Summary

The fertility rate in Australia, like almost all Organisation for Economic Co-operation and Development (OECD) countries, is below the level required for population replacement. This has resulted in an increasingly active debate surrounding possible reasons behind the fall, future likely trends and realistic strategies to stem it.

This paper provides an overview of trends in fertility in Australia and the potential implications of these trends. The various explanations that have been provided for fertility trends and key family policies are discussed.

A number of explanations for the decline in fertility in developed countries have been proposed, yet the topic remains hotly debated. There is growing evidence that a range of factors are related to fertility rates, although there does remain uncertainty as to which factors are most important in explaining the falls. While studies have produced differing results, there does seem to be growing evidence that social and economic policies do have an important role in stemming the declines in fertility rates.

The link between the level of public financial support and fertility rates for OECD countries are discussed. Government support to families with children as a proportion of GDP has increased rapidly in Australia over the last 25 years, with the level having gone from being at the lower end for OECD countries to being well above the average.

The paper then reviews findings from the Fertility Decision Making Project. This project is based on the results of a major national survey of Australian families that was conducted in 2004. The study provided information on the factors that individuals report as affecting their fertility decisions. It was revealed that most people want to have children. Furthermore, the number of children they would ideally like to have is above replacement fertility on average. It also revealed that the number of children they want to have is greater than the number they expect to have. The reasons for respondents revising their family size aspirations downward as they grow older include financial and work-related reasons, partnership issues, age, health and fecundity.

We conclude that policies that lower the direct and indirect costs to families of raising children, and allow women to combine paid employment with childrearing are likely to have a positive impact on the fertility rate (that is, stem the decline or "boost" the rate). Policy changes in Australia over the last 25 years have reduced the direct costs of childrearing through expanding the coverage of the family payment system and increasing the level of some payments. The recently introduced maternity payment also provides significant financial support to families following the birth of a child.

In order to enhance fertility rates, it is important that families have support as the child grows up, not just when the child is a pre-schooler. Parents also need to feel confident that they can manage to raise their children while also enjoying opportunities for personal development, beyond the world of family life. Despite Australia's economic prosperity, people remain concerned about their capacity to create and maintain a family environment in which children can be nurtured and supported financially and emotionally. Such concerns, real or perceived, reflect macro-level trends in economic and employment security; relationship formation pathways and their stability; and micro-level concerns about personal capacity to be a good parent. Strategies directed towards helping people achieve their childbearing aspirations need to tackle people's sense of security in each of these three domains. The data suggest that strategies should target forces both at the macro- and micro-levels - for neither level acts in isolation.

In conclusion, governments need to use a combination of approaches that is based on the recognition that the relatively low fertility rate is not due to a "lack of wanting children". If Australia is to boost its fertility rate - or at least maintain the current level - the message that raising children has an intrinsic richness and is an enjoyable part of life needs to be conveyed widely. To be effective, however, such a message must reflect reality. Couples need to have personal resources such as a secure income stream, a loving and stable relationship, and the skills and confidence to be parents. They also require access to community resources, including family-friendly workplaces and the confidence that they have a strong, continuing commitment from the community. It is of the utmost importance that parents do not feel alone in raising the next generation of citizens.

1. Introduction

Since 1977, Australia, like almost all Organisation for Economic Co-operation and Development (OECD) countries, has witnessed its total fertility rate (TFR) fall below the level required for population replacement.¹ Over the last 10 years, this evidence has resulted in an increasingly active debate surrounding possible reasons behind the fall, future likely trends and strategies to stem it. These discussions are vital in comprehending the potential implications of current and future trends for Australia's longer-term social and economic future.

There has been extensive discussion about the impact of ageing of the population, including potential decreases in the size of the labour force and the total population, the future of immigration, and environmental sustainability. Issues regarding work and family, women's participation in the paid workforce, child care, paid maternity leave and the structure of family payments have also come into question. Such debates, as cited by Stanton in 2002, have sparked eye-catching newspaper headlines, such as "Populate or stagnate", "Procreate or perish", "Marry and multiply" and "Adapt or depopulate".

Australia's recent fertility rates may have been the lowest on record, however, unease about flagging fertility rates is by no means a modern concern. In 1904, a Royal Commission in the Australian state of New South Wales was established to seek reasons for the dramatic fall in fertility rate (the Royal Commission on the Decline of the Birth-Rate and on the Mortality of Infants in New South Wales). Then, in 1942, an official inquiry began into the low birth rate - this time by the National Health and Medical Research Council of Australia, which considered the rate to be a problem, "such as to cause, even now, the gravest anxiety about the future of the Australian people" (cited in Stanton, 2002, p. 3).

A number of explanations for the contemporary decline in fertility in developed countries have been proposed. There is growing evidence that a range of factors are related to fertility rates, although there does remain uncertainty as to which factors are most important in explaining the falls. While studies have produced differing results, there does seem to be growing evidence that social and economic policies do have an important role in stemming the declines in fertility rate (e.g., McDonald, 2006b; Sleebos, 2003).

This paper first outlines trends in fertility, their implications for Australia and the various explanations that have been provided for fertility trends. The potential explanations emphasise the importance of macro-level forces in affecting childbearing decisions, although the precise impact of the different forces is difficult to identify. Second, key government policies directed towards supporting Australian families are outlined. Third, the results of a major national survey undertaken in Australia that explored factors that impact upon decisions about fertility are discussed. Finally, the policies that appear to be likely to influence fertility decision-making are discussed.

2. Fertility rates, demographic structure and population projections

2.1 Fertility trends

Figure 1 plots the TFR for Australia from 1921 onwards, while Figure 2 "amplifies" the changes that took place in the TFR between 2000 and 2005. At the beginning of the 20th century, the TFR was close to 4 babies per woman, having fallen by about one-third over the previous 30 years (Hugo, 2001). The rate fell to 2.1 babies per woman during the Great Depression in 1934 - a trend that apparently resulted more from deliberate birth control within marriage than from the postponement of marriage that occurred at the time (Ruzicka & 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, as well as the influx of immigrants of childbearing age. By 1961, the fertility rate was at a record high - 3.5 babies per woman - but then fell for the first time to below 2.1 (the replacement level) in 1976. The rate then stabilised in the late 1970s and 1980s to between 1.8 and 1.9, and fell in small progressive steps in the 1990s. The rate was at its lowest level on record in 2001 (1.73), but had increased to 1.81 by 2006. Overall, however, the trend has been fairly steady since the late 1990s, fluctuating between 1.7 and 1.8.

