Rising trend of childlessness in China: analysis of social and regional disparities with 2010 and 2020 census data

Abstract

Introduction

In recent decades, there has been a significant increase in childlessness. This paper analysed childlessness in China, specifically examining its socio and regional disparities.

Methods

With data from China’s 2020 population census, supplemented with data from China’s 2010 population census and 2015 inter-censual 1% population sample survey, we used a basic indicator of age-specific childlessness proportion, a decomposition method, and probability distribution models to analyse, fit and project childlessness.

Results

We presented age-specific childlessness proportions for women as a whole and by socioeconomic features, decomposition and projection results. The childlessness proportion increased markedly from 2010 to 2020, reaching 5.16% for women aged 49. The proportion is highest for city women, followed by township women, and is lowest among village women, at 6.29%, 5.50% and 3.72 % for women aged 49, respectively. The proportion for women aged 49 with high college education or above was 7.98%, and only 4.42% for women with junior high school education. The proportion also exhibits marked provincial discrepancies, and the total fertility rate is negatively correlated with childlessness at the province level. The decomposition results distinguished the different contribution of change in educational structure and change in childlessness proportion for subgroups to the total childlessness proportion change. It is projected that city women, women with high education will have higher childlessness proportion, and the proportion will further increase with the rapid increase in education level and urbanisation.

Conclusions

Childlessness has risen to a relatively high level, and varies among women with different characteristics. This should be taken into consideration in China’s countermeasures to reduce childlessness and curtail further fertility decline accordingly.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• Data were obtained from China’s 2020 population census, supplemented with data from China’s 2010 population census and 2015 inter-censual 1% population sample survey.

• Childlessness was analysed for women aged 15–49 years based on the information collected on the number of children ever-born.

• We used three methods to analyse the data: age-specific childlessness proportion, decomposition method, and fitting and projection models.

• The decomposition method is used to examine the contribution of structural changes, such as changes in education and urban–rural population structures, to the change in the level of childlessness.

• The best-fitting model is selected from varieties of exponential, gamma, Gompertz, inverse-Gaussian, log-logistic, log-normal and Weibull to predict the level of future childlessness curve for 15–35 years old with a 95% CI.

Introduction

There are various definitions of childlessness, making it difficult to measure accurately. Depending on the context, it can refer to demographic childlessness, de facto childlessness or actual childlessness.¹ For a woman, childlessness can mean different things, such as never having given birth, having no living children with whom they are in contact or having no children with their spouse or partner.² The majority of the literature considers childlessness as having no live birth,³ which is the same as nulliparous, and that is the definition adopted in this paper. Childlessness can be classified as involuntary or voluntary, the latter also known as being childfree.^3–5

The trend of childlessness is increasing globally, and its impact has become a subject of interest. In some European nations, the proportion of childless women at the end of childbearing years is at least 20%.⁴ In 2010, around one-third of women born in the late 1960s in Hong Kong and Japan, 16% in Taiwan, 8% in South Korea remained childless. And similarly, 21% of women born in 1961–1965 in Singapore were also childless. While these statistics go over a decade into the past, recent data shows that childlessness has continued to increase.⁶ Based on 2020 data, the proportion of childless women born between 1975 and 1980 in Singapore has reached 28%. In Japan, the percentage of childless women born between the early 1950s and 1974–1976 has also reached 28%. Furthermore, women born in the late 1970s in South Korea had a childlessness proportion of almost 19%, and those in Hong Kong and Taiwan also had a high proportion of permanent childlessness.⁷ This increasing trend of childlessness worldwide suggests a shift in societal norms and a need to better understand the causes and consequences of childlessness.

