Research report no.11 2004

'It's not for lack of wanting kids...'
A report on the Fertility Decision Making Project

by Ruth Weston, Lixia Qu, Robyn Parker and Michael Alexander

1. Introduction

According to the Australian Bureau of Statistics (ABS), the total fertility rate in Australia has plummeted to an all-time low of 1.75 babies per woman (ABS, 2003). This rate is well below that required for replacement (2.1).1 Alarm bells are thus ringing and there is now much public debate about the social and economic consequences of this trend.

This chapter first outlines trends in the total fertility rate that have occurred since the late 19th Century and consider some of the events that took place that appear to explain shorter-term aberrations. The chapter then focuses on some of the challenges current fertility trends present for Australia. Finally, the literature concerning apparent reasons for these trends is reviewed, for some of the explanations that have been proposed have played a major role in shaping the Fertility Decision Making Project.

1.1 Trends in the total fertility rate

While much of the attention in Australia has focused on patterns of fertility since the post-war baby boom, the fertility decline well predates the boom. Indeed, at the start of the 20th Century, the total fertility rate was nearly 4.0, having fallen by about one third over the previous 30 years (Hugo 2001).

By 1934, during the Great Depression, the fertility rate had fallen to replacement level for the first time (2.1) - a trend that apparently resulted more from deliberate birth control within marriage than from the postponement of marriage that occurred at the time (Ruzicka and Caldwell 1982). In the late 1940s and early 1950s, the economic outlook improved and fertility increased, as births postponed during World War II took place. This rate was sustained by the trend towards earlier and near universal marriages at that time, as well as the influx of immigrants of child bearing age (Hugo 2001; Ruzicka and Caldwell 1982). In fact, the "baby boom" which began after World War II was really a "marriage boom" as fertility rates within marriage changed little (Ruzicka and Caldwell 1982).

By 1961, the total fertility rate was the highest for the century (3.5) - but 15 years later, it had for the first time fallen below replacement level. The rate levelled out in the late 1970s and 1980s as births that were postponed took place (Carmichael and McDonald 1999), but then fell in small progressive steps in the 1990s, as increasing proportions of couples restricted their family size to one or two children, or remained permanently childless.2

While the fall in fertility in the early years of the 20th century applied to virtually all age groups of women, the fall in recent decades has been restricted to those under the age of 30 years. In fact, the proportion of women in their thirties giving birth has increased in recent decades, although women of this age today are less likely to have a child compared with women of this age during the baby boom period and before the Great Depression of the 1930s. Nowadays, women giving birth when at least 30 years old are increasingly likely to be first-time mothers (38% of all first births in 2001 were to women of this age, compared with 28% in 1993) (Australian Institute of Health and Welfare 2004; ABS 2001). In other words, women are delaying having their first child. Nevertheless, the rise in the fertility rate of women aged 30 or more years has not made up for the fall in the rate for younger women.

These trends reflect dramatic changes in the number of children women have had by the time they have reached their early forties. By this age, women born between 1937 and 1941 were more likely than women born twenty years later (1957-1961) to have four or more children (28% compared with 15%) and less likely to be childless (9% compared with 13%) (ABS 2002a). Of women who are currently in their early childbearing years, around one quarter are projected to remain childless (ABS 2002b). But according to McDonald (1998), future trends in Australia's total fertility rate will be strongly affected by the proportions of women who have more than two children. He estimated that the total fertility rate would fall to 1.4 if those who currently have more than two children stopped at two.

The decline in fertility is not unique to Australia. Beginning in France from the 19th century, the decline in fertility, resulting not from catastrophe as in early times, but from deliberate birth control, spread globally but unevenly. Total fertility rates now fall below replacement level in countries that account for almost half the world's population, including virtually all populations of European origin, as well as Japan and China (Eberstat 2000). In fact, as de Vaus (2004) notes, Australia's total fertility rate is higher than the rate in most OECD countries. Public concern about flagging fertility rates is by no means new. In Australia, a Royal Commission was established in New South Wales soon after Federation to seek reasons for the dramatic fall in recent decades (the Royal Commission on the Decline of the Birth-Rate and on the Mortality of Infants in New South Wales [the "Mackellar Commission"]). And in 1945, British Prime Minister Winston Churchill encouraged women to have "One [child] for the mother, one for the father, one for accidents and one for increase" (Riley 2004). However, some countries have been quicker than others to develop policies to address this issue (see McDonald 2000a; Demeny 2004). But is there really any need for concern?

1.2 Implications of fertility trends

In earlier times, concerns about the declining fertility rate tended to focus on the need to increase the population - needs that could also be addressed through immigration. After World War II, the message, "populate or perish" was widely promoted. In recent times, however, attention has turned to the fact that Australia's population is ageing as a result of the combined effects of a falling fertility rate and increasing life expectancy. Empirical analysis by McDonald and Kippen (1999) suggests that the capacity for immigration to affect the age structure is quite limited. While the first 80,000 migrants affect age structure, these authors show that as levels of net immigration continue to increase above 80,000, the magnitude of their effect on the age structure of the population diminishes. But what are the consequences of an ageing population?

An important question relating to population ageing is whether or not there will remain a sufficient labour supply to support the elderly, taking into account the fact that the proportional representation of the other main dependent group (those too young to work) is shrinking. The issue is becoming increasingly critical, given that that the first of the baby boomers (defined by the ABS 2001 as those born between 1946 and1966) will turn 60 years old in 2006.