Figure 1: Total fertility rate, Australia, 1921-2006

Sources: Australian Bureau of Statistics [ABS] (various years)

Figure 2: Total fertility rate, Australia, 2000-2006

Sources: ABS (various years)

Figure 3: Age-specific birth rates, Australia, 1921-2006

Sources: ABS (various years)

Figure 3 shows the age-specific fertility rates for women since 1921. Between 1960 and the mid-1970s, the fertility rate fell in all age groups, particularly for those in their early and late twenties and those in their early thirties. From the late 1970s, the rates continued to fall for women under the age of 30 (especially those between 20-24 years), but increased for women aged 30 or over. From the 1980s, a slight increase was seen in women aged 40-44, however this rate remained very low. Since the early 1980s, the increase in fertility rates for women aged 30 years and over has been significant enough to offset the decrease in rates for younger women, though less so in the 1990s. More recently, the level of increase in the rates for older women has more than compensated for the level of decrease in the rates for younger women. Nevertheless, for all age groups except women aged 35-39 years, the fertility rate in 2006 was lower than in 1960.

The contemporary decline in the fertility rate overall stems from the fact that an increasing proportion of women either remain permanently childless or are having only one or two children. Between 1981 and 2006, the proportion of women aged 40-44 years who had two children increased from 29 to 38 per cent, while the proportion that had four or more children fell from 28 to 11 per cent. The proportion of women in this age range who had three children also declined (from 27 to 22 per cent), while the proportion having only one child fluctuated (between 8 and 13 per cent). There was also an increase in the proportion of women who remained childless (from 9 to 16 per cent).

Women who give birth over the age of 30 are increasingly likely to be first-time mothers - 41 per cent of all first births in 2003 were to women in this age range, compared with 28 per cent in 1993 (Australian Bureau of Statistics [ABS], 2001b; Laws & Sullivan, 2005). As a result, for many women the timeframe in which they can achieve their preferred number of children is progressively shortened. However, it is worth noting that women in their thirties are now 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.

A limitation of the TFR and cross-sectional age-specific fertility rates is that they do not capture the long-term experience of a birth cohort. Examination of the long-term experience of birth cohorts shows a continued decline in the completed cohort fertility rate (CCFR) for all cohorts born since 1932. The most recent cohort for whom the CCFR is available is the 1960 birth cohort, which had a fertility rate of 2.15 children per woman (Kippen, 2004).

2.2 Australian fertility trends compared with those in other countries

Despite the alarm bells ringing within Australia, its current TFR is considerably higher than those in a number of other countries. Table 1 shows that fertility rates vary from 1.1 in South Korea to 2 or higher in New Zealand, USA and Iceland. Australia's fertility rate at 1.8 is at the higher end for OECD countries and is comparable with a number of European countries, particularly those in Scandinavia (other than Iceland). One measure of the demographic impact of fertility rates is provided by the projected change in the population between 2006 and 2050. The long-term fall in TFR to very low levels in some countries and difficulties experienced in attempting to reverse this trend have further fuelled the concerns felt in Australia (McDonald, 2001). As McDonald notes, such trends are inconsistent with the commonly held assumption that below replacement fertility would be a temporary phenomenon.

Notes: Projected populations based upon reasonable assumptions on the future course of fertility, mortality and migration. Projections are based upon official country projections, series issued by the UN or the US Census Bureau, or PRB projections. 
Source: Population Reference Bureau (2006)

2.3 Population age structure

While the fertility rate in Australia has fallen to between 1.7 and 1.8 since 1994, life expectancy is at an all-time high, having increased by more than 20 years since 1901. Boys born at the beginning of the 20th century could expect to live for 55 years and girls for 59 years, while those born in 2001-03 can expect to live for 78 and 83 years respectively (ABS, 2001b, 2006b). Although the huge reduction in infant and child mortality is a key factor, the significant increase in life expectancy has also occurred in middle and older ages since the 1980s (Jackson, 2007). For example, men and women aged 50 years at the turn of the 20th century would look forward to another 21 and 24 years of life respectively, while today men and women of 50 years could expect 30 and 34 years of life ahead respectively.

The fall in fertility rates and the significant improvement in life expectancy have inevitably resulted in an "ageing" of the population, both in absolute and relative terms. The median age of the total population has increased by nearly 13.8 years, from 22.6 years in 1901 to 36.4 in 2004 (ABS, 2006a; Hugo, 2001).

The ageing of the population is captured in the changing shape of the so-called "age-sex pyramid", presented in Figure 4 for the years 1911, 1961 and 2004. In 1911, there were relatively many children and relatively few elderly people, and so the picture resembles a pyramid. In 1961, there was a swell in the population aged 15 years or under, representing most of the post-war "baby boomers" - defined by the ABS as residents born in Australia or overseas between 1946 and 1965 inclusive (ABS, 2005). By the end of 2004, this group was between 38 and 58 years old. Despite the relatively low fertility rates of baby boomers, the fact that they are a large cohort means that their children represent a large cohort - a factor which, along with increased life expectancy, has resulted in an age-sex profile that no longer resembles a pyramid.

Figures 4: Age and sex structure of Australia's population, observed (1911, 1961, 2004) and projected (2051)

Notes: The baby boom population (born 1946-1965) is highlighted in grey. The 2051 age-sex structure is based on the ABS "medium" set of assumptions: TFR equal to 1.7 from 2018; net overseas migration equal to 110,000 from 2004-05; and life expectancy of 84.9 and 88.0 for men and women respectively from 2051-52. 
Sources: ABS (1997, 2005)

Figure 4 also depicts the age-sex structure for 2051, as projected by the ABS on the basis of a set of specified assumptions regarding the TFR, life expectancy and net overseas migration (see note to Figure 4 for details). Under these assumptions, the age-sex structure in 2051 would bear no resemblance at all to a pyramid.

Figure 5 depicts the changing proportions of the population since 1901 of those under the age of 15 and over the age of 65 (here called "older people"), and future proportions in these same age groups to the year 2101, according to ABS projections. The representation of children under the age of 15 years in the total population is projected to fall from 20 per cent in 2004 to between 13-16 per cent by 2051. The growth in numbers of children aged 15 or under is projected to be much slower than the overall population growth.