The phenomenon of childlessness in China has also gradually attracted attention.^{8 9} For decades, the percentage of childlessness among women remained low, with a rate of 1.46% for women aged 40–49 in the 1982 population census,¹⁰ 1.69% for women aged 49 in the 1987 1% population sample survey,¹¹ 1.25% for women aged 49 in the 1990 population census,¹² less than 2% for women aged 49 in the 2017 National Fertility Survey¹³ and 4.16% for women aged 49 in the 2015 1% population sample survey.¹⁴ However, over the past few decades, there has been a gradual increase in the prevalence of childlessness, specifically as a result of infertility.^15–18 The National Health Commission of China indicated that the incidence of infertility was between 7% and 10% in 2021.¹⁹ Additionally, according to the latest 2020 population census, there has been a significant rise, reaching 5.16% in the proportion of childless 49-year-old women.⁷

Childlessness in China is a concern as it contributes to the country’s low fertility level. Experiences from European and East Asian countries have shown that childlessness is a significant factor in the decline of fertility rates and birth numbers.^{4 6 7} For China, the decline in period total fertility rate for first births declined markedly from 1990 to 2015, the tempo-adjusted and parity-adjusted total fertility rate for first births was only 0.9 in 2015, indicating a dramatic decline in first childbirths and a rise in childlessness.²⁰ Regarding the cohort fertility level, about 4% of 45 to 49-year-old women had no live births in 2015. It is predicted that the rise in childlessness will be the main reason for the country’s fertility decline in the coming years.²⁰

In this paper, we analyse the proportion of childless women in China using data from 2020 census, supplemented by data from 2010 census and 2015 1% population sample survey. Below we first present the Chinese context, and then we introduce the data and method for this analysis. Next, we present the results, including the trend of childlessness from 2010 to 2020, focusing on the latest data from 2020, the differences in childlessness proportion among women based on their residence (city, township and village), education level and province, as well as the decomposition and projection results. We then discuss and conclude.

Chinese context

Historically, Chinese cultural norms, influenced by Confucian ideology, expect women to get married and give birth, with their worth measured in their ability to have children, particularly male children.²¹ However, with the rapid socioeconomic development, attitudes towards childbearing have shifted dramatically. One theoretical perspective germane to explaining China’s paradigm shift from ‘traditional childbearing’ to this emerging childlessness trend is developmental idealism.

Developmental idealism is an important cultural and ideational model consisting of values and beliefs about socioeconomic development and its causal connection to other elements of societies.^22–24 Regarding fertility, developmental idealism contributes to global fertility decline,²⁵ and it has also had a significant impact on family and demographic change in China.²⁶ Based on the analysis of survey data from Gansu Province in China, it was posited that people’s attitudes about family and demographic behaviours are influenced by their beliefs about development.^{25 27} The ideational change has been embodied in persistent low fertility intention,²⁸ and recently in the childlessness trend in China. Having a child is no longer viewed as a necessity. According to recent nationally representative surveys conducted in 2017, 2019 and 2021, the prevalence of women born after 1990 who have no intention of having children was 4.9%, 4.5% and 9.5%, respectively.⁷ The number of women who are voluntarily childless has been increasing.²⁹

China’s fast urbanisation has had a substantial impact on the country’s rural–urban inequities. The urbanisation system has resulted in economic growth, expanded access to education and healthcare, and raised the standard of living for many Chinese individuals. This has encouraged individual autonomy, as people move from rural to urban areas in search of better opportunities, resulting in a breakaway from traditional family-centred ideology.³⁰ To a certain extent, it may explain why childlessness is characterised by rural–urban disparities in China.³¹ The rise of individualism and the pursuit of personal ideals may lead individuals to prioritise their own desires and goals over having children.³² The 2001 Family Planning and Reproductive Health study, which surveyed 39 586 women of childbearing age, found that among those who had no children, 6.5% of urban women and only 1.1% of rural women had no intention of having children.³³ Similarly, the 2017 National Fertility Survey showed that 3.6% of women with urban household registration and 1.4% of women with rural household registration intended to remain childless.³⁴

In addition, as economic development varies across provinces in China, it follows that voluntary childlessness, owing to the choice to pursue a career over family, will also be characterised by provincial factors as some provinces have higher levels of economic development and education compared with others. For example, provinces in eastern China, such as Shanghai and Beijing, have higher levels of economic development and education than those in western China. China’s birth policy³¹ and fertility levels³⁵ also exhibited provincial differences.

