What economic forces have driven the profound changes in the Western family in the last five decades and what should be the policy responses?
(pages 11-15 of printed journal)
‘The family in the Western world has been radically altered – some claim almost destroyed – by the events of the last three decades.’ Gary Becker, Nobel Laureate, Economics (1981)
For the three decades after Becker’s comment on the transformation of the Western family at the beginning of his revised Treatise on the Family, we have continued to experience profound changes in family organisation and dynamics. These changes – delayed marriage and childbearing, increased cohabitation and nonmarital childbearing, higher rates of divorce and remarriage resulting in complex blended families – have been labelled by demographers the Second Demographic Transition. Despite the benefits of increased personal freedom to choose our preferred family arrangements, these trends have sparked anxiety that we may be witnessing the demise of traditional marriage, and concern for the wellbeing of children in increasingly unstable families. What are the economic forces that have driven these demographic revolutions, and how should we think about policy responses?
The term ‘Demographic Transition’ (or the First Demographic Transition) refers to the dramatic decline in both fertility and mortality rates experienced by industrialising European countries in the nineteenth century. Improvements in sanitation and a more abundant and reliable food supply contributed to falling death rates, especially in the early years of life, though the fundamental causes of the subsequent decrease in fertility are still the subject of controversy.1 Some part of the fall in fertility was doubtless a response to the decline in infant and child mortality but, since the number of surviving children per family also fell dramatically, this cannot explain most of the change. Others have pointed to the decreasing economic value of children as the need for their labour in agriculture decreased and as state social insurance systems began to replace an absolute reliance on children as a source of old age support.
Another explanation for declining fertility can be based on the increased demand for human capital that resulted from new technology introduced in the second stage of the Industrial Revolution. The expansion of railways, electrical power, and the use of the internal combustion engine led to a dramatic increase in investment in plant and machinery. New technology and the accumulation of capital increased the value of labour skills relative to muscle power, and this increase in the demand for labour skills reduced fertility in two ways.2 First, there is a shift in parental incentives towards having fewer children, but investing more in the capabilities of each. Rising levels of education attest to this shift in the quality/quantity trade-off facing parents. Second, the Industrial Revolution increased women’s employment opportunities, and the resulting growth in the value of women’s time increased the cost of raising children. The demographic transition contributed to economic growth by facilitating investment in human capital, and was a key feature of the transition to a modern economy.
The years immediately following World War II were characterised by early marriage and large families in the US and throughout much of the developed world. The first signs of a new regime in patterns of partnership and parenthood began as early as the 1950s and 1960s in Scandinavia and the US – increasing divorce rates followed by falling fertility and an increasing age at first marriage. The concept of the Second Demographic Transition (SDT) was developed by demographers Lesthaege, van de Kaa and collaborators in the mid-1980s to explain cross-national trends towards postponed and less stable marriage and low fertility.3 The SDT is now well-established in North America, Europe and Australia, and is clearly beginning in industrialised Asia.
The common features of the SDT include fertility rates that fall near or below replacement levels, postponed childbearing and increased childlessness, and an increase in extramarital childbearing. Replacement-level fertility is an important threshold, but some countries, including Japan, Korea, Germany, and much of South and East Europe, now have fertility rates below 1.3 children per woman. This level has been labelled lowest-low fertility, and leads, without substantial immigration, to rapid ageing and population decline. Increased cohabitation rather than legal marriage, and a high level of divorce has meant that the partnership context in which childbearing takes place is more heterogeneous and less stable than during the post-war baby boom. Despite widespread trends in these family conditions, there is considerable variation across countries: in Italy and Japan marriage and fertility rates are very low, but there is little cohabitation or childbearing outside legal marriage, while in the Nordic countries cohabitation is common and fertility rates are higher.
The so-called divorce revolution in the US led to a doubling of the divorce rate during the 1970s. Some part of this increase was due to a reduction in the costs of divorce, as many states abandoned fault-based divorce laws to permit unilateral divorce; but divorce rates have increased in recent decades in almost all OECD countries. Separation and re-partnering produces complex families and diverse patterns of parental obligation, and may weaken children’s perceived responsibility for care of their elderly parents in the future. A rise in extramarital births as a proportion of all births has also been a widespread feature of the SDT – about 40 per cent of US births now take place outside legal marriage, and about half of these are to lone parents rather than cohabiting couples.