In contrast, the size of the population aged 65 years and over, which in 2004 accounted for 13 per cent, is expected to continually grow at a faster pace than that of the total population. By 2051, the ABS has projected that older people will represent 26-28 per cent of the population. Similarly, the ABS projects that the increase in the number of people aged 85 and over will be faster than the increase of the overall population.

Figure 5: Proportion of population aged under 15 and aged 65 or over (1901-2004), and projected proportion (to 2100)

Notes: The projections are based on the ABS "medium" set of assumptions: TFR equal to 1.7 from 2018; net overseas migration equal to 110,000 from 2004-05; and life expectancy of 84.9 and 88.0 years for men and women respectively from 2051-52. 
Sources: ABS (1997, 2005, 2006a)

The "working age" population is traditionally defined as being aged 15-64 years, although in practice many of those aged 15-19 years remain in education, and workforce participation rates fall considerably after 55. While the proportion of the population of working age is currently growing (61 per cent in 1901, 64 per cent in 1976, and 67 per cent in 2004), according to ABS projections it will begin shrinking in the next five years and return to around 60 per cent between 2031 and 2041 (Access Economics, 2001; ABS, 2005).

3. Implications of these trends

In the past, 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 shifted to address implications of low fertility rates for the age structure of the population (e.g., Productivity Commission, 2005).

An important question 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 the first of the baby boomers turned 60 years old in 2006.

In relation to this issue, the Intergenerational Report 2002-03 (Treasury, 2002) identified emerging fiscal issues over the longer term associated with an ageing population. While the report suggested that Australia is well placed to meet the challenges of an ageing population, it noted "the current generation of taxpayers is likely to impose a higher tax burden on the next generation" (p. 1). Some social commentators have argued that such circumstances will create growing resentment between the generations (Encel, 2002).

Demand for services will also change, given the different needs of the elderly and younger populations in such areas as housing, health care, leisure and education. There is considerable evidence to suggest that families are the most significant support network for the elderly. The increasing rate of childlessness, coupled with family breakdowns and children pursuing jobs interstate or overseas, will mean that many elderly parents will be either "functionally" or "actually" childless (Rowland, 2003; Weston, Qu, & Soriano, 2003).

The workforce itself will age if policies directed towards encouraging later retirement are effective. This will increase wisdom shaped by an accumulation of knowledge; however, there will be a relative loss of young adults, whose age-related talents often produce important technological advancements (McDonald, 2002; McDonald & Temple, 2006).²

In addition, concerns have been expressed about the repercussions of a stagnation or fall in the size of Australia's labour force (McDonald & Kippen, 2000) and the likelihood that, if the fertility rate fell below 1.6 and Australia's current net immigration levels were sustained, then the absolute size of Australia's population would shrink - a trend that would spiral over time (ABS, 2002; McDonald, 2000a). For those concerned about the impact of Australia's growing population on environmental sustainability, this may be seen as a welcome prospect.

Questions have been raised as to whether some of the challenges of a low fertility rate and an associated ageing population can be addressed through increasing immigration. 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, McDonald and Kippen showed that for levels of net immigration above 80,000 the magnitude of their effect on the age structure of the population diminishes.

4. Macro-level explanations for low fertility rates

Many interacting factors appear to have contributed to Australia's (and other countries) falling fertility rates. These include advances in reproductive technology, especially the development of the contraceptive pill; life course changes, such as delays in those transitions that typically precede childbearing; an overall fall in the formation of partnerships; and an increase in the rate of relationship breakdowns. Women's increased financial independence is also a factor.

Delays in achieving those milestones that typically precede having children restrict women's potential childbearing years to older ages, thereby further limiting the number of children they could have and increasing their risk of childlessness. Such changing life course and fertility patterns tend to become norms, influencing the expectations and preferences of those entering or currently in their childbearing years.

Behind some of these life course trends are changes in the labour market and the economy. There is evidence that economic downturns are associated with a lower level of fertility (e.g., Martin, 2003; Ruzicka & Caldwell, 1982). However, the effects of the economic cycle cannot explain the declines in fertility that have occurred in periods when the economic outlook has been relatively positive.

Several authors have suggested that the globalisation of the economy and associated labour market changes of the last two decades have resulted in lower fertility rates. Low-skilled yet relatively highly paid jobs for early school leavers have virtually disappeared, having been replaced by jobs entailing fixed-term contracts and part-time or casual hours, thereby providing limited economic security (Kohler, Billari, & Ortega, 2001; McDonald, 2000b; Saunders, 2001). McDonald (2001) also argued that this era of job insecurity has been accompanied by a strong economic cycle of "boom and bust" and rising or fluctuating house prices that not only lead young adults to invest in their education and career development, but also encourage couples to maintain dual incomes.

The costs of having children are inextricably linked with these broad structural forces. At the same time, several authors have suggested that since the introduction of child labour laws the benefits of having children have been restricted to psychosocial ones. Such benefits, however, do not accumulate as family size increases (Kagitcibasi, 1997; Kohlmann, 2002).

The costs of having children may be conceptualised as being direct or indirect and financial or non-financial. To some extent, direct financial costs tend to be reinforced by improvements in living standards, for these may lead to today's luxuries becoming tomorrow's necessities. Indirect financial costs include having reduced earnings and potentially curtailed careers when caring for children takes precedence over paid work (foregone earnings). Indirect non-financial costs include any lost or diminished opportunities for social relationships, mental stimulation and prestige, which are linked with giving up paid work or substantially reducing work hours, while direct ones include difficulties experienced by parents in managing both paid work and family responsibilities.

In addition, it is sometimes argued that parents nowadays are more likely than parents of the past to invest much time and energy in providing their children with experiences that they feel will enhance their children's emotional development and educational achievement (e.g., Allan, Hawker, & Crow, 2001). Consistent with these observations are Australian time use data that suggest that parents are spending more time with fewer children (Bittman, 2002). This approach to parenting may be seen as a direct non-financial cost to the parents, although the longer-term benefits for their children may be substantial.

Other changing values that may be incompatible with parenting have been proposed. These include the suggestion that there is now an enhanced emphasis on achieving self-realisation, autonomy and freedom (Coleman, 1999); a desire on the part of young adults to keep their options open (Mackay, 1997); society's increased intolerance of children; and a diminished value attached to parents, especially mothers (e.g., Crittenden, 2001).