One of the strategies by which China was able to achieve its modernisation and rapid development was through education expansion, and education is a key component and an important mechanism for achieving progress as specified in the developmental idealism.³⁶ The higher education expansion began in 1999 and saw a significant increase in the number of women with college or above education, from 16.87 million in 2000 to 53.94 million in 2010 and 105.36 million in 2020. The impact of this is that more women spend more time schooling, consequently postponing childbearing or forgoing it for career development.³⁷ According to China’s national 1% population sample survey conducted in 1987 and 1995, the proportion of childlessness for women aged 45–49 with a college education or above was about 3%, and 1% for women with lower education levels. The 2017 National Fertility Survey data shows that 3.4% of women with college education and above did not intend to have a child, while that for women with junior middle school is only 1.1%.³⁴ In 2020, the childlessness proportion for women aged 49 with college education and above reached 7.98%, and that for women with junior middle school was 4.42%.⁷ As the majority of highly educated women live in urban areas, the educational difference in childlessness proportion further adds to the reason why childlessness may vary in rural and urban areas.

Moreover, China’s birth control policy, in its forms and latitude, promoted small families and low fertility.³⁸ In most provinces, urban couples were allowed to have one child, while rural couples were allowed to have a second child if the first was a girl.³¹ Its impact on China’s fertility and population development has been controversial.^{39 40} Although the policy in itself is not nulliparous, nevertheless, there is the perception that it created a generation of only children, who may be less inclined to have children of their own due to the financial and emotional burden associated with raising children in China.²⁰ This may also be linked to the aspect of the development idealism theory that emphasises ideational change.

Data

Population census data, vital statistics data and survey data can be used to estimate childlessness level. The National Bureau of Statistics provides the most credible data, specifically the census data and population sample survey data. Since 1953, China has conducted several population censuses, with the items in the questionnaires substantially increasing, and the survey techniques significantly improving.⁴¹ In 2020, China conducted its seventh national population census, enumerating 1411.78 million people in mainland China.⁴²

In this paper, we mainly use China’s 2020 census data, supplemented by 2010 census data and 2015 1% population sample survey data. During the implementation of the 2020 census, an entirely electronic method of data gathering was used. The census takers used electronic equipment to collect data and transmitted it directly to the database in real-time; on-site verification with individual IDs in the vital statistics system were carried out. Data processing was centrally deployed across the country, using internet cloud technology, cloud services and cloud applications. According to the postenumeration report, the under-reporting rate was only 0.05%, indicating the high quality and reliability of the census dataset.⁴³

However, the publicly accessible ‘China Census Yearbook 2020’ does not contain tabulated information on age-specific childlessness for females.⁴³ The data used in this article are an excerpted micro-sample of long-form data containing information on 1 388 302 individuals, including 467 848 females aged 15–64. Due to the extremely low proportion of births to unmarried mothers in China, unmarried women were not required to fill in birth information. Ever-married women aged 15–64 registered the number of children ever born but not the timing of childbirths, so no detailed cohort analysis could be conducted with this dataset. In this study, we regard a woman as childless if she had no live births by the time of the survey. In addition to national trends, we also describe the proportion of childlessness for women by residence, education level and province.

Data for childlessness in 2010 are from a micro-sample, including individual information excerpted from the 2010 census, containing 462 124 women aged 15–64. The 2015 data comes from the aggregated data in the 2015 1% population sample survey, which provides the age-specific childlessness data for that year. Although some data quality issues, such as recall biases in this retrospective question of birth history, and the possible misinterpretation of own live births, has been reported,^{41 42 44} the census data are still the most reliable data for this study’s objective.

Method

We analyse the data using three methods: the age-specific childlessness proportion as a basic descriptive indicator, the decomposition method, and the fitting and projection models.

Age-specific childlessness proportion

Childlessness is defined as the absence of any living birth in a woman’s life, and when a woman remains childless lifelong, it is referred to as permanent (definitive) childlessness.³ The age at which childlessness can be considered permanent or definitive is controversial.^{4 45} In this paper, we consider the age range of 15–49, and calculate the age-specific proportion of women who have not yet given birth to a live birth.