The changes of the SDT have led, in the affected countries, to both social anxiety about the future of the family and policy debates about appropriate responses. There are two principal sources of concern. First, below-replacement fertility, increased life expectancy, and the resultant ageing population create long-term problems for government financing of pensions and the support of the elderly. Second, high levels of family instability are believed to have adverse impacts on the wellbeing of children and the elderly.
Demographers have emphasised the ideational change that accompanies increased affluence as the principal cause of the retreat from marriage and declining fertility of recent decades. With rising incomes, they argue, comes a focus on ‘higher order needs‘ such as individual autonomy and self-actualisation. A focus on the individual also increases norms of gender equality and decreases the authority of the church and the state over decisions regarding marriage and childbearing. In contrast, an economic approach to the SDT analyses the demographic choices made by individuals who face constraints on their time and material resources. Changes in norms, attitudes and values may reinforce behavioural trends initiated by economic forces, but they are by and large along for the ride. From an economic perspective, the basic causes of the SDT are rooted in technological advances and market developments that have changed the costs of and rewards to childbearing and traditional marriage.
One important driver of the SDT has been the converging economic lives of men and women in most Western countries, which has both reduced the gains from marriage and increased the cost of children. The educational attainment of young women has been rising, and in all OECD countries except Turkey now exceeds that of men. Gender gaps in labour force participation rates, hours worked in the market, and annual earnings have narrowed substantially in most developed countries. Contributing to this convergence have been factors that increase female rates of pay and reduce the barriers to women leaving the home – the development of and increased access to reliable contraception; the falling prices of labour-saving household appliances; the development of markets for home-produced goods and services such as cooking, cleaning and child-care; and the increasing relative wages of women. Changes in production technology and structure of demand have increased the relative returns to cognitive and people skills in the labour market relative to motor skills or physical strength. These alterations in the wage structure have benefited female workers more than male workers, and helped to pull women into the labour force.4
In an economic model of the decision to marry, individuals compare the expected benefits of marriage relative to those of single life, where the advantages of establishing a joint household (or ‘marital surplus’) depend on potential gains in two spheres: production and consumption. The production benefits to marriage come from economies of scale in providing for multi-person households, and from the returns to specialisation and exchange within the household. When one spouse specialises in market work and the other in home work, each becomes more productive in his or her separate sphere, and gains from trade analogous to those from trade between nations ensue. Consumption gains from marriage, on the other hand, come from the joint enjoyment of household public goods such as housing, the value of shared leisure time, and the rewards of raising children together. As dual-career families become the standard, and more household production is outsourced to the market, marriage (and cohabitation) becomes increasingly based on consumption rather than on gender-specialised production, and the overall benefits to marriage are likely to be smaller.5 One of the key implications of the economic approach to the SDT is that the retreat from marriage is unlikely to reverse. Gender equality in the market, which reduces the returns to establishing joint households, is being advanced by fundamental economic forces.
As an ancillary force, the falling costs of exiting marriage seem to have generated contracting problems within the family. It has become increasingly difficult for men and women to forge long-term, enforceable agreements about family life, and this reinforces the secular decline in marital surplus and in marriage. Many factors have reduced the costs of ending a union – the replacement of legal marriage by cohabitation, the liberalisation of divorce laws, and the less-stringent social prohibitions to single parenthood and to divorce. If couples are unable to make binding agreements about their future – basically, I can’t credibly promise not to leave you – then investments in joint ventures, such as children, may be inefficiently low. Investments that are ‘portable’, such as education and career experience, conversely, will be more attractive. If a couple also has a limited ability to make legally-binding agreements about custody or post-separation financial arrangements, the problem is exacerbated.6 Clarity and predictability in post-divorce support and custody arrangements, at the very least, can help to mitigate the consequences of limited commitment.
Persistent social norms can exacerbate these contracting problems by hampering domestic adjustment to a more gender-neutral labour market. As women’s relative education levels have increased, a more egalitarian division of household labour would be efficient, but it can be difficult for young men and women to make a credible contract to share household work in the face of peer pressures that support more traditional gender roles.7 This may explain why fertility has remained higher in countries with more egalitarian norms, such as Norway and Sweden, while marriage and fertility rates have plummeted in more traditional Southern Europe and Japan.