Some of these cost-related explanations for low fertility apply particularly to women. McDonald (2000b, 2001), for example, argued that in many countries "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 limited if they have children, thereby discouraging them from having the number of children many would like to have. Some authors have argued that this is a key area in which government policy can affect fertility rates (e.g., Andersson, 2005; McDonald, 2000a, 2000b). It seems reasonable to suggest that the gender equity explanation can be extended to marriage itself. In OECD countries where women's opportunities for enjoying a self-fulfilling lifestyle are seriously curtailed by marriage, fewer women are likely to get or remain married.

Costs of having children may also be known or unknown. McDonald (2000a) pointed out that the financial and psychosocial costs of having children can be difficult to decipher, and that negative "unknowns" themselves encourage individuals to err on the side of caution and thus discourage them from having children.

Many of the factors that are thought to influence fertility can be affected, to a greater or lesser degree, by government policy. There are a number of studies on the impact of policy on fertility rates. However, estimating the impact is challenging, requiring a number of methodological issues to be addressed.

First, the range of policies that can potentially affect fertility rates is broad. Policies that may be relevant include the tax and income support system, child care policies, educational policies, and policies that affect parents' ability to balance family and work responsibilities.

Second, often a number of policy changes are implemented at the same time. This makes identifying the separate impact of individual policies difficult. Disentangling the influences of individual policies is made more difficult by the fact that the intended outcomes of specific policies may take a long time to take full effect.

Third, it is likely that some of the variables that are thought to affect fertility decisions will also be affected by decisions about fertility (that is they are endogenous). For example, women's childbearing decisions affect labour supply decisions and labour supply decisions affect decisions about childbearing. In addition, endogeneity may also be present at the macro-level. For example, it may be hypothesised that higher levels of financial transfers to families with children increase fertility rates; however, it may be the case that there is some other factor that affects both fertility rates and the level of transfers to families with children. An example of this may be a growing public concern about low fertility rates are and hence an increased value being placed on children.

Fourth, a number of the explanatory variables that are thought to influence fertility decisions and are affected by policy are difficult to measure accurately. A good example of this type of variable is the private cost of children.

Despite the difficulties in an assessing the effectiveness of government policies, there is evidence that they can have an impact on fertility. Laroque and Salaníe (2005), using data for France, found that family benefits have reduced the cost of raising children and consequently have increased the fertility rate. They estimated that a decrease in the private cost of children by 25 per cent would increase fertility by about 5 per cent. In a recent study, d'Addio and d'Ercole (2005) used panel data from 16 OECD countries for the period 1980-99 to show that a higher level of government payment to families was associated with a higher fertility rate, as were higher replacement wages during parental leave, higher female employment rates and a higher share of women working part-time. Lower fertility rates were associated with higher unemployment rates and with children having a larger negative impact upon mothers' earnings (foregone earnings).³ On the other hand, Sleebos (2003), following a detailed review of research that examined the relationship between policy and fertility rates, concluded that "the evidence provided by this review seems to suggest a weak positive relation between reproductive behaviour and a variety of policies" and that "findings are often inconclusive or contradictory, partly because of methodological differences" (p. 48). Grant et al. (2004) concluded that the literature for France suggests that the increase in fertility rates "is less attributable to a single policy mechanism than to its ability to create an environment that encourages childbearing. This environment is created by a combination of policies that jointly serve this aim" (pp. xv-xvi).

This brief review suggests that the factors explaining the fall in the fertility rate are complex and often mutually reinforcing. One of the most fundamental of these is the postponement of first births, which consequently shortens possible childbearing years and increases the risk of having no children at all. The factors appear to include broad technological; structural, cultural and social changes; and changes in personal financial circumstances.

However, there is a great deal of controversy about the existence or relative importance of some issues, such as the significance of increasing individualism and women's employment in making decisions about having children. Little is known about some issues, such as men's contribution to childbearing decisions and the impact of disagreements between partners in relation to making childbearing decisions. Furthermore, the nature and relative importance of forces deterring couples from having children will vary somewhat for different sub-groups in the population (e.g., ethnic groups). The understanding about these sub-group differences is still a long way off.

The above discussion highlights the importance of broad contextual factors influencing fertility decisions. Government policy is part of this context, and although the extent to which they can affect fertility decision-making is debated, there is, in our view, growing evidence that policy does influence fertility rates.

5. Australian Government family policies

In this section, we focus on those government policies that aim to help families with children through direct financial support: cash payments and tax measures designed to modify income or provide basic income support, assistance with the costs of chid care, and maternity and paternity leave.⁴

The income support system in Australia differs in important ways from the system in most other OECD countries. As Whiteford and Angenent (2002) noted, "Consideration of government policy statements suggests that there are two long-standing values that provide the basis of the Australian income support system. One is the recognition of government and community responsibility to assist those in need. The other is that private provision outside the social security system is to be encouraged as far as possible, with the income support system seen primarily as a safety net. This distinguishes Australia from most other developed countries - the primary focus of Australia's social security system is protection against poverty. In most other OECD countries, the primary principle is one of income maintenance across an individual's life-cycle, although many have poverty relief as an important additional objective" (p. 13).

In Australia, income support benefits are flat-rate and paid from general government revenue.⁵ There are no earnings-related features in the government benefit system and payments are made on the basis of categories (e.g., unemployed, disabled, primary carer). There is also an extensive system of supplementary payments for families with children. This includes direct cash assistance for over 80 per cent of all families with children (with those excluded being the highest income earners), and higher levels of assistance for those receiving income support payments or in low-paid jobs. Other payments include the maternity allowance, assistance with child care costs, housing costs for those renting privately, health care costs, and a range of concessions on other services.

Benefits are subject to income and assets tests, but these tests are generous compared to the income and means tests applying to social assistance in other OECD countries (Whiteford & Angenent, 2002). Benefits (in one form or another) are effectively available on an indefinite basis, subject to the means tests. Because payments are not contributory, coverage of the system is universal, subject to a range of residence requirements.

The significance of family payments has increased substantially in recent years. While comparing the level of family benefit expenditure over the last 15 to 20 years is not straightforward since the structure of family payments has changed significantly, it is estimated that between the years 1993-94 and 2003-04, expenditure on family payments increased in real terms by about 115 per cent, from $7 billion to $15.3 billion in 2003-04 dollars (Ministerial Taskforce on Child Support, 2005).

