The second demographic transition: also a 21st century Asian challenge?

Abstract

This introduction aims at placing the unfolding sub-patterns of the Asian “Second Demographic Transition” (SDT) in a global context by contrasting them with those of societies with other than patriarchal histories. Firstly, fertility transitions to below-replacement level can be achieved as part of the first “altruistic” transition without any SDT traits being present. Secondly, Asian societies are by no means immune to genuine SDT developments, as illustrated by the emergence and spread of pre-marital cohabitation. Thirdly, the SDT cohabitation pattern is still conservative: it is followed by marriage, pregnancies result in shotgun marriages or abortions, and parenthood within consensual unions remains rare. Also divorce rates are low. But it is also argued that all Asian cases are still at the beginning of the possible SDT evolution or have barely started it, and that old ways can die off rather quickly with the succession of generations. Finally, it is shown that the cultural component, i.e. the “Willingness” condition, can act as a bottleneck slowing down the transition to a new pattern of behaviour. We therefore illustrate on a global scale how the spread of cohabitation is part of a broader ethical revolution stressing individual rather than societal discretion in matters of life and death. On the basis of these profiles we expect stronger resistance to SDT patterns of partnerhip formation in Hindu and Muslim societies.

In 1985 the Dutch sociology journal “Mens en Maatschappij” (Man and Society) asked D.J. van de Kaa and myself to edit a special issue on the topic of family change. Both our countries, the Netherlands and Belgium, had lived through the “roaring sixties” with not only student protests and hippies, but more profoundly, also through an era of rapid education expansion for both sexes, the end of the breadwinner model as the standard, a major secularisation wave, a political reaction against the system of pillarisation typical for plural democracies, and the rise of post-materialist and expressive values stressing self-development and individual autonomy. In addition to the political revolution, there was also an ethics one, a gender revolution, and a major contraceptive breakthrough. It was in this climate that van de Kaa and I wrote the introductory chapter of the 1986 book volume of that journal. But when everything was done, we did not have a title, and in an unguarded moment I suggested “Two demographic transitions ?” followed by a question mark (Lesthaeghe & van de Kaa, 1986). The title, then in Dutch, was accepted, and shortly thereafter van de Kaa wrote “Europe’s second demographic transition” in English (van de Kaa, 1987). The question mark was gone. At that time we had not the fogiest idea that the concept would catch on, but more strikingly that the SDT would spread, be it under different forms, to many more areas in the world other than Northern and Western Europe.

At present, some 36 years later, we are now asking the same question as in 1986, but for Asian societies: is there a second demographic transition? We have to be cautious in answering it for various reasons. First, the notion of the SDT has two meanings. The SDT refers to the actual demographic changes in the life course pertaining to partnership formation and dissolution, fertility and household composion. As such it can be monitored statistically. But the notion also refers to the theory behind it. And of course that is a different matter. To get the record straight, although we focussed on the ideational changes that accompany the SDT demographic developments, we by no means wanted to do so at the detriment of socio-economic factors. And even less so at the detriment of ethno-historical traditions of belief and social organisation. In fact, I contributed to the re-crafting of the classic “Demographic Transition Theory” by the Princeton European Fertility Project, and the formulation of A.J. Coale’s “Ready, Willing, and Able” paradigm (Coale, 1973). These analyses of the historical fertility transition had shown that industrialisation and urbanisation were by no means the only driving forces, but that three conditions had to be satisfied for the adoption of voluntary fertility control. Firstly, the new behaviour must be in one way or another beneficial (economic advantage = Ready), secondly, it must be culturally acceptable (religious or moral acceptability = Willing), and, thirdly, there must be means to accomplish it (technical means, legal environment = Able). Moreover, the three conditions must be met simultaniously, because if one is evolving more slowly than the other two, the entire transition will be stopped or slowed down.¹ Applied to the SDT, for instance, cohabitation must entail extra advantages over singledom or marriage, it must be ethically or religiously acceptable, and it must be permitted by law. Similarly, fertility postponement must be beneficial materially or psychologically, it must be accepable, and there must be efficient methods of contraception.

The discussion whether Southern, Central and Eastern Europe experienced the SDT abated after an initial reaction of the type “not us, we’re different”. Of course context specific effects arose as historical, socioeconomic and cultural determinants are always present.² But, the overall set of SDT demographic characteristics is now widely present in all Western countries, and all of them experienced the ideational shifts as well, such as rising gender equity, greater individual autonomy, secularisation, and rising tolerance for non-conformism in ethical matters.

