Final Report Summary - CODEC (Consequences of Demographic Change)
What are the consequences of demographic change? The immediate answer is population ageing, which comes about as a direct effect of low fertility and an increase in life expectancy. It is a development we are observing across all OECD countries - but the degree to which it is happening differs - largely because of differences in fertility levels. Fertility levels in the Nordic and Anglo-Saxon countries have remained rather high in recent decades, whereas it has been worryingly low in the Mediterranean and East European countries. Yet, Western countries are experiencing many other changes in the demographic realm: increasing childbearing outside marriage, increase in divorce and an increase in remarriage, followed with an increase in step-parenthood. These changes have raised serious concern in the more conservative corners of our societies, and often these changes are given negative connotations. Cohabitation and out-of-wedlock childbearing are for instance associated with bad child outcomes and economic disadvantage, especially for women. One simple way to frame the research questions of this project is to divide demographic change in terms micros and macro perspectives.
This project has tackled both aspects, and given recent demographic change, there are good reasons to question many of the conventional wisdoms. Despite the worries about new demographic behavior, and its potential detrimental impact on our societies, it turns out that the highest performing societies, are also the ones that have advanced farthest in adopting new demographic behavior. The Nordic countries, for instance, are the ones where divorce is the highest and out-of-wedlock childbearing is now more common than having children within marriage, yet they are the countries which performs highest on a multitude of indicators.
Lets start at the micro level. Is divorce bad for children? Despite the long list of studies claiming that this is so - by taking data from the Millennium Cohort Survey, which follow children over a long time period, and using fixed effect estimation, it is clear that divorce has no effect on children's cognitive outcomes and only modest effect on their behavior. That is not to say that bad family background does not affect the wellbeing of children, but the marital split by itself is not necessarily to blame. Quite on the contrary, a divorce might enhance the wellbeing of the children if there was a very high level of conflict. CODEC follows up on several other applications of similar nature. For instance, the project consider to what extent a partnership breakup affects psychological wellbeing - putting special focus on married as opposed to cohabiting couples, with the argument that a marital split might be worse - as seems to be suggested in the literature. Again using fixed effect estimation techniques, we find there is no impact. What does matter, however, is whether children are present. Having children make a split more painful independent of whether it is from a marriage or a cohabiting union.
Another relevant question asked in this project is - given that fertility is decreasing in so many countries, do children make individuals less satisfied than before? Can low fertility be explained by the fact that raising children is becoming more demanding and stressful in a world where there is ever increasing pressure in terms of work performance and economic success? At the individual level the evidence is mixed, and it clearly depends on whether one considers the first child as opposed to the second or the third. From longitudinal studies we know that individuals' subjective wellbeing increases during the period before the child arrives, which might in turn explain much of the mixed results in the literature (it declines in the period after childbirth). What is clear however, is that subjective wellbeing associated with childbearing differs substantially across countries. Using information from the European Social Survey (ESS) we find that happiness associated with childbearing is clearly higher in high fertility countries. Moreover, there is a strong gender difference. In countries where fertility is low, mothers are always unhappier than non-mothers, whereas fathers are always happier than non-father - no matter the country. These discrepancies lend support to recent theoretical perspectives developed by MacDonald and Gosta Esping-Andersen, and in many respects crystallize the shortcomings of the more established theories (such as Becker's new home economics or the Second Demographic Transition). In the latter part of the project, the team have put a strong emphasis on the importance of institutions. The "usual suspect" in the demography literature is that fertility might become low if there is insufficient provision of childcare, especially because it makes it difficult for women to combine a working career and having children. There is no coincidence that in those countries where childcare is poorly developed, we also find low fertility and low labor force participation among women.
When childcare is scarce, the most common resource for parents with young children is that of grandparents. However, as we show in this project, the mechanism in which grandparents play a role for childbearing is complex. Importantly, the extent to which grandparent may help out in childcare, and therefore boost fertility, depends on their own fertility history. For instance, if they themselves have many children, they will not necessarily have a capacity to help out all grandchildren even if their children have few children. More importantly, the postponement of childbearing will bring about a larger age gap between the generations, which in turn may hamper their ability to effectively care for children. Finally, the role of grandparents interact closely with the provision of childcare by the state. Again, we see a clear North-South divide: grandparents play a critical role in the South, and much less so in the North.
The project has highlighted new issues concerning the interaction between institutions and demographic change, and it has also asked the more ambitious question - starting from the male breadwinner model of the sixties - why did institutions change so differently across Western countries (or similarly - why did some change and others not) - and which role did it have for demographic change? This question is important, because it necessarily affects the consequences for demographic change. We have argued in this project that one of the most important societal shocks to Western societies is one of mass expansion in education. This has lead to much higher educational levels among women, and with it, a massive change in attitudes and aspirations. It is clear that the way countries have adapted and changed their institutions in response to these developments - vary dramatically. In Nordic countries we have seen massive outsourcing of traditional family activities, such as care for children and the elderly, which in turn has had the effect of keeping both fertility and female labor force participation rates high. These countries are also the ones where we see new demographic behavior emerging, manifested through out-of-wedlock childbearing and divorce, yet they are the most successful societies no matter which indicators considered. The project has made suggestions for why institutions has responded differently to these changes. The Nordic countries, typically characterized by a flatter organizational structure, and where trust is higher, might have been better able to adjust to the new aspirations of women. Instead, countries where institutions are more rigid and hierarchical, adaption to new preferences seems much harder. This has important implications. Whereas out-of-wedlock childbearing and divorce may not have much detrimental effect in a society well prepared for these behaviors, they may have more serious effects in those countries where institutions are not able to adapt. This raises certain worries since very recently, we have seen very sharp increases in divorce and out-of-wedlock childbearing also in those countries, which arguably do not have the same institutional protection and preparation. It seems that demographic change is moving faster than institutions, and in this sense the project is opening up a new line of research.