Decomposition method

We employ decomposition method to examine the contribution of structural changes, such as changes in education or in urban–rural population structures, to the change in the level of childlessness. In order to decompose the total change in childlessness proportion into education structure changes, we divided the education level into n categories. The childlessness proportions for two female groups at two different points or for two different ages, group 1 and group 2, is denoted by P₁and P₂, the proportion of women with different education levels in the two groups is ${\displaystyle {\text{FP}}_1^{\text{i}}}$ and ${\displaystyle {\text{FP}}_2^{\text{i}}}$, respectively, where i represents a certain level of education, ${\displaystyle {\text{P}}_1^{\text{i}}}$ is the childlessness proportion of women with education level i in group 1. Then the difference between the proportions of childlessness for two groups can be expressed as

$$\begin{gathered} {\displaystyle {\text{P}}_2-{\text{P}}_1=\underset{\text{i}}{\mathrm{\Sigma }}\left({\text{FP}}_2^{\text{i}}\times {\text{P}}_2^{\text{i}}-{\text{FP}}_1^{\text{i}}\times {\text{P}}_1^{\text{i}}\right)= \\ \underset{\text{i}}{\mathrm{\Sigma }}\frac{{\text{P}}_2^{\text{i}}+{\text{P}}_1^{\text{i}}}{2}\times \left({\text{FP}}_2^{\text{i}}-{\text{FP}}_1^{\text{i}}\right)+\underset{\text{i}}{\mathrm{\Sigma }}\frac{{\text{FP}}_2^{\text{i}}+{\text{FP}}_1^{\text{i}}}{2}\times \left({\text{P}}_2^{\text{i}}-{\text{P}}_1^{\text{i}}\right)} \end{gathered}$$

Where ${\displaystyle \underset{\text{i}}{\mathrm{\Sigma }}\frac{{\text{P}}_2^{\text{i}}+{\text{P}}_1^{\text{i}}}{2}\times \left({\text{FP}}_2^{\text{i}}-{\text{FP}}_1^{\text{i}}\right)}$ denotes the contribution of the educational structure changes, and ${\displaystyle \underset{\text{i}}{\mathrm{\Sigma }}\frac{{\text{FP}}_2^{\text{i}}+{\text{FP}}_1^{\text{i}}}{2}\times \left({\text{P}}_2^{\text{i}}-{\text{P}}_1^{\text{i}}\right)}$ represents the contribution of differences in the proportions of childlessness in the same education level among two groups to the overall difference in the proportion of childlessness, which can be regarded as level changes.

Fitting and projection models

We adopt probability models to fit and project the proportion of childlessness. Some studies have attempted to project permanent childlessness in female cohorts, but the results may be overestimated and misleading as it is difficult to capture the recovery of postponed first birth at late childbearing ages.⁴⁵ Some specific models can be used to fit observed cumulative proportions.^{37 46} For China’s population census, there is some information on the number of children ever born and childbirth during the past year prior to the survey time, but no detailed birth history record for cohort analysis. Given the limited data availability, we fitted and projected period age-specific childlessness proportions using a variety of probability distribution models, including exponential distribution, gamma distribution, Gompertz distribution, inverse-Gaussian distribution, log-logistic distribution, log-normal distribution and Weibull distribution. We then compared and selected the best-fitting model. In the results, the fitted value fits the entire age group of 15–49 years old, while the projected value fits only the period age-specific childlessness curve for 15–35 years old and predicts the level of future childlessness, with the upper bound and lower bound of a 95% CI.

Patient and public involvement

Patients or the public were not involved in the design, conduct, reporting or dissemination plans of our research.

Results

Overall trend in childlessness

The proportion of childlessness at all ages rose between 2010 and 2020, as indicated in figure 1A. In terms of age groups, the age-specific proportion among young women aged 20–30 rose markedly by more than ten percentage points from 2010 to 2020. The proportion increased from 5.39% to 10.91% in 35-year-old women, from 2.66% to 7.85% in 40-year-old women, from 1.55% to 5.86% in 45-year-old group, and the proportion of childlessness in 49-year-old group increased from 1.29% to 5.16%. The increase in the proportion among the lower age groups partly reflects the postponement of childbirth. The increase among the higher age groups indicates the increasing possibility of permanent childlessness.

Figure 1.

Childlessness proportion. (A) Childlessness proportion 2010–2020. (B) Childlessness proportion by residence in 2020. (C) Childlessness proportion for city women aged 45–49 in 2020. Numbers in parentheses represent the number of provinces falling in this range.