As far as Latin American and the Caribbean populations concerned, all of them had to varying degrees a history of consensual unions among the poorer classes and specific ethnic subgroups. But since the 1970s they experienced, be it at different paces, a real cohabitation boom, with even university graduates increasingly moving through the cohabitation phase or remaining permanently in that state. This paralelled further increases among the poorer classes as well (Esteve & Lesthaeghe, 2016). But until recently the other key SDT characteristic, i.e. fertility postponement, was missing. Only very recently has the postponement trend emerged in Argentina and Uruguay as a sub-pattern, but not yet as a nationwide feature. Possibly this is currently also materialising in other large urban areas among educated segments, but it is still too early to come with a firmer answer.³ Hence, the Latin American experience is one of only a partial SDT, with the main ingredient of cohabitation having both historical roots in the so called “pattern of disadvantage” (POD) and in the recent SDT-type combined.

And the Asian far more patriarchal societies ? Aparently the most industrialised ones such as Japan, South Korea, Singapore, or mainland China experienced their fertility transitions early and swiftly. Evidently all three conditions R,W and A were met for fertility control to widely gain acceptance. But cohabitation ? The formation of consensual unions goes very much against the basic principles of patriarchy, such as women leaving the parental home upon marriage and move in with in-laws (patrilocal or uxorilocal marriage), endogamous marriage, if not within the own kinship group, then within the same class or caste, a preference for boys and priviliges for them, and arranged or brokered marriages. Cohabitation is based on individual choice par excellence and hits patriarchal rules head on. As these rules are part and parcel of all Asian religions, Judaïsm, Islam, Hinduism, Buddhism or Confucianism alike, the obstacle to this particular element of the SDT must be formidable.

At the start of the Century, also all Asian colleagues were convinced that cohabitation was “not for us, we are different”. Different indeed, but when the Japanese newspaper Mainichi Shimbun started probing, the existence of premarital cohabitation was revealed. Shortly thereafter the Japanese official fertility survey of 2009 confirmed that 41 percent of women aged 20–40 in a union had experienced cohabitation, and that no less than 24 percent of university graduates had as well.⁴ At that time, we had no idea what was happening in China, but the 2014 Chinese Family Panel Study brought to light that among women with primary education in the age group 25–29 (of all women, singles included and not just those in a union) a quarter had experienced cohabitation, and that the figure rose with education level to 36 percent for those with completed high school, and still 31 percent for university graduates. As is typical for a universally rising trend, all percentages for older women drop, but among women 40–45 approximately 10 percent had already been cohabiting (Yu et al., 2015). Aparently the education gradient in Japan followed the negative POD-pattern, but in China a clearer SDT positive one was observed. But it is likely that in both countries the practice has spread to other population categories as well, so that these gradients could have been become flatter or reversed. What is still missing in the Asian context is parenthood and further family expansion occuring within the consensual union. In fact, as is well documented for Japan, if not aborted, an accidental pregnancy during courtship immediately results in a shotgun marriage.⁵

Aside from the recently emerging feature of sub-replacement fertility in a number of other Asian countries, such as Buthan (TFR = 1.5), Sri Lanka (1.7), Malaysia (1.7), Thaïland (1.5), Iran (1.7), Brunei (1.8), Nepal (2.0), or Bangla Desh (2.0) (The Population Reference Bureau, 2022), there is only scant news about rising cohabitation. The Catholic population of the Philippines is a clear exception. Their 2010 census estimated that 30 percent of women in a union at ages 20–24 were in consensual unions and that this percentage was 17 percent of partnered women 25–29. The DHS survey of 2013, however yielded considerably higher figures: 52 and 35 percent respectively. The truth may be somewhere in between as persons may be more reluctant to give a true answer when confronted with an official census agent, and when, by contrast, the sampling frame of the DHS surveys may contain an urban bias. Also, the earlier surveys displayed a negative educational gradient, but this was no longer so for the 2013 one (Lesthaeghe, 2020a, 2020b).

Possibly we may glean some expectations about the future of cohabitation in the Asian patriarchal societies by looking at how far they had come with respect to the “ethics revolution”, i.e. their views about individual autonomy of choice re ethics issues such as divorce, abortion, homosexuality, euthanasia and suicide. From the World Values Studies of the period 2005–07, we can obtain an idea of the enduring strength of the stigma against these five practices via the answer category of “Never justified” (as opposed to “depending on circumstances “or “always justified”). In Fig. 1 we have reproduced the cumulated percentages “Never justified” for all 5 topics together, and with a simple scale from 0 to 500. A high score reflects a strong opposition against individual free choice but instead support for a conformist stigmatisation.