Regarding this issue, the Australian Government's Intergenerational Report (2002) estimates that the net costs to the Government of an ageing population will increase, with today's taxpayers probably subjecting tomorrow's taxpayers to a heavier tax load. Some social commentators have argued that such circumstances will create growing resentment between the generations (see Encel 2002).

Demand for services will also change, given the different needs of the burgeoning elderly population and younger generations in such areas as housing, health care, leisure, and education. And while there is considerable evidence that families are the most significant sources of support for the elderly, increasing rates of childlessness coupled with family breakdown and children pursuing jobs overseas, will mean that many elderly parents will be either "functionally" or "actually" childless (see Rowland 2003; Weston, Qu and Soriano 2003).

In relation to young children, Mackay (2001) has noted that little attention has been given to the difficulties that children may face in the middle of this century, when one quarter of the population will be over the age of 65.

The workforce itself will age, if policies directed towards encouraging later retirement are effective. Thus any wisdom shaped by an accumulation of knowledge will increase, while there will be a relative loss of young adults whose age-related talents often produce important technological advancements (McDonald 2002).3

But how will Australia fare if its labour force stagnates or falls in absolute size? According to McDonald and Kippen (2000), this issue has received insufficient attention. Allied to this problem, there is the concern that, if the fertility rate falls below 1.6 and Australia's current net immigration levels are sustained, then the absolute size of Australia's population will shrink - a trend that will spiral over time (ABS 2002b; McDonald 2000b). Of course, for those concerned with the impact of Australia's current population on environmental sustainability, this might seem a welcome prospect.

Another issue, often overlooked in discussions of the implications of the fall in the fertility rate, is that women appear to be experiencing dashed hopes and unfulfilled expectations and intentions: they initially want to have, and expect and intend having, more children than they achieve (Fisher 2002; McDonald 1998; Qu and Soriano 2004; Quesnel-Valee and Morgan 2002).4 However, there is some evidence from overseas countries that the fertility aspirations of future generations of young adults will be lower than those of young adults today, for it appears that aspirations are eventually affected by fertility trends. For instance, Goldstein, Lutz and Testa (2003) found that the average family sizes considered to be personally ideal by young men and women in German-speaking parts of Europe has fallen to as low as 1.7 children. Nevertheless, the evidence available so far suggests that fertility aspirations are not being met by many people in many developed countries. Why is this so?

The following section examines reasons for the declining fertility rate. Some of these reasons are clearly evident, and others are more speculative. Furthermore, some of the clearly evident ones, such as postponement of births, require their own explanations - explanations that in turn have causes that need deciphering. At the base of these layers of explanations is interesting speculation.

1.3 Reasons for the decline in fertility

The search for reasons for the low fertility rate is by no means a recent phenomenon - a case in point being the above-noted "Mackellar Commission" that was established around the turn of the 20th century. One very early theory, which came to be known as the "demographic transition theory" in the 1940s, explained the decline in fertility in terms of the move from one level of equilibrium in which fertility levels were high to compensate for high mortality, to another in which fertility levels fall as mortality levels falls so that population replacement is achieved (see McDonald 2001a). However, fertility is now well below replacement level in many countries. As McDonald notes, homeostatic explanations involve such a flexible time frame that they verge on tautology: they suggest that corrections to imbalances via fertility change will eventually occur, but "who knows when?" McDonald thus argues that generalised frameworks are needed to explain current fertility changes - changes that will nonetheless be influenced by the specific social, economic and institutional contexts in which they occur.

Many interacting factors underpin the fall in the fertility rate - a complexity that makes the choice of where to begin this review somewhat arbitrary. As noted above, some of these factors are transparent and certain and relate specifically to contextual factors; others are not. Perhaps the most obvious are the development of modern fertility control measures and the related changing patterns of family formation, including couple formation. The review thus begins with these issues.

1.3.1 Advances in reproductive technology

A key explanation for the fall in fertility is that first births have increasingly been postponed thereby shortening women's total childbearing years and increasing their chances of not having any children (Jain and McDonald 1997). While some women attempt to extend their reproductive life through assisted reproductive technology, such procedures account for only 2 per cent of all births (McDonald 2001a).

As noted above, it appears that women today end up having fewer children than they want, expect and intend to have. Such trends are in marked contrast to those of some 40 years ago when couples often had more children than they expected because of relatively inefficient means of contraception (Petersen 1961). Clearly, the effects of the development of the contraceptive pill and its widespread acceptance have been revolutionary.

In Australia, the contraceptive pill became available in 1961 for distribution via medical prescriptions from those practitioners who approved of its use. In 1972, sales tax on all contraceptives was removed and the pill was placed on the Pharmaceutical Benefits List, thereby lowering the price substantially, increasing its use, and leading to widespread acceptance. Other advances in contraception that contributed to the modern fall in fertility include new versions of the intra-uterine device from the mid-1960s, developments in sterilisation procedures in the 1970s and 1980s, and liberalisation of abortion laws (Cica 1998; Carmichael 1998).

However, the average age of women giving birth to their first child continued to increase well after the pill became readily available. The median age of married women giving birth to their first child of the union of the time was 25.3 in 1981, 29.3 in 1991, and 29.8 in 2001, although some of these women would have had children from previous relationships. As outlined below, advances in reproductive technology facilitated a number of other life course changes that combined to accentuate the delay in first births and fall in fertility (ABS 2001, 2002a).