5.1 Family payments

The core of the family payments system in Australia is the Family Tax Benefit (FTB). The system of family payments comprises:

• FTB Part A, a two-tiered payment linked to the number and age of children;
• FTB Part B, to provide extra help for families with one main income, including sole parents;
• Child Care Benefit (CCB) and Child Care Rebate, to assist families with their child care costs;
• Baby Bonus (formerly known as the Maternity Payment), to assist families following the birth or adoption of a child; and
• Maternity Immunisation Allowance, to encourage immunisation of children aged 18-24 months.

Families receiving FTB Part A may also be eligible for several additional components:

• Rent Assistance, for private renters;
• Multiple Birth Allowance, for the birth of triplets or more; and
• Large Family Supplement, for the third and subsequent children.

The base FTB Part A rate recognises the costs of children for all but the highest income parents.⁶ Many families with relatively low household incomes are also given a supplementary payment as a component of FTB Part A.⁷ It is provided to ensure that parents on low incomes have enough money to maintain their children adequately. In the financial year 2007-08, parents with a household income of less than $41,318 are entitled to the maximum rates of FTB Part A.⁸ Payments of FTB Part A are for each child.

FTB Part B provides additional assistance to sole-parent families and two-parent families with only one main income earner. The amount received is based on the age of the youngest child.⁹ Unlike Part A, Part B is not paid for each child. For a couple, this payment is not income tested on the higher earner's income, but on the income of the lower income earner. Under the FTB Part B income test, the lower earner can earn $4,380 each income year before the payment is tapered out at 20 cents for each dollar of income. For sole parents, there is no income test.

5.2 Parenting Payment

The Parenting Payment is an income support payment for both sole and partnered parents. However, it is only payable to one member of a couple. An alternative income support payment, such as Newstart Allowance (for unemployed people looking for work), may be payable to the other member of the couple.

A number of changes were made to the Parenting Payment from 1 July 2006. These changes were made as part of the Welfare to Work package - measures that are aimed at increasing the rates of paid employment of those in receipt of income support payments.

Principal carers who went on to Parenting Payment (Partnered) on or after 1 July 2006 receive the Parenting Payment until their youngest child is 6 years old. When their youngest child turns 6 years, these parents need to test their eligibility for another income support payment if they still require such support. People on unemployment-related payments who are principal carers have an obligation to look for part-time work of at least 15 hours per week. Principal carers who are single parents and went on to the Parenting Payment (Single) on or after 1 July 2006 are eligible for Parenting Payment until their youngest child turns 8 years. At that time, they also need to test their eligibility for another income support payment, usually an unemployment-related payment (e.g., Newstart Allowance).

Those who started receiving Parenting Payment before July 2006 are eligible to continue to receive Parenting Payment until their youngest child turns 16 years (subject to continuing eligibility). From 1 July 2007, eligibility to continue receiving Parenting Payment is subject to meeting a part-time participation requirement, or when their youngest child turns 7 years, whichever is later.¹⁰

5.3 Child care

Assistance with the costs of child care is provided through CCB. CCB is paid for hours of care used in approved or registered child care. The maximum rate of CCB ($3.37 per hour for children who are not yet at school) applies to those with a family income under $35,478 or who are in receipt of income support payments.

Additional assistance with the costs of children is provided by the Child Care Tax Rebate. The Child Care Tax Rebate covers 30 per cent of out-of-pocket child care expenses paid for approved child care, less any CCB received. The rebate is payable to a maximum of $4,354 per child per year. The Child Care Tax Rebate applies to child care costs incurred from 1 July 2004. The rebate is received at the end of the tax year in which the child care expenditure was made, that is, after a full year of eligibility for CCB has been completed.

5.4 Baby Bonus

The Baby Bonus is paid to families following the birth or adoption of a baby. It is payable to: a parent of the baby; families who have care of a newborn child within 13 weeks of the child's birth and who are likely to have care of the child for no less than 13 weeks; and families who have a child entrusted to their care for adoption before the child is two years of age. (In the case of overseas adoptions, families are eligible for this payment if the child enters Australia before his or her second birthday.) Usually, the payment is made as a one-off lump sum payment of $4,133. For multiple births, the Baby Bonus is paid for each child; for example, $8,266 is currently paid for twins. No income or assets tests apply. The Baby Bonus came into effect in the financial year 2004-05 at a value of $3,000. It will increase to $5,000 for babies born after 1 July 2008.¹¹

5.5 Maternity and paternity leave

Australia does not have a statutory system of paid maternity or paternity leave. Permanent full-time and part-time employees, as well as regular casuals who have at least 12 months of continuous service with an employer, can take up to 52 weeks of unpaid parental leave following the birth or adoption of a chid. There is no statutory right to paid maternity or paternity leave. Employees can negotiate paid maternity or paternity leave with their employer. In 2004, 41 per cent of women and 31 per cent of men had access to paid maternity or paternity leave in their job (ABS, 2004).

While these are key supports to families, it is important to note that families may have access to several other forms of support, such as access to free education for school children, bulk-billing of basic medical services, and concession cards that enable low-income families to receive various goods and services at a reduced cost (e.g., pharmaceuticals, public transport).

The purpose of this benefit package can be better understood when it is compared to the amount of assistance in other countries. This can be achieved by expressing the amount of a country's public spending on families as a percentage of its GDP. Figure 6 provides such data for 2003 in a number of OECD countries, including Australia. Australia's family spending was about 3.3 per cent of GDP, which is towards the higher end among OECD countries.

Figure 6: Family spending in cash, services and tax measures, percentage of GDP, 2003

Notes: Public support accounted here only concerns public support that is exclusively for families (e.g., child payments and allowances, parental leave benefits and child care support). Spending recorded in other social policy areas, such as health and housing support, also assists families, but not exclusively, and is not included here. Columns that are in black indicate a TFR of 1.5 or less in 2004.
Sources: OECD (2007) and Population Reference Bureau (2006)

Figure 7 shows that public expenditure on families in Australia has increased considerably since 1980. While the latest figures available are for 2003, the level of family payments has increased quite substantially in Australia since that time.

