birth or to a higher-order birth | ECEC availability (0–3) at the municipal level | Longitudinal census and register data for the 2000s combined with childcare coverage rates at the local level | Both multilevel models, and municipality fixed-effects models | Belgium (2001–2005) | 157,476 dual-earner couples at risk of a first birth; 216,331 couples at risk of a second birth and 321,576 couples at risk of a third birth (censored at women’s age 50) | Positive effect of ECEC availability on transition to a first birth or to a higher-order birth |
Wood (2019) | Transition to a first birth among dual-earner couples | ECEC availability (0–3) at the municipal level | Longitudinal census and register data for the 2000s combined with childcare coverage rates at the local level | Random effects and fixed effects discrete-time hazard models | Belgium (2001–2005) | 157,476 couples at risk of a first birth, 216,331 couples at risk of a second birth and 321,576 couples at risk of a third birth (censored at women’s age 50) | Positive effect of ECEC availability on transition to a first birth |
Sobotka et al. (2019) | Fertility rate | ECEC availability and costs | Available literature | Literature review on ECEC effect on fertility | Many OECD countries and medium–high income countries | Literature review and empirical illustrations | Positive effect of ECEC availability on fertility |
Bergsvik et al. (2021) | Transition to a first birth or to a higher-order birth | ECEC services | Available literature | Systematic literature review on quasi-experiments | West Germany, Sweden, Norway, Belgium | 5 studies on formal childcare | Positive effect of ECEC availability on transition to a first birth |
Aassve et al. ( 2021b) | Period TFR | ECEC coverage (% municipalities with ECEC services) | Aggregated data at the county level (Italian National Institute of Statistics—ISTAT) | Difference-in-Differences approach | Italy (2004–2013) | 103 provinces | Positive effect of ECEC coverage as a buffer in times of uncertainty |
Schuss and Azaouagh (2022) | Transition to a first birth or to a second birth | Childcare coverage | Longitudinal data (German Socio-Economic Panel—SOEP) combined with data on childcare services at the county-level | Semiparametric Cox hazard models with a Difference-in-Difference approach | West Germany (1998–2012) | Women married or cohabiting (age 21–45) | ECEC increases the transition probability to first birth by 11.9% for native childless couples who were in the labour force before childbearing; No significant effect of ECEC increase in the transition probability to second birth |
Dimai (2023) | Transition to a second birth or to a higher-order birth | ECEC subsidy | Longitudinal administrative data on the means test certification, matched with the actual subsidy requests | Propensity score matching and event history analysis model | One region in North-East Italy (2017–2020) | Households who had a child in 2016 and obtained a means test certification in the years 2017–2020 | Positive (although small) effect of ECEC subsidy on the probability of having another child in the Friuli-Venezia Giulia |