Rural–urban difference

As shown in figure 1B, the childlessness proportion among women living in cities was higher than that of women residing in townships and villages. Conversely, the childlessness proportion for women in townships and villages was relatively similar, with little variation between the two groups. At age 35, the proportion of childlessness among city women was 13.78%, 7.29% among township women and 9.09% among village women. At age 49, the proportion was 6.29%, 5.50% and 3.72%, respectively. The difference in childlessness proportion between urban and rural areas is the result of a combination of factors, including differences in marriage and childbearing attitudes that stem from the socioeconomic and cultural divide created by China’s dual urban–rural system.

Figure 1C shows the childlessness proportion among city women aged 45–49 by province. The proportion was higher than 7% in 11 provinces, and exceeded 9% in 5 provinces. Such a high proportion indicates a deviation from the traditional universal childbearing pattern for city women.

Difference by education level

Figure 2A, B demonstrates a clear correlation between a woman’s level of education and her likelihood of being childless, as evidenced by the age-specific childlessness proportion among women with different education levels in 2010 and 2020. In 2010, the childlessness proportion among 49-year-old women with junior high school education level or below was 1%, while the proportion was 1.56% for women with a high school education, and 4.94% for women with college education or above. However, in 2020, the childlessness proportion for women of the same age and education level had risen significantly, to approximately 5%, 7% and 8%, respectively.

Figure 2.

Childlessness proportion by education. (A) Childlessness proportion by education level in 2010. (B) Childlessness proportion by education level in 2020. (C) Childlessness proportion for women aged 45–49 with college education or above education by province in 2020. Numbers in parentheses represent the number of provinces falling in this range.

The childlessness proportion among women with varying education levels has become increasingly disparate, heralded by the rise in childlessness proportion across all education levels between 2010 and 2020. In 2010, the age-specific childlessness proportion for women with a college education or above was 3.6 percentage points higher than that of women with a junior high school education level or below. However, by 2020, the difference had widened to nearly 5 percentage points.

Furthermore, as indicated in figure 2A, B, the proportion of childlessness among the 15–24 age group in 2020 for women with primary school and below, junior high school and high school education was lower than that for women of the corresponding age and education level in 2010, particularly for those with junior high school education. This probably suggests that women with lower education, who are more likely to be rural residents, did not delay their marriage and childbearing as their counterparts with higher education. This group of women is more likely to be rural residents because the proportion of low education is higher in rural areas, and early marriage is also more prevalent in rural areas.⁴⁷ While the legal marriage age in China is 22 and 20 years for males and females, respectively, in rural areas, females sometimes get married before this age. In fact, studies indicate that the number of underage marriages, defined as marriage before 18 years of age, rose between 2000 and 2010.⁴⁸ And the ever-married proportion for those aged 15–19 years increased from 1990 to 2015.⁴⁹

As also shown in figure 2C, the childlessness proportion among women with a college education or higher was greater than 10% in 10 provinces. Considering the substantial growth in higher education, it is likely that the childlessness proportion will continue to increase.

Difference by province

Between provinces, there were significant differences in the proportion of childlessness among women aged 45–49, as illustrated in figure 3A, B. The proportion exceeded 4% in 25 provinces, and exceeded 5% in 15 provinces.

Figure 3.

Childlessness proportion by province. (A) Childlessness proportion for women aged 45-49 by province in 2010. (B) Childlessness proportion for women aged 45-49 by province in 2020. (C) TFR versus childlessness proportion for women aged 45–49 by province in 2020. Results for some provinces are not displayed due to small observations. Numbers in parentheses represent the number of provinces falling in this range. TFR, total fertility rate.

In provinces with lower fertility, the proportion of childlessness among women aged 45–49 was higher, as shown in figure 3C. Childlessness will contribute to China’s fertility decline in the coming years,²⁰ and this applies at the provincial level. The childlessness proportion for Beijing and Shanghai was 11.92% and 8.18%, respectively, and the childlessness proportions for Jilin, Heilongjiang and Liaoning were 8.59%, 7.9% and 7.11%, respectively. The fertility level for those provinces was relatively low.³⁵

Table 1 indicates that women born in later cohorts had a higher proportion of childlessness, as demonstrated by the proportion of childlessness across various birth cohorts by provinces in 2020.

Table 1.

