1 Introduction

The decoupling of the connections between “marriage, sex and childbearing” stands out as an important feature of the second demographic transition. Although cohabitation was recognized as an illegal behavior and was uncommon before the end of the 1960s (Trost, 1979), it has grown in popularity and has even become an important type of family, existing alongside the marriage family now (Van de Kaa, 1987; Lesthaeghe, 2010). These developments have been accompanied by more and more nonmarital pregnancies. However an out-of-wedlock pregnancy is increasingly unlikely to prompt a marriage. Between 1970 and 1980, there was an increase in all the Northern and Western European countries in the proportion of out-of-wedlock pregnancies that ended in an out-of-wedlock birth (Höpflinger, 1985). There was only 5% pregnant unmarried women transit to marriage before the birth in America (Lichter et al., 2014). Consequently, the out-of-wedlock birth rate increases quickly in Europe and North America (Kiernan, 2001; Wu, 2015). In some European countries, out-of-wedlock births have begun to outnumber in-wedlock births, becoming the top source of births.

In the wake of China’s reforms and opening up (beginning at the end of the 1970s), urbanization, and modernization, the second demographic transition is well underway in China. In some of China’s most developed cities, people's attitudes and behaviors towards premarital sex, cohabitation, extramarital affairs, divorce, homosexuality, late childbearing, voluntary childlessness, and sexual relations have seen seismic changes (Jiang, 2002). Although it is not clear when exactly the second demographic transition began in China, all of the marriage and childbirth related characteristics of the second demographic transition have taken shape and become increasingly visible in large Chinese cities, especially the megacities (Liu et al., 2012).

With individualism influencing people's sexual attitudes and behaviors over time (Li & Chen, 1993), premarital sex has become increasingly commonplace in China (Wu, 2019), and this has led inevitably to growth in the number of premarital pregnancies. Studies have found that for the marriage cohort from 1987 to 1991 in seven Chinese provinces and municipalities including Shanghai, Shandong, and Guangdong, the proportion of premarital pregnancies was 16.8% (Xu, 1998). In 2013, among migrant couples aged 15–59, the proportion of premarital pregnancies was as high as 30.5% (Qi & Yang, 2015). Another survey conducted in 2011 found that among migrants who were born in 1980 or later and who were married and had children, 42.7% conceived their first child before marriage (Song et al., 2012). Among unmarried adolescents in Shenzhen, however, the incidence of premarital pregnancy was less than 5% (Huang, 2016; Xu, 2003). Li Ding and Tian Siyu found that premarital pregnancies were most prevalent in women around 20 years of age (Li & Tian, 2017), whilst Zhong et al. (2013) and Xu & Shi (2019) found that premarital pregnancy was becoming more common among young teens.

Nevertheless, although premarital pregnancy occurs more often in certain provinces (Liu et al., 2019), and despite the fact that since the 1980s, attitudes towards sex and sexual behavior have become considerably more open, both urban and rural Chinese are still conservative on the whole (Wu, 2019). Whilst the overwhelming majority of rural residents remain intolerant of premarital sex (Liu, 2010), urban residents display relatively diverse attitudes towards premarital sex (Wu, 2019).

A few scholars have attempted to probe the outcomes of premarital pregnancies. In the late twentieth century, most women chose abortion to terminate a premarital pregnancy (Wang, 1999). A survey of women having premarital check-ups in Laibin (Guangxi Province), Foshan (Guangdong Province), and Longhua New District (Shenzhen Municipality) found that the incidence of out-of-wedlock births among these women ranged from 10 to 17% (Chen, 2010; Jian et al., 2015; Yang et al., 2017).

Studies like those noted above have helped uncover the facts of premarital pregnancies and out-of-wedlock births in certain regions and groups, yet these studies sometimes reach different conclusions on the incidence of premarital pregnancies and out-of-wedlock births, and data in some studies on sexual attitudes seems inconsistent with the behavioral data on premarital pregnancies. Is premarital pregnancy really commonplace among Chinese women? Are there any age-specific characteristics and patterns? What percentage of premarital pregnancies eventually end in out-of-wedlock births? Answers to these questions would help shed light on the choices available to Chinese women choices when they make decisions about marriage and childbearing, and would provide a strong basis for evaluating the current status and future development of China's fertility rate.