1.3.2 Life course trends

Life course changes affecting fertility include delays in those transitions that typically precede the family formation process, most particularly leaving home and forming partnerships, an overall fall in the formation of partnerships and an increase in their rate of breakdown, consequent diverging reproductive histories of potential partners, and women's increased financial independence.

Delayed transitions preceding having children

Although leaving the parental home remains an important transition in life, it is no longer quite the symbol of adulthood it once was. Young adults are taking this step at an older age than in the past, and the first move is likely to be temporary rather than permanent. Between 1979 and 2000, the proportion of those in their early twenties living with their parents increased from 46 per cent to 52 per cent for men and from 25 per cent to 39 per cent for women. In two surveys conducted by the Australian Institute of Family Studies in 1981 and 1998, the proportion of those in their early twenties who had returned after leaving home increased from 52 per cent to 67 per cent for men and from 44 per cent to 57 per cent for women (McDonald 1995; Weston, Stanton, Qu and Soriano 2001).

While marriage used to be the main reason for leaving home, these days young adults leave for a variety of reasons, and form their first (living-together) union later in life than previously (Qu and Weston 2001). Furthermore, their first union is unlikely to start with marriage - in 2001, for example, 69 per cent of marriages that were first marriages for both the bride and groom were preceded by cohabitation (or de facto relationships). Yet couples still typically wait until they are married before having children, although this trend is weakening. While the proportion of babies born outside marriage has been increasing progressively, a trend that probably largely reflects the increase in cohabitation, more than two-thirds of babies are born within marriage (69 per cent in 2001, compared with 90 per cent in 1976) (ABS 1995, 2002a).

Research in the United States reported by Barber and Axinn (1998) suggests that, while some individuals opt for cohabitation rather than marriage because they do not want to have children, the experience of cohabitation itself may lower intentions of having children. Although such research is by no means definitive, it is possible that one partner may opt for cohabitation because he or she does not want to have children and, over time, changes the other partner's views about having children.

More broadly, individuals' life decisions are usually influenced by general societal trends, and these decisions can then perpetuate these trends. Delayed milestones that precede having children represent an example of this process (ABS 2001). In turn, as marriage becomes increasingly delayed, people who eventually wish to marry can increasingly afford to wait because the pool of eligible partners remains large at a later point in the life span (Kohler, Billari and Ortega 2001).

Nevertheless, the rate of overall partnering has fallen. In essence, while cohabitation rates have increased, this increase is not large enough to offset the fall in marriage rates (Birrell and Rapson 1998; Birrell, Rapson and Hourigan 2004) - a trend that would contribute to the increased rate of childlessness.

While partnering rates have fallen, the majority who do enter unions are faced with the prospect that their relationship may break down. The fragility of relationships, coupled with postponement of childbearing, increases the chances of not having children, or of smaller than intended families. These issues are considered below.

Those with partners

Despite delays in the timing of partnering and falls in the overall rate of partnering for different age groups, the majority of men and women in their early thirties were living with a partner in 2001 (59% and 55% respectively). Of those in their late twenties, just over half the women (53%) but only 41 per cent of the men had partners (Birrell et al. 2004). Given the decline in fertility overall and various apparent constraints in having children (outlined below), it is increasingly likely that at least one partner will either want to remain childless or will want to defer having children. What happens when couples disagree?

As Greene and Biddlecom (1997) note, most discussion of fertility trends ignore men and the couple dynamics. There is thus scant information about the reasons for, or consequences of, divergent preferences of partners regarding total family size, whether or not to have a first or additional child, or the timing of the next child. Little is also known about the ways in which couples resolve their differences.

The few studies that have taken into account the views of each partner suggest that men's preferences or expectations do influence childbearing outcomes. For instance, Thomson (1997b) found that when couples disagree about having children, intentions tend to shift towards not having a child, while Greene and Biddlecom (1997) cite evidence suggesting that a difference in opinion over the timing of the (next) birth are particularly influential in determining whether or not they have a child.

Fragility of relationships

As noted above, cohabitation is now the dominant pathway to marriage. To outsiders, the period of cohabitation might thus be seen as an "engagement" or at least a symbol that the partners are committed to each other. But for some couples, the meaning of cohabitation may be uncertain, may differ for each partner, or change with time. In the meantime, with the delay in the establishment of this union, the woman's "biological clock" is ticking. There is some evidence that first unions that begin with cohabitation are increasingly likely to end in separation and that relationship breakdown is an important reason for women losing opportunities of having the children they once intended to have (Qu and Weston 2001; Qu, Weston and Kilmartin 2000).

Like cohabitation, the fragility of marriage can also limit opportunities for having children. The divorce rate rose dramatically following the Family Law Act 1975, as the backlog of long-term separations were formalised and some divorces were brought forward. The divorce rate then subsided but has remained at a much higher level than prior to the Act. In the mid to late 1980s, the rate varied between 10.6 and 10.9 divorces per 1000 married women, and fluctuated between 12.0 and 12.9 divorces per 1000 married women during the period of 1995 to 2000.

While relationship separation may disrupt opportunities for childbearing, there is also evidence to suggest that unhappily married couples are less likely than other couples to have a (further) child (Lillard and Waite 1993). This may be more likely in contemporary marriages given that their survival is often under threat when emotional, sexual and companionship needs are not being met (Wolcott 1999). That is, currently intact but unhappy marriages may also result in lowering fertility.

Parents without partners

In a context of high relationship breakdown, there are also many parents without partners, some of who may have hoped for more children. But their status as parents appears to lower their likelihood of forming new partnerships and may also lower their likelihood of having more children should they partner (Greene and Biddlecom 1997; Stewart 2002; Thomson 1997a, 1997b).