Childlessness proportion by birth cohort and province in 2020 in China

Province	Birth cohort
	1970–1974	1965–1969	1960–1964	1955–1959
China	5.58	4.55	3.89	3.87
Beijing	11.92	7.91	5.08	3.21
Tianjin	5.45	8.55	4.11	3.53
Hebei	4.67	4.10	2.77	3.16
Shanxi	3.99	2.68	2.96	3.36
Inner Mongolia	5.98	3.74	3.56	3.72
Liaoning	7.11	4.46	3.69	3.18
Jilin	8.59	5.80	4.88	3.75
Heilongjiang	7.90	7.01	5.35	6.83
Shanghai	8.18	5.40	4.22	3.85
Jiangsu	4.34	3.11	2.85	2.91
Zhejiang	5.21	4.30	4.72	3.55
Anhui	3.35	3.24	3.10	3.12
Fujian	4.18	3.62	2.92	2.78
Jiangxi	4.35	3.84	3.70	3.28
Shandong	4.64	3.65	3.56	4.23
Henan	5.30	4.48	4.03	4.55
Hubei	4.12	3.36	2.90	3.00
Hunan	4.67	4.37	2.96	3.98
Guangdong	7.41	6.34	4.43	3.78
Guangxi	6.55	5.73	4.84	5.12
Chongqing	5.31	4.14	3.40	3.88
Sichuan	4.82	4.52	4.39	4.83
Guizhou	5.77	4.16	5.02	3.86
Yunnan	6.12	4.77	4.58	4.28
Shaanxi	4.98	5.49	3.98	3.57
Gansu	4.69	3.55	3.80	2.14

Results for some provinces are not displayed due to small observations.

Decomposition by education level

Figure 4 presents the decomposition results by education level. Three indicators are shown: (1) total change, representing the difference between two proportions of childlessness for women of a certain age between 2020 and 2010; (2) change in the contribution of the change in educational structure, namely, the change in the proportion of women of specific ages with different levels of education; (3) the change in the effect of the difference between 2020 and 2010 in the proportion of childlessness among women of the same age and education level.

Figure 4.

Decomposition of age-specific childlessness proportion difference between 2020 and 2010.

Overall, the age-specific proportion of female childlessness in 2020 is higher than the corresponding proportion in 2010. From the age of 15, the difference increased rapidly, reaching a maximum of 16.40 percentage points at the age of 26, and then began to decline. This means that compared with 2010, childbearing was postponed among the young ages in 2020, but after the age of 26, as women recuperated in childbearing, the proportion of age-specific childlessness in 2020 decreased, and the difference in childlessness proportion between 2020 and 2010 declined. For women over the age of 40, the childlessness proportion was 5 percentage points higher in comparison to the corresponding age in 2010.

Figure 4 shows that there was a significant contribution of changes in educational structure to the total change in childlessness proportion between 2010 and 2020, with the contribution increasing notably from the ages of 15–23. At the age of 23, the contribution is 15.03 percentage points, indicating a rapid increase in women’s education level in the young ages from 2010 to 2020. More women received high school education or above, postponed their childbearing, and brought about an educational structure change. The contribution of educational structure change was 4.41 percentage points at the age of 30, and decreased after the age of 35. Women aged 35 and above were 25 years or older in 2010 and had mostly completed their education by then; therefore, the educational structure did not change significantly for this age group between 2010 and 2020. As a result, they contributed little to the total change in childlessness proportion.

Regarding the difference in childlessness proportion over time for the same education level, there is a negative change in the total difference for the women in the 15–24 age group. As a result, the difference in the proportion of childlessness at the same education level actually contributed a reduction in the total difference. Especially, the difference for women with junior high school education contributed the most to the overall difference. This shows that from 2010 to 2020, for the age group of 15–24, women with lower education, mainly women with junior high school education, did not delay their marriage and childbearing as their counterpart with higher education.

Projecting the proportion for certain groups

With the period age-specific childlessness proportions derived from the 2020 population census and multiple probability distribution models, we fitted and projected the proportion of childlessness for women by residence and education level. Among them, city women and women with college or above education had the highest projected proportion of childlessness.

The proportion of childlessness among city women aged 15–49 in 2020 was fitted with multiple probability distributions. The best fitting result is from the gamma model and is shown in figure 5A. The fitted proportion for 49-year-old women was 9.16%, with a 95% CI (8.20%, 10.12%). We also fitted the period age-specific childlessness proportions for women aged 15–35, and then predicted the childlessness proportion for women aged 36–49. At the age of 49, the predicted proportion was 12.41%, with a 95% CI (10.52%, 14.31%).