Figure 7: Family spending in cash, services and tax measures, percentage of GDP, 1980-2003

Source: OECD (2007)

There has been a great deal of discussion as to whether the recent up-turn in fertility rates in Australia is a direct result of government initiatives in the area of assistance to families (especially the Baby Bonus). As discussed in this paper, the level of financial support that the Australian Government provides to families with children has increased substantially, a factor that appears to be positively related to fertility rates. While the research evidence about this question is limited, McDonald (2006a) reported that early data on the introduction of the Baby Bonus indicate that "in the first full quarter in which births could have been affected by the new payment (June Quarter 2005), there was an increase of 10 per cent in the number of births compared to the same quarter in the preceding year" (pp. 224-225). This is consistent with the findings of studies from a range of countries into the impact of this type of maternity payment on fertility rates (e.g., Adkins, 2003; Gauthier & Hatzius, 1997; Milligan, 2005). Although there is some evidence that the maternity payments had an impact upon fertility, it is difficult to determine whether this was caused by changes in the financial incentives to have children or by the publicity around the payment that emphasised more broadly the importance of having children to the parents and society.¹²

6. Views about having children

The Australian Institute of Family Studies has undertaken considerable research into the decisions individuals make about having children - the "micro-level" dynamics of their everyday choices. The broad aim of this research program is to enhance understanding of the reasons underlying the fertility decisions of men and women in their key childbearing years, both as individuals and couples. Selected findings from this research that relate to the relevance of some of the above-mentioned explanations are outlined below.

The Institute's research, along with that undertaken elsewhere in Australia (Evans & Kelley. 1999; McDonald, 2002), clearly suggests that most people want to have children. In this section, the following issues are examined: the desire for children; current, ideal and expected family sizes; perceived stability of preferences; and issues men and women consider to be important when thinking about having or not having children.

6.1 The desire for children

There is little evidence from the Institute's research that individualism is an important explanation for Australia's below-replacement-level TFR. According to the Fertility Decision Making Project (FDMP), a national survey of 3,201 men and women aged 20-39 years undertaken in 2004, two-child families were the most popular, followed by three-child families (Weston, Qu, Parker, & Alexander, 2004).¹³ A family of four or more children appeared to be more popular than no children or only one child (taken separately). In fact, each of these latter two alternatives (taken separately) was considered to be ideal by less than 10 per cent of men and women in four age groups (Table 2). The popularity of an ideal of having two children, followed by three, then four or more, applied to men and women in their twenties and thirties regardless of relationship status and educational attainment.

Source: Weston et al. (2004)

These proportions were based on all men and women in each age group, including those who were already parents. Two-child families also represented the most common ideal family size of childless respondents in the four age groups (applying to 51 to 63 per cent of men and 44 to 56 per cent of women). While three-child families were the second most common preference for childless respondents under 35 years old (15 to 24 per cent of men and 18 to 29 per cent of women), remaining childless was the second most commonly mentioned ideal for those in their late thirties (21 per cent of men and 24 per cent of women).

In fact, among the childless, the ideal of having no children or one child became progressively more popular with advancing age, while having three or more children became less popular. For example, 24 per cent of childless women in their late thirties ideally wanted to remain childless, compared with only 5 per cent of childless women in their early twenties.

While to some extent these results reflect the fact that those approaching their forties would include an increased concentration of people who never wanted children, there was also evidence from the FDMP and in previous research (Weston & Qu 2001), that limited opportunities for having a child led some childless people to adjust to their circumstances by modifying their fertility preferences to be consistent with reality. For example, some childless respondents' reasons for their ideal of being childless included being single or having a partner who already had children, having postponed having a child and now feeling too old to cope with raising children, or fecundity problems experienced by themselves or their partner. In other words, preferences tended to be constrained by circumstance.¹⁴

6.2 Current, ideal and expected family sizes

Figures 8 and 9, which are based on the FDMP, show the averages (means) derived for ideal, expected and actual family size for men and women respectively in their early and late twenties and thirties. For each gender taken separately, the average ideal family sizes were similar across the four age groups (men: 2.3-2.4 children; women: 2.5-2.6 children), as were the average expected family sizes (men: 1.7-1.8 children; women: 2.1 children), with ideal family size being higher than expected family size. Women's average ideal and expected family sizes were slightly higher than those for men, although the ideal family size of men and women in their early twenties was very similar.

It is noteworthy that the average number of children that men and women, in their late thirties actually had, is below the average expected family size (men: 1.3 versus 1.8; women: 1.8 versus 2.1). Given that the men tended to be two or three years older than their partner, it is not surprising that the gap between the actual and expected number of children was greater for men than women. The gap would narrow further as some men and women in their late thirties will have a child or more children. Nevertheless, even in their late thirties, neither the women nor the men were expecting to have the number of children they said they ideally wanted.

Figure 8: Men: current, expected and preferred number of children

Source: Weston et al. (2004)

Figure 9: Women: current, expected and preferred number of children

Source: Weston et al. (2004)

Men and women in their thirties, regardless of whether they were married, single or cohabiting, had higher average ideal family sizes than they expected to actually have¹⁵. The same applied to patterns of averages for those in the three educational status groups. Of each of these different groups, the greatest discrepancy between average ideals and expectations occurred for men and women who were single (men: 2.3 versus 1.4; women: 2.4 versus 1.7), while the smallest discrepancy occurred for men and women who were married (men: 2.5 versus 2.3; women: 2.7 versus 2.4).

It is important to note, however, that between 60 and 65 per cent of men and women in the four age groups expected to have (or already had) the number of children they considered ideal - although, as noted above, there was evidence that ideals tended to be limited by perceived opportunities. Just over 30 per cent of men in their twenties and 32 to 41 per cent of men in their thirties believed that they would have fewer children than they considered ideal - a situation that applied to 29 to 37 per cent of women in their twenties and 30 to 34 per cent of women in their thirties. No more than 10 per cent in any group expected to have more children than they considered ideal.

The proportion of respondents expecting to achieve their ideal was strongly related to relationship status for men and women in their twenties and thirties (taken separately). Those most likely to expect to achieve their ideal family size were married men and women in their twenties and thirties, and cohabiting women in their twenties (68 to 70 per cent), while those least likely to expect this were single men and women in their thirties. Indeed, a slightly higher proportion of these older single men felt that they would "underachieve" rather than achieve their preference (50 per cent versus 45 per cent), while the opposite applied for women in their thirties (45 per cent versus 52 per cent).