However, Stewart (2002) points out that few fertility studies have taken into account the impact of stepchildren, and most of those that do, have not only ignored their "step" status (by combining biological children with stepchildren), but have also limited their attention to the children of the mother who are living with her and her partner.

Little is known about the extent to which parenting responsibilities of fathers without partners (including the cost of supporting non-resident children) affect their views about, and their likelihood of, partnering and having further children. In the United States, where child support regimes vary across states, research by Bloom, Conrad and Miller (1998) suggests that child support enforcement reduces the likelihood of marriage for low-income men. On the other hand, if marriage does occur, fathers (low income or otherwise) are just as likely as other husbands to have a child in this relationship.

Women's growing financial independence

Women's increased workforce participation and financial independence appear to be centrally linked with fertility trends. Women now have access to a wider range of job types than in the past, have more opportunities to earn high incomes, and are no longer as reliant on finding a partner for financial support as in the past. Furthermore, women's opportunities to remain in paid work after they marry have increased over the past 35 years. It was not until 1966 that women were permitted to be appointed or to remain as permanent officers of the Commonwealth Public Service upon their marriage and to return to their jobs after the birth of their children. Since this period, participation in the labour force has increased dramatically for married women in all age groups (Weston et al. 2001).

Such a major social change inevitably requires other adjustments within the home, workplace and community to facilitate the new way of life. But as Moen and Yu (2000) point out, some of these adjustments take time to occur. A number of explanations for low fertility relate to such time lags. These include inflexible workplace practices; women's "double burden" and men's reluctance to share this load; limited access to affordable, high quality child care; financial, career and other work-related costs linked with giving up work partially or fully to care for children; and potential clashes in values, attitudes or beliefs about family and paid work responsibilities.

Summary

In short, child bearing depends largely on the formation and continuation of "live-in" partnerships, most particularly marriage. But partnerships are occurring later in life, their rate has fallen overall, and the risk of their breaking down within the first few years of their formation is considerable.

Furthermore, the presence of children from previous relationships may lower chances of partnering and of having additional children. Finally, women now have opportunities for financial independence and career success - opportunities that may interfere with partnership formation or their plans to have children.

As explained above, life course changes affecting fertility involve their own set of causes. What is behind them?

1.3.3 Labour market, economy and other broad structural forces

Several authors maintain that trends in the labour market and the economy in general influence fertility rates, in part through their impact upon life course patterns. Not surprisingly, there is strong historical evidence that Australia's economic circumstances have played a major role in influencing fertility rates since well before modern methods of birth control were introduced. Indeed, the sharp fall in fertility in the 1890s has been explained in terms of the sudden economic collapse that occurred in 1893 (Caldwell 1982; Ruzicka and Caldwell 1982), while the Great Depression of the 1930s saw the total fertility rate falling to 2.1 babies per woman in 1934, then rising as the economic outlook improved. Smaller observable oscillations have also been linked with Australia's economic circumstances (Martin 2003).

But if the economic outlook is so important to the fertility rate, why has the rate continued to fall in recent times, when the outlook has been relatively positive?

Several authors have implicated the globalisation of the economy and associated labour market changes of the last two decades. Low-skilled yet relatively highly paid and secure jobs available to early school leavers have virtually disappeared, having been replaced by jobs entailing fixedterm contracts and part-time or casual hours, thereby providing limited economic security (Kohler et al. 2001; McDonald 2000c, 2001c; Saunders 2001).

McDonald (2001a) also argues that this era of job insecurity has been accompanied by a strong economic cycle of "booms and busts" and rising or fluctuating house prices which combine to encourage continuing dual employment amongst couples as insurance against dual joblessness, and to lead young people to invest in their own "human capital" (their education and career development) before considering having children. As Birrell et al. (2004) and McDonald (2001a) point out, the economy has generated increasing inequity between "winners" and "losers". Furthermore, Birrell and colleagues conclude that men with limited employment prospects are reluctant to take on the responsibilities of marriage and parenting, while women may also be reluctant to accept them as partners.

The pressure to invest in education is strengthened by the fact that the stakes are very high. Investment in education, in turn, increases debts, which may lead to further postponement of childbearing (Jackson 2002). According to McDonald (2000b, 2001d), childbearing is also discouraged by the limited financial, child care and other supports available to couples when mothers wish to participate in the labour force.

Inextricably linked with these broad structural forces are the considerable costs of having and raising children. These costs and their interplay are likely to cast a shadow on the benefits of having children and reinforce the need for educational and career investment.

1.3.4 Are the benefits of having children decreasing?

Schoen, Kim, Nathanson and colleagues (1997) point out that explanations for modern fertility trends can often be reduced to the argument that the fall in fertility has resulted not only from increasing costs of having children but also from decreasing benefits. Is there any evidence that this might be the case? The following discussion focuses on arguments in the literature about links between fertility trends and the nature of the benefits and costs of having children.

While much of the literature has focused on costs of having children, relatively few studies have attempted to assess perceived benefits. Perhaps the most well known is the large-scale study on the Value of Children, conducted in the 1970s in nine countries that varied in terms of development (for example, the Philippines and the United States). Benefits included economic, practical and psychosocial gains such as receiving help with household chores or economic support in old age, experiencing pride, love and companionship, deriving pleasure from watching the children grow up, making life more purposeful, and carrying on the family name (Hoffman and Manis 1979). More recently, Schoen and colleagues (1997) emphasise the importance of children in expanding and strengthening social ties, which in turn provide avenues to many other resources.