Figure 5.

Projected childlessness proportion. (A) City women. (B) Women with college education or above.

For women with a college education or above, the fitting results of gamma model are shown in figure 5B. The fitted childlessness proportion for 49-year-old women was 9.62%, with a 95% CI (9.05%, 10.20%). Based on the fitting of the period age-specific childlessness proportions for women aged 15–35, the projected childlessness proportion for women aged 49 was 10.03%, with a 95% CI (8.71%, 11.35%). This indicates that childlessness remains a significant issue for women in China. The projected proportion also highlights the need for attention to the current and future childlessness of women with higher education levels.

Discussions

China’s fertility rate and annual births have declined rapidly, and the country reached its population peak in 2021, followed by a population decline in 2022. As a critical factor in understanding the decline in fertility, childlessness has also attracted attention.^{7 13 20} Voluntary childlessness is an emerging issue in the low fertility era, presenting a shift in social-cultural diversification and individualisation in some Asian societies. In contrast, involuntary childlessness is an age-long existing reproductive health issue. Due to data limitations, separate analyses could not be conducted for each phenomenon. Instead, we have analysed childlessness as a whole, using census data and 1% population sample survey data, to show changes in childlessness among Chinese women between 2010 and 2020.

The proportion of childlessness at all ages rose between 2010 and 2020, suggesting an increase in delayed parenting and a shift from the traditional family structure. While the trend may be similar to other Asian countries like Singapore, Japan and South Korea, it differs markedly in terms of proportions. The childlessness proportion among women born in Singapore between 1975 and 1980, in Japan between the early 1950s and 1974–1976, and in South Korea in the late 1970s was 28%, 28% and almost 19%, respectively.⁸ Whereas, our results indicate that for Chinese women born between 1970 and 1974, the proportion of childlessness was 5.58%. For China, permanent childlessness is still nascent. While permanent childlessness may have taken a stronghold in other Asian countries, the postponement of childbearing primarily influences the trend in China.

Our result showed that childlessness varies by residence, with the proportion higher in the city than in the township and village. Several possible factors may explain this, including the disparities in education, and economic factors between rural and urban.^{31 32} For voluntary childlessness, the ideological shift from family-oriented to individual autonomy, influencing the individual decision to remain single and childless, can be described as more pronounced in urban than rural areas.³⁰ Urban women spend years schooling and are likely to delay marriage and childbearing and also more likely to be career-oriented than family-oriented.⁵⁰

Furthermore, there are regional differences in childlessness, and the proportion of female childlessness is higher in provinces with low fertility levels. The differences reflected the interplay between fertility levels and economic development. The more economically developed the province, the lower the fertility level and the higher the proportion of female childlessness. Although childlessness and fertility are not always negatively correlated over time,⁵¹ studies on China have confirmed that the decline in fertility level for first births, namely childlessness, may become the main driver of China’s fertility decline in the coming years.²⁰

Childlessness was also shown to vary by educational attainment, with more educated women more likely to be childless. Previous studies have reported mixed results in this regard.^{3 51–56} In places where results have been similar, the reason has been mostly adduced to the high opportunity cost of childrearing,⁵¹ leading more women to postpone marriage and childbearing for the pursuit of education and careers. In the context of China, there is an additional nuance. Women have generally been more likely to marry up the social ladder, which means marrying men who are more educated, accomplished and wealthier than themselves.⁵⁰ For highly educated women, there may be few such men who meet their criteria, creating the phenomenon known as ‘leftover women’. These are women who are still unmarried in their late 30s or early 40s, despite being highly educated and successful. And how does this contribute to childlessness? In China, people are becoming more open and liberal, with premarital sex and cohabitation becoming more common. However, childbearing out of wedlock or single parenthood is still largely frowned on. Therefore, the phenomenon creates a group of involuntary childlessness not due to infertility.