To understand the situation with respect to premarital pregnancy, this study first takes data from the China Fertility Survey 2017 to determine the incidence of premarital pregnancies from the late 1970s to 2017 in Chinese women born between 1957 and 2002. Next, it takes an in-depth look at the decisions Chinese women make when facing choices about marriage, pregnancy and childbearing. Lastly, by comparing out-of-wedlock births in China and in OECD member states, the study discusses the behavioral choices of Chinese women with respect to sex, marriage and childbirth, and expounds on the different characteristics of the second demographic transition in China and the West.

2 Definitions and data

In this study, premarital pregnancy refers to the pregnancy of an unmarried woman. Premarital pregnancies can be divided into two broad categories: (1) pregnancies in women unmarried at the time of conception and remaining unmarried and (2) pregnancies in women unmarried at the time of conception and married after that. Determining if a pregnancy is type 1 or type 2 requires comparing the date the pregnancy ended with the date of the first marriage. Given the low likelihood of preterm births with less than 32 gestational weeks and to avoid overestimation, this study has adopted the assumption of Li and Tian (2017) and Xu (1998) that a pregnancy generally lasts for 8 months or more, With this assumption in mind, we can refine our definition as follows: (1) pregnancies in women unmarried at the time of conception and remaining unmarried now; (2) pregnancies in women unmarried at the time of conception and married after the pregnancies, with an outcome of live birth, and a time interval of 7 months or less between the end date of the pregnancy and the date of the first marriage; and (3) pregnancies in women unmarried at the time of conception and married after the pregnancies, with an outcome of non-live birth, and the end date of the pregnancy coming prior to the start date of the first marriage.

The China Fertility Survey 2017 (the Survey), conducted by the former National Health and Family Planning Commission, produced nationally representative fertility data. Women aged 15–60 (i.e., women born between 1957 and 2002) were interviewed during the first half of 2017. With a sample of 249,946 people, the Survey was an effort to unveil the reproductive histories of women born in the period 1957–2002. Since most women born in the late 1950s got married and gave birth to children in the late 1970s and early 1980s, a review of the fertility histories presented by the Survey can help reveal the incidence of premarital pregnancies in Chinese women from the late 1970s to 2016. Since the age of menarche can be as early as 12 years old, first pregnancies occurring before 12 years old were excluded from our calculations. Eventually, a valid sample of 249,922 women was obtained.

3 Results

3.1 20% of Chinese women aged 15–60 had experienced premarital pregnancy

Using the values for marital status, pregnancy status, starting time of pregnancy and starting time of first marriage, the pregnancy status of each woman in the sample population can be assigned to one of five categories: premarital pregnancy, marital pregnancy, never pregnant, uncertain premarital pregnancy, and uncertain pregnancy. Uncertain premarital pregnancy is defined as: (1) pregnancies with an outcome of non-live birth, and an interval of 7 months or less between the ending time of pregnancy and the starting time of the first marriage; and (2) pregnancies for which the end time of the pregnancy and start time of the pregnant woman’s first marriage are unknown. Uncertain pregnancy is defined as pregnancies reported by unmarried Survey respondents (cohabiting respondents) who also claimed zero pregnancies and failed to complete their pregnancy history. Furthermore, if a respondent gave the number of her pregnancies and the ending time of each pregnancy but answered “No” to the question “Have you been pregnant”, the respondent was deemed to have been pregnant, and this was followed by determination of whether such pregnancy was a premarital one. The weighted statistics are shown in Table 1.

Table 1 Pregnancy status of respondents aged 15–60 from 2017 Survey data (unit:%)
Full size table

Statistics from the 2017 Survey data show that 21.5% of the 249,922 female respondents aged 15–60 (i.e., those born in the period 1957–2002) had experienced at least one premarital pregnancy.

At the time of the Survey, 91.9% of women who had experienced premarital pregnancy, a “deviant behavior” in the traditional sense, were married, and only 8.1% were still unmarried. This was due to the culture of universal marriage in Chinese society.

At the time of the Survey, 22% of women aged 15–60 were unmarried, and 7.9% of them had experienced premarital pregnancy. Of women aged 15–60, 78% were married or divorced, and 25.3% of them had experienced premarital pregnancy, whilst as high as 69.9% had experienced their first pregnancy after marriage. In general, the incidence of premarital pregnancies was higher in married (divorced, widowed) women than in women who were unmarried at the time of the Survey.