In developing countries young children are often relied upon to contribute financially to the family. However, in countries (like Australia) where child labour laws virtually eliminate children's direct contribution to the household economy, child-related benefits are mainly psychosocial. Some authors argue that these benefits do not accumulate with increasing family size. Kagitcibasi (1997) concludes that parents can derive all the love they need from one or two children. Similarly, Schoen and colleagues (1997) argue that few children are required to strengthen family and friendship networks, and Kohlmann (2002) maintains that restricting family size may even increase the chances of receiving some benefits. According to this argument, in countries where economic benefits of children only apply in old age, it is more efficient for parents to have few children thereby allowing strong investment in their education. Indeed, for some parents, such investment may be seen as a ticket to "winning" in a free market economy in which the losses of the "losers" are severe.

In short, while having children can be very rewarding, several researchers maintain that such rewards do not accumulate as family size increases. In fact, the chances of receiving some benefits may decrease as family size increases. The financial costs, on the other hand, can be a force by themselves - as can the many non-financial costs of children.

1.3.5 Are the costs of having children increasing?

The fall in fertility rate has been attributed to a number of rising costs of having children, most particularly direct and indirect financial costs, but also emotional costs relating to giving up paid work and costs that might never eventuate but that nonetheless encourage couples to "think twice" about having children.

Financial and work-related costs

Over time, material living standards have increased in modern societies, with many of yesterday's perceived luxuries becoming today's perceived necessities - thus highlighting the importance of values in shaping considerations about whether or not a couple can afford to have children. Indeed, divergent views about the material necessities in life and associated expenditure patterns may partly explain the fact that the people who can least afford it are having the largest families (see de Vaus 2004).

In addition to the direct financial costs of having and raising children (for example, food, clothing, housing, education and possibly child care) are the indirect costs such as reduced earnings and potentially curtailed careers when caring for children takes precedence over paid work. While it appears that the foregone earnings of women who have children have diminished somewhat since the 1980s, when cohorts with the same level of education are compared (Gray and Chapman 2001), women's improved career opportunities mean that increasing numbers have a great deal to lose should they decide to relinquish full-time work to raise a family. Such losses are emphasised by McDonald (2000b, 2001a) as a central reason for the fall in fertility.

The deprivations are not only monetary, but can be felt in relinquishment of work-related prestige and diminished opportunities for social relationships, job skill development, and mental stimulation and challenge (see Barnes 2000; Perry-Jenkins, Repeti and Crouter 2000). Thus, while the labour market and economic context may encourage the accumulation of educational and career-based accomplishments, the benefits of this pathway may compete with having children as alternative avenues for self-fulfilment (Quesnel-Valee and Morgan 2002).

Parents may, of course, use non-parental child care to enable each partner to remain in full-time work, but this introduces other monetary and time costs, with women typically carrying most of the child care and other domestic responsibilities (Bittman and Pixley 1997; Bittman and Matheson 1996; Moen and Yu 2000). These costs, along with potential difficulties in accessing high quality child care and worries about the suitability of child care for very young children, may contribute to couples' decisions to have few if any children, or to continue to defer decision-making.

The above issues relate to the "gender equity" explanation for low fertility (which will be discussed further below). Basically, McDonald (2000c, 2001b) argues that gender equity has progressed well in those institutions that are oriented to the individual, such as education and employment, but family-oriented institutions have lagged behind by continuing to assume a male breadwinner model of family life. Under these circumstances, women's opportunities are seriously diminished if they have children, thereby discouraging them from having the number of children many would like to have. The potential for indirect discrimination around pregnancy and family commitments are also work-related costs that may influence fertility decisions. (For incidence of complaints, see Human Rights and Equal Opportunity Commission Annual Report, 2002-2003).

While several policies have been introduced to facilitate workers' achievement of their caring responsibilities, access to such work benefits appears to vary both between and within organisations (Gray and Tudball 2001). Furthermore, workers with family responsibilities may be reluctant to take advantage of existing benefits if they believe that doing so is likely to cause major disruptions at work, or jeopardise either their career advancement or their relationships with others in the workplace (Judiesch and Lyness 1999; Galinsky 1999; Marshall and Barnett 1993). The prevalence of such hidden barriers in Australia and their relevance to decisions about having children remains uncertain.

But how much value do women attach to paid work? Some authors argue that McDonald (2000a, 2001a) gives undue emphasis to the importance of paid work in women's lives (Hakim 2001; Manne 2001). In the United Kingdom, Hakim (2000) concludes that, while some women are career-centred (giving priority to work rather than raising children) and others are home-centred (giving priority to family and thus preferring not to work), the majority try to achieve a balance between the two. This debate highlights the relevance to fertility trends of values and beliefs about children's needs, and has sparked lively discussions about the appropriateness of some policies directed towards supporting women (or couples) in their roles as paid workers and parents.5

In summary, it seems quite easy to identify many potential financial and work-related costs that may contribute to the falling fertility rate. Most of those discussed above are tangible costs, but the spectre of the intangible may also deter couples from having children, encourage them to postpone starting a family, or defer making decisions about having children.

Dealing with the "unknowns"

McDonald (2000a) points out that economic and non-economic (psychosocial) costs of having children can be difficult to decipher, and that negative "unknowns" may encourage individuals to err on the side of caution. For instance, people may decide against having children because of uncertainties about job stability, housing prices, interest rates, or about chances of returning to paid work should they take time off to care for a child. Some couples may worry about the possible disruptive effects of a child on their relationship, the possibility that this child may follow worrying pathways through life or be harmed in some way, and the difficulties in coping with raising a child.