This project has tackled both aspects, and given recent demographic change, there are good reasons to question many of the conventional wisdoms. Despite the worries about new demographic behavior, and its potential detrimental impact on our societies, it turns out that the highest performing societies, are also the ones that have advanced farthest in adopting new demographic behavior. The Nordic countries, for instance, are the ones where divorce is the highest and out-of-wedlock childbearing is now more common than having children within marriage, yet they are the countries which performs highest on a multitude of indicators.
Lets start at the micro level. Is divorce bad for children? Despite the long list of studies claiming that this is so - by taking data from the Millennium Cohort Survey, which follow children over a long time period, and using fixed effect estimation, it is clear that divorce has no effect on children's cognitive outcomes and only modest effect on their behavior. That is not to say that bad family background does not affect the wellbeing of children, but the marital split by itself is not necessarily to blame. Quite on the contrary, a divorce might enhance the wellbeing of the children if there was a very high level of conflict. CODEC follows up on several other applications of similar nature. For instance, the project consider to what extent a partnership breakup affects psychological wellbeing - putting special focus on married as opposed to cohabiting couples, with the argument that a marital split might be worse - as seems to be suggested in the literature. Again using fixed effect estimation techniques, we find there is no impact. What does matter, however, is whether children are present. Having children make a split more painful independent of whether it is from a marriage or a cohabiting union.
Another relevant question asked in this project is - given that fertility is decreasing in so many countries, do children make individuals less satisfied than before? Can low fertility be explained by the fact that raising children is becoming more demanding and stressful in a world where there is ever increasing pressure in terms of work performance and economic success? At the individual level the evidence is mixed, and it clearly depends on whether one considers the first child as opposed to the second or the third. From longitudinal studies we know that individuals' subjective wellbeing increases during the period before the child arrives, which might in turn explain much of the mixed results in the literature (it declines in the period after childbirth). What is clear however, is that subjective wellbeing associated with childbearing differs substantially across countries. Using information from the European Social Survey (ESS) we find that happiness associated with childbearing is clearly higher in high fertility countries. Moreover, there is a strong gender difference. In countries where fertility is low, mothers are always unhappier than non-mothers, whereas fathers are always happier than non-father - no matter the country. These discrepancies lend support to recent theoretical perspectives developed by MacDonald and Gosta Esping-Andersen, and in many respects crystallize the shortcomings of the more established theories (such as Becker's new home economics or the Second Demographic Transition). In the latter part of the project, the team have put a strong emphasis on the importance of institutions. The "usual suspect" in the demography literature is that fertility might become low if there is insufficient provision of childcare, especially because it makes it difficult for women to combine a working career and having children. There is no coincidence that in those countries where childcare is poorly developed, we also find low fertility and low labor force participation among women.
When childcare is scarce, the most common resource for parents with young children is that of grandparents. However, as we show in this project, the mechanism in which grandparents play a role for childbearing is complex. Importantly, the extent to which grandparent may help out in childcare, and therefore boost fertility, depends on their own fertility history. For instance, if they themselves have many children, they will not necessarily have a capacity to help out all grandchildren even if their children have few children. More importantly, the postponement of childbearing will bring about a larger age gap between the generations, which in turn may hamper their ability to effectively care for children. Finally, the role of grandparents interact closely with the provision of childcare by the state. Again, we see a clear North-South divide: grandparents play a critical role in the South, and much less so in the North.
The project has highlighted new issues concerning the interaction between institutions and demographic change, and it has also asked the more ambitious question - starting from the male breadwinner model of the sixties - why did institutions change so differently across Western countries (or similarly - why did some change and others not) - and which role did it have for demographic change? This question is important, because it necessarily affects the consequences for demographic change. We have argued in this project that one of the most important societal shocks to Western societies is one of mass expansion in education. This has lead to much higher educational levels among women, and with it, a massive change in attitudes and aspirations. It is clear that the way countries have adapted and changed their institutions in response to these developments - vary dramatically. In Nordic countries we have seen massive outsourcing of traditional family activities, such as care for children and the elderly, which in turn has had the effect of keeping both fertility and female labor force participation rates high. These countries are also the ones where we see new demographic behavior emerging, manifested through out-of-wedlock childbearing and divorce, yet they are the most successful societies no matter which indicators considered. The project has made suggestions for why institutions has responded differently to these changes. The Nordic countries, typically characterized by a flatter organizational structure, and where trust is higher, might have been better able to adjust to the new aspirations of women. Instead, countries where institutions are more rigid and hierarchical, adaption to new preferences seems much harder. This has important implications. Whereas out-of-wedlock childbearing and divorce may not have much detrimental effect in a society well prepared for these behaviors, they may have more serious effects in those countries where institutions are not able to adapt. This raises certain worries since very recently, we have seen very sharp increases in divorce and out-of-wedlock childbearing also in those countries, which arguably do not have the same institutional protection and preparation. It seems that demographic change is moving faster than institutions, and in this sense the project is opening up a new line of research.