Fig. 1

Ethical conservatism. Sum of percentages “Never justified” (NJ): divorce, abortion, homosexuality, euthanasia, suicide. Source: World Values Surveys 2005–07

Obviously Norway and Sweden, the core SDT-countries in Europe, have hardly any stigma left, and customary cohabition fits the very liberal position in the ethics sphere. They are followed by Finland and the Western European countries, but not yet to the same extent as Norway and Sweden. Spain and Italy are reported next. What is striking here is the considerably wider diffusion of cohabitation in Spain than in Italy, which corresponds to a matching distinction with respect to the ethics revolution. In fact, Italy still stands out among the most conservative countries in both respects. The Eastern and Central European populations come next: more conservative ethically and not as far along the cohabitation spectrum. Noteworthy is a Bulgaria versus Romania contrast, which mirrors that of the Spain-Italy pair: considerably more liberal ethics and concomitantly more cohabitation in Bulgaria.

The ethics profiles of the Latin American populations resemble those of the Eastern Europeans, but of course they had the widely pre-existing historical base for cohabitation which the latter did not have. Remarkable is again the position of Uruguay: far ahead on ethical liberalism, and correspondingly by 2010 experiencing the most spectacular cohabitation boom of all.

When moving to Asian countries next, we are on speculative grounds. Figure 1 shows quite substantial differences with Hong Kong, Taiwan, South Korea and also Buddhist Thaïland exhibiting quite liberal profiles,⁶ but with Vietnam and mainland China adhering to traditional conformist ethics. Would this imply that there is a high degree of social acquiesence affecting opinion surveys in mainland China and Vietnam, given their political history of strong state control ? Or has practice, at least in mainland China, evolved faster than public opinion ? Also note that the ethics profile or Japan is the most liberal one in Asia, and at par with the profiles of the Anglo-saxon world.

The conservative responses in India are less of a surprise, and those in the set of Muslim countries even less so. Only Malaysia in the set of Muslim countries listed shows a more liberal stance with respect to these ethics issues. Evidently, the emergence and spread of consensual unions in the wake of individual free partner choice are not yet in the books in the Hindu and Muslim world. And again, sub-replacement may be achieved, as mentioned earlier, long before any feeble rise in cohabitation appears.⁷ In this instance these populations would form the other extreme of the overall Asian pattern with sub-replacement fertility emerging decades before the patriarchal pattern of marriage shows any significant cracks. Evidently, the Latin American pattern of a partial SDT is exactly the opposite, as can be understood from its roots of non-patriarchal kinship organisation.⁸

If the SDT still fully unfolded in the old and new Europes during the previous Century, it is highly likely that those which have already partial patterns could now add missing characteristics during the current one. In parts of Latin America fertility postponement may be imminent in forerunner groups. The industrialised Asian societies are likely to accept a further diffusion of pre-marital cohabitation, but remain reluctant to allow parenthood as a part of it. In the Buddhist world, cohabitation may also emerge in the near future as a result of more free partner choice, but an incipient trend toward cohabitation is not yet likely in the Hindu and Muslim worlds. In these populations sub-replacement fertility can develop at the end phase of the “altruistic” first demographic transition,⁹ but the SDT-type of parenthood postponement, inspired by the actors’ self-interest, may be emerging more slowly, and requires, amongst other factors, more advanced higher female education and concomitant female emancipation in private and public spheres, a firm shift in the direction of more individual autonomy, and hence changes in ethical-religious codes. This is likely to occur first in specific forerunner sub-groups before the pattern spreads to a majority of the population.

Finally, together with low out-of-wedlock fertility, another characteristic of most Asian societies is their low divorce rate. This feature is currently being heralded as proof of Asian resilience against the so called “decadent Western family culture”. But again, this argument has the familiar ring of “not us, we are different”. In patriarchal societies divorce is difficult, and as good as impossible if initiated by women. But, with rising female education, more free partner choice and resistance to arranged marriages, and concern with individual well-being, low divorce rates can quickly become things of the past. Older generations will be alarmed once again, but younger generations are likely to accept new ways as perfectly normal. Again, the model of generational succession is mighty powerful: old ways die off with old generations.

The documentation and analysis of these changes ahead is a challenging task for the current generation of young social scientists, and in particular for demographers and sociologists. And the question is: “how far will the Asian sub-narratives of the SDT deviate from the other ones in the world ?”. It should be stressed that the SDT-related evolutions are at present by no means at an end point, or have barely started (if at all) in most Asian societies. As these trends become more apparent, all present accounts will have to be rewritten. But, as they stand now, the chapters in this issue constitute significant historical landmarks, and they will be lasting reference points for what is there to come.