According to the fitting and projection results, the percentage of Chinese women without children will keep increasing. This trend will be most noticeable among certain groups of women, such as those with higher education and those living in urban areas. Additionally, it is expected that the percentage of women without children and the percentage of women who will never have children will continue to rise. This trend could lower China’s birth rate further, threatening population growth. If the course is unreversed, China’s fertility level will inevitably follow the way of other Asian countries like Japan and South Korea.

The consequences of childlessness extend beyond the threat to population growth. Given the vital role of children in providing financial, spiritual and social support to older adults,⁵⁷ childless older adults will be at a significant disadvantage in terms of their economic level, social network, and physical and mental health, and overall life satisfaction.⁵⁸ They will also be at higher risk of morbidity, mortality, isolation and exclusion.⁵⁹ As childless older adults lack support from children, the individual old age support will be partially transferred to the public social security systems. Consequently, the burden on the public social security systems will surge, resulting in higher maintenance costs.

Policy measures to reduce childlessness should be aimed at both involuntary and voluntary childlessness. For involuntary childlessness, the Chinese government has already implemented policies addressing infertility, including expanding assisted reproductive technology (ART) services. The number of assisted reproductive centres has increased over time, growing to over 500 as of 2019.⁸ Furthermore, a concerted effort is being made by considering proposals to gradually include ART in medical insurance, with Beijing already adding 16 ART procedures in its medical insurance as of March 2022.⁶⁰ While these policies are in the right direction, there are opportunities for improvement. Given its relatively high cost, it is imperative to expand the medical insurance for ART to other regions and subsidise the out-of-pocket cost to create more accessibility. Other strategies to adopt include awareness and improved access to treatment. The government might launch awareness and education campaigns about the causes, prevention strategies and available therapies for infertility. This may consist of school-based sex education programmes and other focused outreach initiatives. The ‘Decision on Optimizing the Fertility Policy and Supporting the Long-term Balanced Development of the Population,’ released by the Chinese government in 2021, called for specialised research into enhancing fertility and standardising infertility diagnostic and treatment services. Budgetary support for research might help expand access to high-quality infertility diagnostic and treatment programmes nationwide.

For voluntary childlessness, the above-mentioned decision also stipulates that, by 2025, the government should basically establish a pronatalist fertility policy, significantly improve the level of prenatal and postnatal care services, accelerate the construction of an inclusive childcare service system, and reduce the cost of childbirth, parenting and education in order to encourage childbirth. Although this is not directly aimed at voluntary childlessness, these measures will to some extent eliminate some of the anxiety of women who are ready to choose childlessness. Just like in other developed countries, the level of childlessness in China will rise rapidly, for which the Chinese government must take countermeasures in advance.

Some limitations are associated with our study. First, the study’s definition of childlessness did not explicitly account for adopted or deceased children; thus, caution should be exercised when generalising the findings. Second, the retrospective nature of data collection could introduce recall bias, which may distort data and affect the accuracy of the results. Third, the information collected did not include the timing of childbirth; therefore, we could not conduct a cohort analysis. Furthermore, given the unique characteristics of China, information on birth history was only collected for ever-married women. Also, due to data limitations, childlessness was analysed as a whole without separation into voluntary and involuntary. Finally, we did not conduct a correlation or causality analysis due to the limitation of the population census data. This can be done with survey data in the future.

Conclusions

China’s modernisation process has led to increased childlessness, with education and urbanisation being the key factors. In China, women with higher education levels have a higher proportion of childlessness, as supported by the decomposition results that showed the contribution of increased educational attainment to the rise in childlessness. Although the paper did not show the decomposition result of urban–rural population structure changes on the proportion of female childlessness, urbanisation has also driven the increase in childlessness during the past decade. International experience indicates that in the process of modernisation, changes in lifestyles and fertility concepts tend to increase the number of people who choose not to marry or have children.⁶¹ With the expansion of high education in China, it is projected that 70% of childbearing-age women will have higher education by 2050.⁶² The rapid urbanisation process has also resulted in more women moving from rural to urban areas and changes in lifestyles and fertility concepts, leading to the rise in childlessness.

Data availability statement

All data relevant to the study are included in the article or uploaded as supplementary information.

Ethics statements

Patient consent for publication

Not applicable.

Ethics approval

Data are a sample from China’s population census which were conducted by China’s National Bureau of Statistics. We tabulated the sample of which individuals were not identifiable.