Whatever the family size ideals held by men and women, most respondents in each educational status group indicated that they expected to achieve (or had achieved) their ideal family size. Nevertheless, men in their thirties who neither had, nor were pursuing, post-school qualifications, were less likely than other men to expect to achieve their ideal family size (41 per cent versus 32-34 per cent). Men in their thirties in the lowest educational status group were also less likely than other men of the same age to be partnered, although single men in the lowest educational status group were more likely to have children than other single men. These trends are generally consistent with the arguments by Birrell, Rapson, and Hourigan (2004) that men with poor financial prospects are less likely than other men to be able to have and raise children in a secure relationship.

6.3 Perceived stability of preferences

In the FDMP, those who were 23 years or older were asked to recall how many children they wanted when they were 20 years old. While most respondents indicated no change in preferences, across all age groups, those who reported a change in views were more likely to indicate a conversion from not wanting a child to wanting a child (a "positive conversion") rather than the reverse (a "negative conversion"). Positive conversions were indicated by 22 to 24 per cent of all men and by 15 to 18 per cent of all women. Negative conversions, on the other hand, were indicated by no more than 5 per cent of respondents in any group.

Among the majority who wanted children at age 20, family size preferences were more likely to be revised downwards than upwards - except for men in their twenties (Figure 10 and Figure 11). An upward revision was reported by 13-17 per cent of men and women in each age group, while the tendency to revise family size preferences downwards appeared to increase with increasing age. For instance, a revision downwards was indicated by 20 per cent of men in their late twenties and by 32 per cent of men in their late thirties. For women, the proportions were 21 per cent and 39 per cent respectively. Nevertheless, with the exception of women in their late thirties, more than half the respondents indicated no change in desired family size.

Figure 10: Men at age 20 who wanted to have a child: number of children wanted now (2004) and then (at age 20) by current age

Source: Qu and Weston (2004)

Figure 11: Women at age 20 who wanted to have a child: number of children wanted now (2004) and then (at age 20) by current age

Source: Qu and Weston (2004)

Interestingly, the reasons for respondents revising their family size aspirations downward and upward often related to the same domains. While financial and work-related reasons were often seen as constraints that led to a downward revision, a few respondents who wanted more children than when they were 20 years old mentioned their improved financial resources or job situation. Perceptions of financial resources were sometimes couched in terms of the desire to give children a good start in life, including sending them to private schools. For some, personal values and beliefs about how to maximise children's life chances led to perceptions of financial constraints.

Partnership issues, along with age, health and fecundity, emerged as reasons for changing family size aspirations. While lack of a partner and relationship breakdown led to revisions downwards, finding a partner and feeling secure in this relationship was often mentioned as a reason for upward revision. And while some of those who were already approaching their forties saw their age and fecundity problems as reasons for downward revisions, upward revision of family size was sometimes explained in terms of having "grown up" and changed priorities in life accordingly.

The pressures of parenting sometimes appeared to trigger downward revision, while for some, the joys experienced in parenting led them to consider having more children. Clearly, improving access to strategies that help those who are finding the task of parenting a stressful one is important. There also needs to be a change in the way in which parenting is perceived by many of those who are yet to have children. As noted above, parenting has within some sections of society, been devalued, and the FDMP suggests that some people are unaware of the positive aspects of having children until (and unless) they become parents.

Others who had revised their views downward explained this change of heart in terms of the problematic state of the world for raising children or over-population. On the other hand, upward revisions were sometimes explained in terms of a desire to have both a boy and a girl or to give a child a sibling.

6.4 Important considerations when thinking about having children

Respondents in the FDMP were also asked to indicate, on a scale from 0 to 10, how important each factor was when considering whether or not to have a child. Table 3 ranks the 28 items according to the proportions of men and women who appeared to see the item as very important (here defined as ratings of 8 to 10).

Source: Weston et al. (2004)

The capacity to support a child financially and the ability of each partner to be a good parent were issues that were most commonly considered to be of the highest importance. The male partner's job security was also a key issue for most men and women. Other matters rated as highly important by close to 50 per cent or more of men and women were: having someone to love, the female partner's age and the uncertainty about a relationship's future.

By contrast, only one-quarter of men and women emphasised concerns that may reflect emphasis on individualism, such as having time for leisure and social activities, and the ability to make major purchases.

Women were more likely than men to attach strong importance to the male partner having the time and energy to put into his career, and to the stress and worry of raising a first or additional child. The latter finding is not surprising given that the female partner usually assumes most of the responsibility for the everyday care of the children (Baxter, 2002; Bittman, 2004; Morehead, 2001).

In conclusion, these results clearly suggest that most Australians in their primary childbearing years either have or want children, and while most respondents expected to achieve their preferred family size, for some this had come about by a downward revision of aspirations to better fit personal circumstances. Furthermore, sizeable numbers did not expect to achieve their aspirations.

An important finding is that for both men and women the ideal family size was higher than that required for population replacement, but also higher than the family size that respondents expected. The proportion of young childless respondents who expected to remain childless was clearly lower than the proportion of young women who are projected by the ABS (2002) to be permanently childless (25 per cent).

A core set of issues appeared to shape aspirations and expectations, including achievement of an adequate income stream and ongoing secure employment; the perceived difficulties in having the time for both work and family life; and, for older respondents, age and fecundity problems. Importantly, this report clearly suggests that people are very concerned about their capacity to be a good parent and to provide the emotional security for their children that comes from a secure relationship.

7. A closer look at the importance of partnerships

Partnership formation and relationship stability is a common theme running through the above analysis and it is clear that finding a partner early in life and being happy in this relationship were important to respondents in the FDMP. Relationship breakdown, on the other hand, appeared to be linked with a downward revision of fertility preferences. The following section considers, in the first instance, partnership trends in Australia. It then outlines research that highlights the importance of these trends as factors influencing Australia's current fertility level.

Table 4 shows that partnership rates (covering both partners who are married to each other and those who are cohabiting) have fallen across all age groups (Birrell et al., 2004; Weston, de Vaus, & Qu, 2003).

Source: Birrell et al. (2004)

The decline in overall partnership rates arises from delays in partnership formation, the increased instability of relationships, and the increased proportions of people who remain continuously single.