Kohler et al. (2001) point out that postponing having children can reduce some of the uncertainties, such as those relating to financial and relationship stability. They further note that starting a family is irreversible while planning or lack of planning is not. In their view, this situation encourages couples to postpone the decision to have children. Over time, choices thereby become increasingly restricted to having few if any children - and for those who postpone too long, there is no choice.

Costs linked with societal norms and personal values

Some authors have argued that people have increasingly emphasised "post-materialist values" and that such values are incompatible with parenting (see Coleman 1999; McDonald 2001a). These values include achieving self-realisation, autonomy, and freedom from the bonds of traditional forces, including religion. But in a similar, though more moralistic vein, the above-noted "Mackellar Commission" that was established around the turn of the 20th Century, concluded that the fall in the birth rate was ultimately caused by "growing selfishness", as reflected in part by "a love of luxury and of social pleasures" (Hicks 1978: 23). Furthermore, McDonald (2000b) questioned the notion that post-materialist values explain low fertility rates on two grounds: among industrialised countries, more liberal societies have higher fertility rates compared with traditional societies, and surveys suggest that women want to have more children than they end up having. In his view, low fertility can be explained in terms of the combined forces of costs, uncertainty, and social institutions that make it difficult for women to combine work and family life.

Perceptions of the financial costs of children are also inextricably linked with people's values. Allan, Hawker and Crow (2001) maintain that today's parents (in Britain and Europe) express greater concern about the quality of their children's everyday experiences, their emotional development and educational achievement compared with parents in earlier times. Consistent with these observations, Australian time use surveys suggest that parents are investing more time in fewer children (Bittman 2002). Given these issues, children are expensive in terms of time, money and energy. In an era in which there is growing economic inequity involving "winners and losers" (see Birrell et al., 2004), it is understandable that parents may invest more heavily in fewer children.

These and other macro-level forces may also contribute to a possibly growing emphasis on the part of young people to be reluctant to commit to a pathway of no return. Mackay (1997) suggests that, as a consequence of having grown up in an era of relatively constant and rapid technological, social and cultural change, men and women born in the 1970s are bent on keeping their options open. Ironically, as Kohler et al. (2001) also noted, a consequence of "drifting" is that the doors of having children are closing.

Changing values attached to children and parenting may also add to the costs of having children. King (2002) argues that Australian society has become increasingly intolerant of children and that parents tend to be disparaged, while in the United States, Crittenden (2001) maintains that a job devoted to nurturing a child full-time tends to be equated with "doing nothing". The diminishing importance of motherhood in women's lives is revealed in three Australian surveys. The proportion of married women under the age of 35 who agreed with the statement, "Whatever career a woman may have, her most important role in life is still that of becoming a mother" fell from 78 per cent in 1971 to 46 per cent in 1982 (McDonald 1983), and to 26 per cent in the 1991 Family Formation Project conducted by the Australian Institute of Family Studies.

Also potentially relevant to decisions about starting a family are prospective parents' beliefs about whether young children need full-time parenting and, if not, whether they can access suitable child care. Research by Evans and Kelley (2002) suggests that most Australians believe that mothers should stay home full-time to care for children under school age. Under these circumstances, the psychological costs of using non-parental child care for children under school age may be high for some parents. The psychological costs might also be high for working parents who reject the notion that young children need full-time parenting, but who are unhappy about the suitability of services available. According to Wise (2002), parents' judgments of "suitability" of child care are based on a range of issues, including the warmth displayed by carers, and health and safety issues. She also notes that some of the issues emphasised by parents vary according to their cultural background.

Drago, Scutella and Varner (2002) have specifically linked fertility with clashes in values. According to these authors, the trend towards part-time work for mothers in Australia reflects the presence of two beliefs that have a negative impact on fertility: that responsible mothering cannot be achieved with full-time work, and that ideal workers are committed to uninterrupted employment involving long work hours. These authors maintain that such beliefs lead many career-oriented women to opt out of having children.

As mentioned above, the fall in fertility may itself generate social norms that further reinforce the fall - norms that would not necessarily increase the costs of having children. This issue, along with other social effects and personal beliefs that might contribute to the fall in fertility, will be considered next.

1.4 Other social influences and personal beliefs

Changing life course patterns that result in smaller families and childlessness themselves lead to greater tolerance and eventually become "norms", shaping the expectations and preferences of those entering adulthood. Increasingly, partnering and having children have become "choices" rather than inevitable pathways in life. Furthermore, as family size falls, those who decide to have children will be increasingly likely to restrict their choice between having only one or two children. Indeed, above-noted evidence from Europe suggests that the size of a family that young adults consider to be ideal may fall below two children, if it has not already (Goldstein et al. 2003).

The decision to delay having a family may also be influenced by beliefs about the success of assisted reproductive technology as women (and men) age. Successes, not failures, tend to be publicised. Little is known about the prevalence of such beliefs, how they may have shifted with technological advances, and their impact on fertility decision-making.

The experiences of personal friends and other associates are also likely to play a major role in influencing family formation decisions. In fact, Kohler et al. (2001) argue that the tendency of prospective parents to draw on the experiences of others in their social networks has contributed to the postponement of decision-making and, consequently, low fertility rates. They maintain that prospective parents prefer to wait until they get a good idea of how others fare. But this takes time thus further delaying the decision.