As mentioned above, 25.3% of married (divorced, widowed) women had experienced premarital pregnancy. This proportion is higher than Xu's (1998) inference from a1992 survey of married women in seven Chinese provinces and municipalities, suggesting that—at least for married women—the incidence of premarital pregnancy has risen over time.

The incidence of premarital pregnancy in migrant women (i.e., women living in a locale other than that of their household registration) was 6 percentage points higher than that of women living in the locale of their household registration (hukou), a finding in accord with the results of existing research. Specifically, this study found that 28.7% of migrant women had experienced premarital pregnancy. This percentage is consistent with the 28.1%, 28.9% and 30.6% extrapolated by Li and Tian (2017) from migrant monitoring data for 2011, 2013 and 2015 respectively, and the 30.5% extrapolated by Qi and Yang (2015) using a broader statistical caliber. All these extrapolated results suggest that the incidence of premarital pregnancy was at least 28.7% to 30% for migrant women who participated in the 2017 Survey. With the passage of time, the incidence of premarital pregnancy has also increased for migrant women.

3.2 Higher incidence of premarital pregnancy in younger birth cohorts and women born after the reforms and opening up began

Most 50-year-old woman have passed their reproductive years and, of course, have completed all of their pregnancies. Therefore, this study takes women who were 50 years old in 2017 (i.e., those born in 1967) as the baseline birth cohort and then examines changes in the incidence of premarital pregnancy for other birth cohorts relative to the baseline cohort.

As shown in Fig. 1, in birth cohorts for 1986 and earlier, the incidence of premarital pregnancy gradually rose as the birth cohorts got younger.

Fig. 1
figure 1

Age-specific pregnancy status of women aged 15–60. Number of obs = 249,922 Pearson: Uncorrected chi square (180) = 1.74e + 05 P = 0.0000. All figures were obtained through weighting of data for 2017

Full size image

There are four segments with distinctly different incidences of premarital pregnancy: (1) in the first segment, the incidence of premarital pregnancy gradually edged up in the birth cohorts from 1992 to 2002, but remained below the incidence of premarital pregnancy in 50-year-old women; (2) in the second segment, the incidence of premarital pregnancy in the 1979–1991 birth cohorts surpassed that of 50-year-old women; (3) in the third segment, the incidence of premarital pregnancy in the 1966–1978 birth cohorts fluctuated along the horizontal line representing the incidence of premarital pregnancy in 50-year-old women; and (4) in the fourth segment, the incidence of premarital pregnancy in the 1956–1965 birth cohorts dropped below the horizontal line.

Women within the second segment were still in their prime reproductive years, and thus the incidence of premarital pregnancy in that group might continue to rise until these women reach the end of their reproductive years. In this segment, the incidence of premarital pregnancy in younger birth cohorts exceeded that in older cohorts aged 50 years or above.

From the perspective of temporal characteristics, the lowest incidence of premarital pregnancy was observed in women falling within the fourth segment, which represents the 1956–1965 birth cohorts. Enacted in 1950 to put an end to the feudal practices of arranged and forced marriages, the Marriage Law of the People's Republic of China advocates voluntary marriage, monogamy and equal rights for men and women, and fully protects the legitimate rights and interests of women and children. People have been freed from the shackles of outdated thoughts and the new idea of independency in marriage has spread. This might have contributed to the slow increase in the incidence of premarital pregnancy. However, the principal reason for the slow increase was that a traditional and puritanical asceticism was in the ascendant at that time; very few Chinese had even rudimentary notions of individualism.

The third segment represents cohorts born during the turbulent years between 1965 and 1978. For these cohorts, the incidence of premarital pregnancy hovers around 0.236, and shows no obvious signs of going up or down. Given the chaotic social atmosphere during those years, it is possible that people born during this period were more cautious and avoided any kind of activity that might be considered improper or over the line. Therefore, the rising incidence of premarital pregnancy in the fourth segment failed to continue into the third segment.