The pronounced changes in partnership and fertility patterns that characterise the SDT have progressed very differently across socio-economic groups: marriage rates have declined more rapidly and extramarital fertility is much higher among women with lower education levels in the US. Nonmarital childbearing (to both lone and cohabiting parents) also has a distinct education gradient across Europe. This implies that children of less-educated mothers are more likely to experience family disruption and are less likely to have a father present than children of more-educated mothers. 8 The diverging family lives of those at the top and the bottom of the income scale have implications for investments in the next generation and for increasing levels of inequality.
The Fragile Families and Child Wellbeing Study, a survey of 5,000 newborns and their parents in US cities in 1999, found that unmarried parents were very likely, at the time of the birth, to be either cohabiting or in a romantic relationship. Five years later, however, only 35 per cent of these parents were still together, and the children had experienced many relationship transitions among their parents, and very unstable living arrangements. Why are marriages so rare and relationships so unstable, particularly among poor parents in the US? Possible explanations include limited gains from legal marriage, which are higher when there is property to protect or when higher incomes increase the returns to joint consumption. An alternative theory emphasises the economic woes of men in low-income communities. Real wages and employment levels have fallen among men with a high school diploma or less in the US since 1980. Considerable empirical evidence supports the dependence of stable relationships on male income contributions, despite the increasing earnings power of women. This implies that men with poor employment prospects and low wages have little to offer a potential marital partner. In low-income communities, particularly those affected by the high incarceration rate of minority men, men with higher earnings and more-stable employment will be sufficiently scarce that they will have little incentive to commit to a long-term relationship.
The SDT has altered the structure and dynamics of families in industrialised countries over the past five decades – making families smaller, less stable, and more heterogeneous. It has been driven by changes in technology and the labour market that have reduced household gender specialisation and the gains from marriage, and increased the costs of children. In some countries, collisions between these economic forces and traditional social norms have driven fertility rates to very low levels and hastened the ageing of the population. Among the poor in America and elsewhere, the economic prospects of men are hindering the formation of stable families and adequate investments in children.
In designing policies to promote higher fertility and greater family stability, it is important to realise that both men and women must expect to reap gains from forming and maintaining joint households and bearing children. When women who are experiencing improved labour market opportunities realistically expect to bear a double burden of work and domestic responsibilities if they marry and have children, then they will have little incentive to follow a traditional family path. Policies that make it easier for women to raise children effectively on their own, though they work against the mutual dependence of the traditional family, do support investment in children, and are probably not measures that we want to roll back in order to promote marriage. Finally, when there are severe barriers to economic contributions to households by men in low-income communities, stable marriages and marital childbearing will be discouraged.
Galor, O and Weil, D, 1996, ‘The gender gap, fertility, and growth,’ American Economic Review, vol. 86, no. 3, pp. 374-387.
Guinnane, T, 2010, ‘The historical fertility transition: a guide for economists,’ forthcoming in Journal of Economic Literature.
Lesthaeghe, R, 2010, ‘The unfolding story of the Second Demographic Transition’, Population and Development Review, vol. 36, no. 2, pp. 211-251.
Lundberg, S, 2010, ‘The changing sexual division of labour’, in Solow, R and Touffut J-P (eds), The Shape of the Division of Labour: Nations, Industries and Households, Edward Elgar, Cheltenham, UK.
Lundberg, S and Pollak, R, 2007, ‘The American family and family economics’, Journal of Economic Perspectives, vol. 21, no. 2, pp. 3-26.
McLanahan, S, 2004, ‘Diverging destinies: how children fare under the Second Demographic Transition’, Demography, vol. 41, no. 4, pp. 607-627.
McLanahan, S and Beck, A, 2010, ‘Parental relationships in fragile families’, The Future of Children, vol. 20, no. 2, pp. 17-37.
Sevilla-Sanz, A, 2005, ‘Social effects, household time allocation, and the decline in union formation’, Congressional Budget Office, Washington, D.C.
Stevenson, B and Wolfers, J, 2007, ‘Marriage and divorce: changes and their driving forces’, Journal of Economic Perspectives, vol. 21, no. 2, pp. 27-52.
1 An excellent recent review is provided by Guinnane (forthcoming).
2 These arguments are developed formally in Galor and Weil (1999) and Galor and Moav (2002).
3 For a review, see Lesthaege (2010).
4 Lundberg (2010).
5Stevenson and Wolfers (2007).
6The problem is more acute in specialised marriages. Since specialists in home production invest little in market skills, these arrangements rely on a permanent, or at least long-term, commitment between husbands and wives to protect the home-based partner from future poverty.
7 Sevilla-Sanz (2005).
8 McLanahan (2004).