VoxEU Column Gender Labour Markets

Labour in the past, present, and future: A trilemma in China

In a country where grandparents provide a significant amount of childcare, China’s plans to gradually delay retirement over the next few decades may significantly impact the labour supply and lifetime earnings of young women. Using the China Family Panel Studies survey data, this column demonstrates that the provision of grandparental childcare affects females’ income, in particular better-educated, urban females with younger children. An increase in public childcare subsidies may be required to complement the phasing-in of the retirement policy in China.

China has become one of the first major emerging market economies to suffer from an ageing population. Its working population has been steadily shrinking since 2012 (World Bank Open Data 2017). To ease the increasing labour supply shortages and pressure on the social security system, the state has decided to gradually delay retirement over the next few decades – the statutory retirement age (currently 60 for men, 55 for women in the public sector, and 50 for women who work in factories) will be raised gradually until it reaches the level of advanced economies (65 or more) by the 2040s or so.

Economic research has often shown that the older generation and the young do not compete directly in the labour market, so that retaining the older generation in the labour force does not necessarily crowd out the employment of the young (Gruber et al. 2010). However, we hypothesise that this may not be the case in China, because young children in China are often looked after by their retired grandparents. Such an intra-family, downward intergenerational labour transfer effectively reduces the childbearing-induced interruption in the careers of working young females, as well as their opportunity cost of working. Raising the retirement age may shift more of the burden of childcare from grandparents onto young females, thus deterring young females’ willingness to work or forcing them to take a longer break in their career paths. This implies a larger interruption in young females’ human capital accumulation and a lower level of their lifetime labour income flows.

From the parents’ point of view, childcare as the production of future labour supply is characterised by a strong positive externality: parents cannot get the full benefit from producing the future labour supply, but they have to incur the full cost. As a consequence, inputs to childcare will be insufficient unless it is subsidised privately (e.g. by grandparents, sacrificing their own supply in the labour market) or publicly. For this reason, without sufficient public subsidies, China’s demographic policies are very likely to be stuck in a trilemma whereby it is impossible to achieve optimal labour supplies from the older, younger, and future generations at the same time.

Figure 1 Providers of daytime childcare for children of different ages

The China Family Panel Studies (CFPS) 2014 survey clearly shows the importance of grandparental childcare as a private subsidy for childbearing females (Yu et al. 2018). Out of 15,000 surveyed families across China, we find 4,277 females living with children below age 11. As shown in Figure 1, within those families, grandparents share almost half of the childcare with mothers for children aged 2-3, after young females’ maternity leaves are over and before children are admitted to kindergarten at age 3. Overall, at least 50% of grandparents take major responsibility for daily childcare, compared with 16% of grandparents in the US, 15% in Germany/Austria, and less than 2% in Scandinavia (Lumsdaine and Vermeer 2015). In addition, in the absence of a credible mechanism to properly monitor childcare workers, few families in China rely completely on babysitters hired from the market, mainly for safety reasons – in fact, only three families do so in the survey.

The importance of grandparental childcare explains a seemingly paradoxical combination in China. The country has one of the world’s highest labour-force participation rates for females (currently 61.5%,1 compared with 55.7% in the US, 50% for the euro area, and around 60% for Scandinavia). Yet its childbearing-induced penalty for female employment – computed as the childbirth-induced drop in the employment rate – is only 5.6% (for the urban population) or 3.0% (for the rural population) (Zhang 2011), much lower than 26.3% in the US (Cristia 2008), even though social protection for childbearing female employees in China is poor. 

Just because a large part of their childcare burden is shared by the older generation, as is shown in Table 1, females in our sample are 15.3% more likely to stay in the labour force, compared with those who do not get support from the older generation. The difference is greater for females with younger children who need more intensive care. By contrast, the difference is negligible for the males in the same families.

Table 1 Labour market participation rate for females/males with (Yes) or without (No) grandparental support

The impact of grandparental childcare on females’ annual labour income is even more striking, as shown in Table 2. Overall, females with grandparental support earn about 81% more than those without support; those with children aged 0-2 earn almost three times as much if they are supported by the grandparents. Furthermore, the difference in labour income is substantial even for females with older children – females without grandparental support have to incur a larger interruption in their human capital accumulation (either by dropping out of the labour force or by switching to more flexible, lower-paid jobs), and this has a persistent impact on their long-term income flows. Again, the impact of grandparental childcare on labour income is negligible for the males.

Table 2 Average annual income for childbearing females/males with (Yes) or without (No) grandparental support (in renminbi, CNY)

Our analysis also reveals that the employment and income penalties for females without grandparental support are stronger for better-educated, urban females with younger children. This is in contrast to the established evidence from advanced economies that less-educated females with lower incomes suffer more from childbearing-induced penalties in wages and employment (Anderson et al. 2002, Arpino et al. 2014). The reason is that better-educated females with higher incomes in China can hardly substitute grandparental support with childcare provided by the market, even if they can afford it. 

Given the trilemma in China’s demographic policies, the new retirement policy may reduce grandparents’ downward labour transfer towards their grandchildren, thus forcing young females to allocate more of their labour to childcare and crowding out their labour supply, especially among young females with the highest-quality human capital. Overall, the potential adverse impact on young females’ post-childbirth employment and lifetime income may even undo the new retirement policy that aims to increase China’s aggregate labour supply and the gross income of its pension fund via raising the statuary retirement age.

We therefore call for more social protection policies to be implemented along with the phasing-in of the retirement policy in China. Increasing public subsidies for childcare can provide policymakers with a way out of the trilemma, by giving incentives for young females to increase their input to childcare, or the production of the future labour supply. A functional market offering sufficiently qualified childcare services is needed too, which provides a substitute for falling intra-family downward labour transfers caused by elderly workers’ postponed retirements and enables young females to maintain their jobs after childbirth. In addition, policies against discrimination against childbearing females, as well as policies that provide insurance to birth-related career breaks and increase the involvement of males in childcare, may also alleviate the negative childbearing-induced shocks to females’ human capital accumulation and lifetime income flows.

Authors’ note: The views expressed in this column are those of the authors and do not necessarily reflect the views of their affiliated institutions.


Anderson, DJ, M Binder and K Krause (2002), “The motherhood wage penalty: Which mothers pay it and why”, American Economic Review 92(3): 354–358.

Arpino, B, CD Pronzato and LP Tavares (2014), “The effect of grandparental support on mothers’ labour market participation: An instrumental variable approach”, European Journal of Population 30(4): 369–390.

Cristia, JP (2008), “The effect of a first child on female labor supply: Evidence from women seeking fertility services”, Journal of Human Resources 43(3): 487–510.

Gruber, J, K Milligan and D Wise (2010), “Introduction and summary”, in Gruber, J, and D Wise (eds), Social security programs and retirement around the world: The relationship to youth employment, University of Chicago Press.

Lumsdaine, RL, and SJC Vermeer (2015), “Retirement timing of women and the role of care responsibilities for grandchildren”, Demography 52(2): 433–454.

Yu, H, J Cao and S Kang (2018), “Fertility cost, intergenerational labor division, and female employment”, CESifo Working Paper 7293.

Zhang, CC (2011), “Impacts of number of children on the labor supply and wage earnings of married women”, Population and Economics 2011(5): 29–35 (in Chinese).


[1] Computed from the International Labour Organization database (ILOSTAT).

1,575 Reads