Women in the second segment were born in the period 1979–1991, i.e., they were born after China's reforms and opening up began in the late 1970s. Coming of age in the vibrant atmosphere created by the reforms and opening up and the emergence of a market economy, they witnessed the take-off of China’s economy and the introduction of liberalism, individualism, and sexual liberation from the West. These developments are key to understanding the high incidence of premarital pregnancy in the second segment.

It is worth noting that 31-year-old women (i.e., those born in 1986) had the highest incidence (33.1%) of premarital pregnancy of any group, meaning that one out of three women born in 1986 experienced a premarital pregnancy. Given that 31-year-old women were still in their prime reproductive years in 2017 and that there were women in this group who had experienced zero pregnancy and zero childbirth at the time of the Survey, it is possible that more than 33% of these 31-year-old women will have experienced zero pregnancy and zero childbirth by the time when they reach the age of 50 in 2036.

Women in the first segment were born in the period 1992–2002, and were 15 to 25 years old at the time of the Survey. Because young people spend more years in school than in the past, many of this group were still students in 2017, meaning that premarital pregnancies were less likely to occur. Therefore, the incidence of premarital pregnancy was low in these birth cohorts at the time of the Survey.

3.3 The rising in the age at first premarital pregnancy

According to the Marriage Law of the People's Republic of China (1980), the legal age of marriage is 22 years old for men and 20 years old for women. Because a 20-year-old woman can legally marry and have children, this study has divided the our sample population into four groups to determine the age of women at the first premarital pregnancy: aged 12–19, aged 20–24, aged 25–29, aged 30–34, and aged 35 and above.

Though some scholars believe that premarital pregnancy is gradually become more common among young teens, this study shows that the incidence of first premarital pregnancy in the 12–19 age group displayed an inverted V-curve. In fact, the incidence of first premarital pregnancy gradually moves downward in younger cohorts born after 1970, but spikes quickly at age 20 and beyond, and especially at age 25 and beyond (Fig. 2).

Fig. 2
figure 2

Age structure of first premarital pregnancies by birth cohort. Number of obs = 59,770 Pearson: Uncorrected chi square (24) = 3181.8218 P = 0.0000. Note: The cohorts born in the period 1985–1989 and after 1990 were still in their prime reproductive years when the Survey was conducted in 2017. The incidence of first premarital pregnancy at lower ages will keep going down as the number of premarital pregnancies increases at higher ages

Full size image

Women in their 20 s are well past the age of majority, so they are unlikely to lack knowledge of contraceptive methods. The surge in the incidence of premarital pregnancy in cohorts born after 1970 (who were at least 20 years old at the time of the Survey) is by no means a consequence of unwitting mistakes.

Surprisingly, in the 1956–1964 birth cohorts, the incidence of premarital pregnancy in women aged 20 and older was higher than in the 1965–1969 and 1970–1974 cohorts. This might be a result of the “later, longer and less” fertility policy implemented in the 1970s. In the period 1970–1980, young men and women who planned to marry had to wait for the approval of their workplaces, and hence during these years, women experiencing their first premarital pregnancies were often older.

From a temporal perspective, with the passage of time, both the average age of women at first premarital pregnancy and the average age at first marital pregnancy increased, with the average age of women at first marital pregnancy increasing more than the average age at first premarital pregnancy.

As shown in Fig. 3, from 2006 to 2016, the average age at first premarital pregnancy remained stable in the period 2006–2010 before reaching 22.73 years in 2016—an increase of just more than 1 year of age in a six year period.

Fig. 3
figure 3

Average ages at first premarital pregnancy and first marital pregnancy in women aged 15–49 in the period 2006–2016

Full size image

The average age at first marital pregnancy also showed a similar trend, although there was a greater increase in the period 2011–2016. This also led to a widening of the gap between the average age of women at first premarital pregnancy and the average age at first marital pregnancy—from a gap of 2.27 years to one of 3.21 years in just 11 years.

The increased average ages of women at first premarital pregnancy and first marital pregnancy point to the fact that the decision to delay the time of the first birth was commonplace among Chinese women. With Chinese women generally postponing childbirth, their age at first marital pregnancy increased.

3.4 Significant decline in the incidence of multiple premarital pregnancies

Our findings show that the incidence of multiple premarital pregnancies dropped significantly, compared to the spike in the incidence of single premarital pregnancies. This phenomenon is more noticeable in younger cohorts.