1.5 Drawing some threads

In this discussion so far, attention has been paid to a complex and multi-layered set of factors that may help explain the decline in fertility. McDonald (2000b; 2001a) has attempted to integrate some of this literature by categorising the explanations for low fertility provided by Coleman (1998) into four theoretical perspectives. These have already been mentioned above, but will be restated here, given their relevance to the Fertility Decision Making Project.

One of these perspectives focuses on "post-materialist values". It assumes that people have become increasingly concerned about achieving self-realisation, autonomy and freedom from the bonds of traditional forces including religion. Arguments that women are selfish in not having children fit within this category of explanations. According to McDonald, low fertility is better explained in terms of the combined forces of costs, uncertainty, and the difficulties women face in attempting to combine paid work and family life.

According to the "rational choice" (or "demand") theory, people weigh up the costs and benefits of having a first or additional child. Economic consequences represent costs only, while psychological repercussions may include costs and/or benefits. McDonald proposed that there may be several psychological benefits associated with having the first child, including achieving the status of being a parent and being a "family", and meeting the expectations of other people. However, he argued that increasing age might deter couples from having additional children, because psychological costs may rise or psychological benefits may fall.

Financial costs include direct expenses incurred by having children as well as lost earnings resulting from the need to care for the child ("indirect costs" or "opportunity costs"). According to McDonald, indirect costs are particularly salient in decisions about having a first child, while direct costs are more important in decisions about having additional children.

In his discussion of the "rational choice" perspective, McDonald mostly focused on psychological benefits and financial costs. Women who feel that having a child would interfere with their pursuit of a personally rewarding career would be confronted not only with indirect financial costs but also with direct psychological costs.

Whereas the "rational choice" perspective assumes that people are aware of the rewards and costs of having children and weigh these up in arriving at a decision, McDonald noted that the existence of uncertainties is at the core of another set of explanations, called "risk aversion theory" or "theory of risk and opportunity". This perspective maintains that there are many uncertainties surrounding the impact of having children. If people feel unsure about what the future may hold, then they will avoid the insecurities associated with having children and invest in any security that is within their reach, such as education and employment.

The fourth perspective, called "gender equity", focuses on the difference in women's opportunities relative to men's opportunities before and after they have a child. McDonald maintains that, in countries where educational and work opportunities for women are similar to those for men but are seriously diminished for women if they have a child, then women will have fewer children than otherwise. In McDonald's view, lack of "family-friendly" work policies, high cost child care and tax benefits that operate as disincentives for taking up part-time work combine to encourage women to have fewer children than they wish to have. However, it seems very likely that these various considerations would also help shape men's views about having children. Indeed, McDonald argued that surveys suggest that most couples want to remain dual income earners.

Thus, in some form or another, all four explanations take into account the costs of having children - with the "post-materialist" perspective emphasising the loss of personal freedom and impediments to self-interested goals (which are likely to require two incomes for couples); the "rational choice" perspective suggesting that the decision to have a child is an outcome of a costbenefit analysis; the "risk aversion" perspective focusing on concerns about uncertain costs; and the "gender equity" explanation taking into account curtailment of employment opportunities and thus income for women, along with direct financial costs of having children.

1.6 Conclusion

Australia's total fertility rate is well below replacement level, at an all-time low, and may continue on a downward slide. This trend has created considerable angst, given its impact on the ageing of the population and the projection that, if the rate falls below 1.6, the population will decline this century (ABS 2000a).

Of course, not everyone agrees that the economic and social consequences for Australia of this scenario are necessarily dire. For example, Hamilton (2002: 1) argues that: "Population growth will not make us richer in economic terms and it will almost certainly make us poorer in terms of environmental amenity".

Nevertheless, if the master plan is to prevent total fertility from falling below 1.6, a good understanding of the factors at play that are driving the birth rate down are required. These factors appear to be complex and often mutually reinforcing, with the most fundamental being postponement of first births and consequent shortened childbearing years, and increased risk of having no children at all. The factors appear to include broad technological, structural, cultural and social changes, shifting pathways of friends and associates, changes in personal financial and life course circumstances, and shifts in the beliefs and values of prospective parents.

However, there is a great deal of controversy about the existence or relative importance of some issues, such as increasing selfishness and the significance of women's employment in decisions about having children. Little is also known about other issues, such as men's contribution to childbearing decisions and the impact on future childbearing of disagreements between partners. Furthermore, the nature and relative importance of forces deterring couples from having children will vary somewhat for different sub-groups in the population, and understanding some of these sub-group differences and why they may be changing is a long way off.

Under these circumstances, it is little wonder that governments are struggling to develop policies that might curtail the falling birth rate. Perhaps the greatest hope stems from the fact that, despite all the forces operating against having children, most people want to become parents (Fisher 2002; Qu and Soriano 2004) and it appears that prospective parents want and expect to have more children than they end up having (McDonald 2001a; Quesnel-Valee and Morgan 2002). As noted above, there is evidence to suggest that in some European countries, the number of children considered personally ideal has already fallen below replacement (Goldstein et al., 2003).

1.7 The Fertility Decision Making Project

The Fertility Decision Making Project, a collaborative project of the Office for Women and the Australian Institute of Family Studies, was designed to examine ways in which some of the broad social forces suggested in this review are translated into the decisions individuals make about having children - the "micro-level" dynamics of their everyday choices. The broad aim of the study was to enhance understanding of the reasons underlying the fertility decisions of men and women aged 20-39 years, both as individuals and couples.