As shown in Fig. 4, the frequency of premarital pregnancies decreased gradually as the cohorts became younger. In women who were born before 1964 (inclusive) and experienced premarital pregnancy, the average frequency of premarital pregnancies was 1.48, compared to 1.29 in the 1980–1984 birth cohorts, a decline of 13.2 percentage points.

Fig. 4
figure 4

Average number of premarital pregnancies by birth cohort. Number of obs = 59773 Pearson: Uncorrected chi square (66) = 904.4673 P = 0.0000. All figures were obtained through weighting of data for 2017

Full size image

Of the women who had experienced premarital pregnancy, 78.7% had had one premarital pregnancy and 13.96% had had two premarital pregnancies, together accounting for 92.7% of the women how had had premarital pregnancies. The proportion of women who had experienced more than two premarital pregnancies shrunk. The incidence of four or more premarital pregnancies dropped from 5.67% in the cohorts born before 1964 (inclusive) to 1.84% in the 1980–1984 cohorts, a decrease of 67%, whilst the incidence of three premarital pregnancies also fell from 7.17% in the cohorts born before 1964 (inclusive) to 4.04% in the 1980–1984 cohorts, a decrease of 43%.

This decrease in the incidence of multiple premarital pregnancies is associated with the reduction in the number of high parity births on the one hand and the prevalence of contraceptives on the other. Though there are no national data for the using percentage of condoms among unmarried women by birth cohort and year, but according to the 2017 Survey, those unmarried (cohabiting) respondents got a higher percentage of using contraceptives with the respondents becoming younger. The percentage was 61.5% for those unmarried (cohabiting) respondents born from 1975 to 1979, and the number increased to 72.67% for those born from 1980 to 1984, and 74.79% for those born from 1985 to 1989. And the study (Zou et al., 2018) has also shown, among married reproductive aged women, the percentage of using the condoms had risen to 16.1% in 2016, which was nearly two times higher than in 2010, the using percentage of the condoms among unmarried women should be higher because of the convenience and availability of condoms.

3.5 The “cohabitation-pregnancy-childbearing” relationship: cohabitation is associated with a high incidence of premarital pregnancy and a high rate of live birth

Some scholars argue that cohabitation is a kind of “trial marriage” (Yu and Xie, 2017)—that is, a test of what married life is like before entering into a legally recognized marriage. Cohabitation leaves women in such relationships fully exposed to the risk of pregnancy, thus pushing up the likelihood of getting pregnant. Pregnant women in cohabitation relationships are more likely than single women to carry the child to term and give birth.

How likely are women in a cohabiting relationship to get pregnant out of wedlock? According to the 2017 Survey, 92.6% of unmarried women who cohabited experienced premarital pregnancy, compared to only 2.7% of unmarried women who did not.

Cohabitation also drives up the likelihood of a women carrying her first premarital pregnancy to term and having a live birth. Only 70% of premarital pregnancies in unmarried non-cohabiting women culminated in a live birth, compared to 93.4% in unmarried cohabiting women. As high as 96.3% of marital pregnancies resulted in a live birth. Cohabitation makes it highly likely that premarital pregnancies have an outcome akin to that of marital pregnancies (Fig. 5).

Fig. 5
figure 5

Marital status vs. pregnancy status. Number of obs = 249922; Uncorrected chi square (12) = 2.36e + 05 P = 0.0000

Full size image

The abortion rate in unmarried non-cohabiting women reached 25.7%, a rate much higher than the rates for unmarried cohabiting and married women. Considering that some unmarried non-cohabiting Survey respondents might have been unwilling to report their pregnancy and abortion history, the actual abortion rate for this group could be higher.

3.6 The “pregnancy-marriage–childbearing” relationship: premarital pregnancy is associated with a higher rate of in-wedlock births

This study examines the relationship between pregnancy, childbirth and marriage. Based on the outcome (live birth or non-live birth) of premarital pregnancies in married (or divorced) women and the interval between the date of conception and the date of marriage, we have compiled statistics for the following four outcomes of premarital pregnancies in married (or divorced) women: (1) marriages, either shotgun or consensual, which resulted in an in-wedlock live birth, including marriages during a first premarital pregnancy, and marriages during a premarital pregnancy following one or more premarital pregnancies which all ended in abortion; (2) marriages that occurred after childbirth, i.e., the woman was pregnant and gave birth out-of-wedlock; (3) premarital pregnancies ending in abortion and followed by marriages, marital pregnancies and in-wedlock live births; and (4) premarital pregnancies in which the date of conception or marriage could not be determined.