Key issues examined included the nature of fertility aspirations and expectations, ways in which aspirations and expectations are related, and the importance individuals attach to various costs, benefits and risks (including employment-related and financial issues) when considering having children. The extent to which these views vary according gender, age, relationship status, educational level and employment status was explored.

The project thus led to examination of a number of policy-relevant specific questions, for example:

As de Vaus (2003) has pointed out, adding this subjective dimension will help us gain a better understanding of the factors behind declines in the fertility rate. The Fertility Decision Making Project was designed to achieve this aim and thereby provide policy with a greater evidence base.

However, research into the subjective dimension is beset with difficulties in defining and measuring psychological concepts. This problem is apparent in the research into key issues that are examined in this report: aspirations and expectations.

Interpreting "aspirations" and "expectations"

The concepts, "ideals", "preferences", "desires", "expectations" and "intentions" are related cognitions and have been variously examined in theories and studies of decision-making. The concepts "expectations" and "intentions" are often used synonymously (see Quesnel-Valee & Morgan 2002). Regardless of the possible inaccuracy in this approach, these two concepts are certainly closer in meaning to each other than to "preferences", "desires" or "ideals". (Throughout this report, the terms "preferences" and "desires" are used interchangeably.)

In the Fertility Decision Making Project, the question on respondents' ideal number of children was designed to tap into inclinations that, as much as possible, are removed from individual circumstances and other life goals. It seems likely that "desires" for another child are more influenced by current circumstances (including competing goals), compared with personal ideals. On the other hand, "expectations" are likely to be the most influenced by such circumstances.

These arguments diverge somewhat from those of Miller (1994). In his model of fertility decisionmaking, Miller focused on "intentions" (rather than "expectations") and "desires". He argued that "desires" refer to what one wants, whereas "intentions" refer to decisions made based on personal desires that are constrained by reality (with partner's views representing one potential constraint). To Miller, reproductive behaviour is influenced by "intentions".

Azjen (1985, 1988, 1991) also emphasises that intentions are influenced by perceived constraints. According to his theory of planned behaviour, intentions shape behaviours that are under volitional control - with such circumstances applying to most everyday behaviours. Intentions are thus seen as instigating behaviour when the opportunity to engage in that behaviour arises.

However, life is more complex than that suggested above. Although "ideals" may be the most removed from the practical realities of life and thus the least predictive of behavioural outcomes, they are not necessarily divorced from constraints. Firstly, it can be difficult to envision possibilities that are very remote from personal life experiences - an issue relevant to the "structure versus agency" debate. Secondly, "ideals" may not only be strongly affected by societal norms to some extent but are also likely to be influenced by personal life experiences. Thirdly, perceived constraints may help shape ideals and preferences through a process of defensive reappraisal or dissonance reduction (Festinger 1957; Lazarus & Folkman 1994).

Indeed, the extent to which personal circumstances and alternative aspirations are taken into account when establishing ideal or desired family size is likely to vary across individuals. For instance, some individuals may view two children as ideal for them because it is no longer possible for them to have a larger family or because they would not be able to afford private education for more than two children. Others may take less account of their competing goals and practical constraints when establishing their "ideal" family size, while acknowledging that this ideal is unachievable. In other words, "ideals" may not mean the same thing to all respondents in a survey.

Nevertheless, there is evidence to suggest that "ideals" and "desires" are further removed from, and thus less predictive of, behaviour compared with intentions or expectations (see Van Peer 2000). This does not mean, of course, that measurement of ideals and preferences has little policy relevance. Of those who do not intend having children (at least in the near future), some will want children and others will not. Of the latter, there will be some who have never wanted children, but not all of these people may continue to feel this way.

In the following chapter, design and sample of the Fertility Decision Making Project is outlined. Chapter 3 then provides a socio-demographic snapshot of the current situation of our respondents in terms of the following key factors: gender and age, relationship status, educational level and employment status. The next three chapters focus on fertility aspirations (Chapter 4), fertility expectations (Chapter 5), and the relationship between the two (Chapter 6). In Chapter 4, the extent to which respondents' views about having children had changed over time is explored. The final set of analyses (Chapter 7) examines the extent to which various potential costs, risks and benefits are taken into account when considering having children.

In each chapter, the extent to which the views of respondents about having children vary systematically according to the above-noted five factors (gender and age, relationship status, educational level and employment status) is assessed. Many other factors no doubt play an important role, but their examination is beyond the scope of this report. The final chapter draws together all these sets of analysis to address the key issues that formed the basis of this study and associated policy implications.


  1. The total fertility rate is a synthetic measure that refers to the number of children a woman would have if she experienced the current age-specific fertility rates across her reproductive life. [back]
  2. While the distinction between "voluntary" and "involuntary" childlessness is at times difficult to draw given that some women voluntarily postpone having children until they reach an age when fecundity problems develop, for others external circumstances hinder their chances of having children. Nevertheless, McDonald (2001a) estimated that 7 per cent of people are infertile, although this trend increases with age. [back]
  3. See edited transcript of interview by Kerri Phillips for the ABC's The Europeans program, printed in Family Matters, No.63, pp.50-59. [back]
  4. So far, little attention has been paid to men's aspirations and how they may change. [back]
  5. See Family Matters 2002, No. 63, for edited transcript of Keri Phillips' interview with McDonald and Hakim for the ABC's Europeans program. [back]

Research Report 11: Contents | Previous | Next