In cases (1) and (3), although the woman became pregnant out of wedlock, the first childbirth took place inside marriage.

The Survey found that nearly 70% of women who had experienced premarital pregnancy gave birth to their first child inside marriage, with outcome (1) accounting for 59.64% and outcome (3) accounting for 7.17%—the two combined accounted for 66.81%.

Marriages, either shotgun or consensual, can be further divided into two cases: marriages in which the bride was pregnant for the first time (accounting for 56.19% of the 59.64% noted in the paragraph above), and marriages in which the bride's premarital pregnancy followed one or more previous premarital pregnancies (accounting for the remaining 3.45%). In the first of these cases, according to Chinese traditions, the man marrying the pregnant woman is the man who made her pregnant. There are two possibilities: (1) the couple previously had no definite plan to marry, and decided to marry because of the premarital pregnancy; and (2) the couple did have plans to marry, and the premarital pregnancy was not entirely unintended. In both outcomes, premarital pregnancy catalyzed marriage. In the second case, premarital abortion may result because the couple failed to marry before the woman became pregnant. It is also possible that the woman marries a man other than the man who originally made her pregnant.

Another 32.24% of premarital pregnancies culminated in out-of-wedlock births, with 30% of these unwed mothers marrying before their children reached one year of age. It was not easy to determine if the men these unwed mothers married were the partners that made these women pregnant for the first time.

Despite their significant differences, outcomes (1) and (3) are both similar to the extent that both outcomes culminate in in-wedlock births; childbirth takes place after marriage and the first child is born within wedlock. For Chinese people, “family” is the linchpin of Chinese traditional culture and social order, and an individual’s social identity comes, first and foremost, from the family. Bearing children and carrying on the family’s legacy are the most important purposes and tasks of the family. Such social and cultural traditions are enormously powerful incentives behind the transformation of premarital pregnancies into in-wedlock births.

This is quite at odds with the Western societies. Taking the United States as an example, the 2006–2010 National Survey of Family Growth conducted by the National Center for Health Statistics shows that 18% of premarital pregnancies transformed into cohabitation before the child was born, and only 5% led to marriage (Lichter et al., 2014). Compared with the high transformation rate in China, the gap is huge (Table 2).

Table 2 Childbirth-marriage interval in married (divorced) women who had experienced premarital pregnancy
Full size table

3.7 Out-of-wedlock births accounted for less than 9% of all live births

During the second demographic transition in Western societies, out-of-wedlock births have provided a vital source—or even a major source—of live births. The situation in China is, however, are quite distinct from that of Western societies.

Based on the marital status at the time of a pregnancy, the interval between the ending time of that pregnancy and the starting time of marriage, and the type of live birth, this study has divided live births into the following types of cases: (1) out-of-wedlock birth: a live birth out of wedlock; (2) premarital birth: live birth by a married (or divorced) women, with the ending time of pregnancy being earlier than the starting time of first marriage; (3) in-wedlock birth of a child conceived before marriage: live birth by a married (or divorced) women, with an interval greater than 0 but less than 8 months between the ending time of pregnancy and the starting time of first marriage; (4) in-wedlock birth of a child conceived after marriage: live birth by a married (or divorced) women, with an interval greater than or equal to 8 months between the ending time of pregnancy and the starting time of first marriage; and (5) unknown: live birth by a married (or divorced) women, for which the ending time of pregnancy or starting time of first marriage was unknown.

Cases (1) and (2) pertain to out-of-wedlock births, and cases (3) and (4) pertain to in-wedlock births.

Among all 396,022 live births, out-of-wedlock births accounted for 8.96%, in-wedlock births accounted for 91.02%, and the remaining 0.01% were case (5).

In all out-of-wedlock births, which accounted for less than 9% of all live births, 2.73% of the women who gave birth were unmarried at the time of the Survey, and 6.23% were married (or divorced) at the time of the Survey.

By year, the incidence of out-of-wedlock births peaked in 2009 at 10.4% of all live births, followed by a steady decline thereafter, dropping to 4.3% in 2016.

The fall in out-of-wedlock births since 2009 is closely linked to the continued decline in first births in recent years (Chen & Duan, 2019). As shown in Fig. 6, 50% of out-of-wedlock births were the women’s first births. Constraints such as late marriage and decisions to delay childbearing have led not only to the slump in first births, but also contributed to the downward slide in out-of-wedlock births year by year (Fig. 7).

Fig. 6
figure 6

Out-of-wedlock births in women aged 15–60. Total live births = 396,022

Full size image
Fig. 7
figure 7

Proportion of out-of-wedlock births to all live births in women aged 15–49 (2006–2016). Total live births = 106,463

Full size image

3.8 Comparison of out-of-wedlock births between China and major OECD member states

This study examines the relative level of out-of-wedlock births in China from a global perspective by comparing out-of-wedlock births in China and major OECD member states. As early starters, OECD member states are relatively advanced in terms of socioeconomic development, with major OECD member states showing signs of the shifting “marriage–childbirth” relationship during the second demographic transition.

OECD member state varied largely in terms of the proportion of out-of-wedlock births. In the first group of countries represented by Chile, the five Nordic countries, and France, the proportion of out-of-wedlock births was above 50%, meaning that more than half of the children born in 2016 were born out of wedlock. Chile in South America, where 72.7% of all births in 2016 were out of wedlock, had the highest proportion of out-of-wedlock births (Fig. 8).

Fig. 8
figure 8

Changes in the proportion of out-of-wedlock births in China and OECD member states

Full size image

In the second group of countries represented by long developed economies such as the United States, the United Kingdom, and Germany, and several other Western European countries, the proportion of out-of-wedlock births ranged from 30 to 50%.

In the third group of countries, which were mainly Eastern European countries that have undergone seismic institutional transformations, the proportion of out-of-wedlock births soared to more than 30% after2000, with Slovenia and other countries topping 50%.

In the fourth group of countries, the proportion of out-of-wedlock births was below 30% in 2016 (Fig. 9). These countries include, in part, nations with strong religious traditions, such as Poland, Switzerland, Greece, and Turkey. Italy is an outlier that embraced secularization in the mid-1980s, and saw its proportion of out-of-wedlock births rise quickly and surpass 20% after2000. Japan and South Korea, both in this group, are in the Confucian cultural orbit and had proportions of out-of-wedlock births lower than 10%. Countries in the fourth group are characterized by a low initial proportion of out-of-wedlock births in 1960, and grow that an extremely slow pace thereafter. These countries had lessthan5% increases in the proportion of out-of-wedlock births in the 20 long years from 1960 to 1980. Although the proportion of out-of-wedlock births edged up after 2000, the uptick remained small. Looking at growth in the proportion of out-of-wedlock births for countries in this group over the past 50 years, Poland and Switzerland clearly outpaced countries adhering to Confucian traditions.

Fig. 9
figure 9

Source of data for OECD member states: http://www.oecd.org/els/family/database.htm

Comparison of out-of-wedlock births in China and OECD members states with proportions below 25% (1960–2016).

Full size image

China also ranks among countries with the lowest proportion of out-of-wedlock births. Although China, Japan and South Korea all have strong Confucian cultural traditions, the proportion of out-of-wedlock births among Chinese women was slightly higher than that among Japanese and South Korean women; China was more closely aligned to the proportions in Greece and Turkey.

4 Conclusion and discussion

The mounting incidence of premarital pregnancies points to a more pronounced separation between sex and marriage, which, according to the analysis in this study, has become an increasingly evident trend for Chinese women. This separation is manifested not only in the prevalence of premarital sex, but also by the more than 20% of Chinese women who have experienced premarital pregnancy, with some younger women even experiencing multiple premarital pregnancies.

In earlier studies, premarital pregnancy was considered “deviant behavior” that was found mainly among teenagers. Now, the incidence of premarital pregnancy has increased rapidly in adult women over 20 of age, and this has occurred in parallel with a rapidly declining incidence of multiple premarital pregnancies. This suggests that most premarital pregnancies are not unintentional behaviors, but rather a reflection of how women's attitudes toward marriage and sex are changing, and how China’s reforms and opening up, the development of a market economy, and the emergence of individualism have reshaped people's attitudes toward sex and marriage. This change in attitudes well coincides with characteristics of the second demographic transition.

Traditional Chinese beliefs about marriage have, nonetheless, refused to make room for this change, and the “marriage–childbirth” bond remains strong. More than 70% of women who have experienced premarital pregnancy would still prefer to have the first birth in wedlock. Although they have defied the tradition of “marriage first and sex later”, they still hold to the traditional idea that “marriage comes before childbirth,” and want to ensure that their first child is born within marriage. At the same time, more than 90% of live births in China occur within marriage. Compared with OECD countries—be they Nordic countries, Western European countries, countries separated from the former Soviet Union, or countries with strong religious roots—China has an extremely low incidence of premarital birth.

In fact, China manifests a variation of the second demographic transition, one that incorporates traditional Chinese values of family, and that features at least partial liberation from the “sex-marriage–childbearing” connection. Although the traditional value that puts “marriage first and sex later” is fading, a rupture of the “marriage–childbirth” bond has not followed. Whilst childbirth remains an important and meaningful event to Chinese families and Chinese society, the traditional “marriage–childbirth” bond also remains intact.

Renowned Chinese anthropologist and sociologist Fei Xiaotong pointed out the functional significance of marriage for childbearing by arguing that “marriage…is to establish dual-line parenting” (Fei, 2013). Although modernity constantly penetrates Chinese society, traditional family values and the pivotal role played by family in Chinese social structures and institutions assure the continuation of the functional significance of marriage for childbearing.

Citing an old Chinese saying “self-cultivation, family harmony, state-governing, and world peace (修身齐家治国平天下) ”, JinYaoji and Liang Shuming argued that family is the key node connecting self-cultivation and efforts to rule the country and bring peace to all (Jin, 2002). Whilst state-governing might be an impossible dream for individuals and self-cultivation hinges largely on self-discipline, family is the most important community for one to be in contact with. For a family to run well, individuals must submit to the moral norms practiced by the family.

Family has been endowed with many functions in the Chinese social system, and stands out as an important part of the nation and society. An individual first acquires his/her identity from the family before getting his/her social status. From the time of birth, an individual is assigned a position in the family before finding a place in the social system. After death, an individual takes a position within the ancestral temple. Even in modern times as ancestral temples have gradually disappeared, the core symbolic significance and the cultural image have yet to perish.

Although childbearing is a personal act, its significance goes way beyond the reach of an individual, a couple or even a small family. Not only is childbearing bound up with reproduction, it is also about passing on a legacy to future generations.

From this point of view, individualism and liberalism do have an impact on individual activities, such as sexual behavior and pregnancy. But when it comes to childbirth, the individual's will must give way to a family decision. Whether it is to “give the baby a sweet home” or “let the baby become a part of a big family”, or to avoid the embarrassment of having an out-of-wedlock child, an overwhelming majority of premarital pregnancies have translated into in-wedlock births; it is extremely rare to see out-of-wedlock births in China. Unmarried women not in stable cohabitation relationship with a partner whom they can marry also have a higher abortion rate. These outcomes demonstrate the on-going influence of China’s family culture, which has also given rise to institutional norms that hamper out-of-wedlock births to a considerable extent.

Therefore, the choices Chinese women face when making decisions about marriage, pregnancy and childbearing combine a mix of characteristics from both the second demographic transition and traditional Chinese patterns. Consequent to the spread of liberalism, individualism and sexual liberation, premarital pregnancies are gradually occurring more frequently in China and are making their way from being an issue of concern among juveniles to an increasingly common choice. However, given the profound weight China’s family culture, Chinese people continue to maintain the inherent link between childbearing and marriage. Pregnancy may occur before marriage, but childbearing most likely occurs inside of marriage, because when bearing children, one must take into consideration both family demands and social and cultural norms. Compared with the traditional marriage and childbearing pattern of dating > marriage > sex/pregnancy > childbirth, the more recent pattern of dating > sex/cohabitation (premarital pregnancy) > marriage > childbirth" does indeed mark an enormous change, yet even the newer pattern adheres to the traditional order of marriage coming before childbirth. Premarital cohabitation and premarital pregnancy mark preparatory or transitional stages rather than alternatives to marriage and childbearing.