Nevertheless, research suggests that most young adults appear to want to find a suitable partner and get married, but they appear to encounter many difficulties in achieving their ambition. These difficulties include: the setting of very high "standards" of suitability (i.e., it was difficult to find someone who could meet their standards); associated caution linked with the instability of relationships; already having young children whose needs were paramount; and lack of time to find a partner, owing to study or work pressures (Qu & Soriano, 2004).

7.1 Changing partnership formation pathways

Changing trends in partnership formation pathways have also contributed to delays in having children. While the proportion of children born outside marriage has increased, most couples wait until they marry before they have children. Around two-thirds of all births occur within marriage, and most of the remainder occur to couples who are cohabiting. However, first marriage rates have declined for all age groups, as shown in Figure 12 for women (the rates for men are very similar).

Figure 12: Age-specific first marriage rate, women

Source: ABS (2001a, 2003)

7.2 Instability of relationships

The instability of relationships is clearly an important factor affecting opportunities for individuals to have the number of children they hoped to have. In Australia, divorce rates were lower in the late 1980s (between 10.6 and 10.9 per 1000 married men and women), but then climbed fairly steadily from 10.9 in 1990 to 12.9 in 1996. Between 1997 and 2001 they have fluctuated between 12.0 and 13.1 (the rates occurring for 2000 and 2001). However, given the increase in cohabitation, divorce rates are no longer an accurate proxy for relationship instability.

Most first unions now start with cohabitation rather than marriage, with around 80 per cent of recent (first-union) cohabitations ending in either marriage or separation within five years of the start of the relationship. Furthermore, the proportion of such unions that end in separation within five years of the start of the relationship has increased. For example, of those who began cohabiting in the early 1970s, 63 per cent had married and 25 per cent had separated within the first five years of the relationship. Of those who began cohabiting in the early 1990s, 43 per cent had married and 38 per cent had separated within five years (Weston, de Vaus et al., 2003).

In other words, cohabitation was predominantly a prelude to marriage for women born in the early 1950s and appears to have become progressively less so for younger generations. Possibly, young couples today may be more likely to cohabit at an earlier stage in their relationship (when "going steady", rather than when considering marriage). Alternatively, young couples may now embark on cohabitation as a "trial marriage", but hold higher expectations about having their needs fulfilled in the relationship and are more prepared to separate if their needs are not met.

7.3 Links between relationship stability and fertility intentions

The FDMP was a large-scale cross-sectional study that relied on respondents' recollections to assess the dynamics of fertility aspirations. Nevertheless, the importance of partnership circumstances for views on having children was also apparent in a 10-year national follow-up study. The relevant analysis based on this latter study focused on fertility intentions and childbearing outcomes of 783 respondents (425 men and 358 women), all of whom were childless and had been between 18-34 years old in Wave 1 (Qu, Weston, & Kilmartin, 2000). Most respondents intended having children in Wave 1 and, regardless of their relationship status, most of them held these intentions in Wave 2, either having had children or still intending to do so. However, changes of intentions were often linked with personal relationships. Those who had separated from their partners were the most likely to change their minds and decide against having children, followed by those who were continuously single (particularly continuously single people aged 35 years or more). Nevertheless, this change of mind for the continuously single appeared to become more entrenched with age. Such trends may well reflect respondents' adjustment to the fact that time was running out. In this case, there appears to be a blurring of "voluntary" and "involuntary" childlessness.

Qu et al. (2000) also found that intentions to remain childless were less stable than intentions to have children, and were more likely to change for those who were single in 1981 and who subsequently partnered over the next ten years than for respondents who were partnered and remained with the same partner. The extent to which such changes arose from enhanced opportunities to have children or home pressures from the new partner is unclear, although as noted above, some respondents in the FDMP remarked that it was not until they met their partner that they first felt the urge to have children.

Another consequence of partnership instability is that, in new relationships, at least one of the partners may have already had children. The FDMP and other research (Weston & Qu, 2001) suggests that this circumstance leads some people to expect fewer children than they ideally want (including remaining childless), and others to revise their aspirations downward (and sometimes preferring childlessness).

While there is evidence at the individual level that relationship breakdown can have a negative impact upon fertility, the evidence at the macro level is not so clear. Firstly, Castle (2002), using data from 1998 for a number of OECD countries, found a positive correlation between having a higher divorce rate and the fertility rate.¹⁶ Castle cited the work of Monnier and de Guibert-Lantoire (1996), which found that countries with higher divorce rates and higher rates of cohabitation tend to have higher fertility rates. As Castle (2002) noted, this apparently counter-intuitive pattern may be a reflection of the fact that countries with higher divorce rates also tend to have higher rates of female labour force participation and gender-equity, factors, as argued above, which are thought to be positively associated with fertility rates.

Secondly, the fertility rate in Australia has been relatively stable over the last 25 years, while the rate of relationship breakdown has increased as the proportion of relationships that are cohabiting increases. Of course, the lack of a clear relationship between the increasing instability of relationships and fertility rates at a macro level may be because other changes that tend to increase fertility rates are happening over the same period. An example of such a change may be the increasing ability for women to combine childbearing with participation in the labour market.

The research described above also highlights the obvious importance of partnership formation for having children. Nevertheless, partnership formation represents a private realm into which Western governments seem reluctant to enter, although Singapore has launched several strategies to help people find a "life-long" partner. While having a partner is almost always a pre-requisite for childbearing, relationship quality is also one of the important factors in fertility decision-making. On the whole, this private area is one in which governments have been willing to help - by introducing "prevention" and "early intervention" measures directed towards promoting partnership quality and stability. Such measures, if successful, may lead some couples who want a child or who want to have more children to achieve their aims. There is evidence that participation in premarital education is associated with higher relationship quality and lower likelihood of divorce. (Stanley, Amato, Johnson, & Markman, 2006). However, the extent to which such positive results are a function of other unmeasured characteristics is unclear,¹⁷ and only a minority of couples use such services (see Halford & Simons, 2005).

7.5 Influence of a partner's views

Once in such a relationship, the role of a partner in fertility decision-making is clearly important. Yet, as Thomson (1997) pointed out, most research focuses on individuals rather than couples. In the FDMP, the importance of a partner's views or reproductive history featured in some of the explanations for not wanting or expecting children, for revising family size preferences upward or downward, or for expecting more or fewer children than preferred (Weston et al., 2004).

Furthermore, there was some evidence both in the FDMP and in the analysis of another large-scale dataset (the Household, Income and Labour Dynamics in Australia (HILDA) survey) to suggest that, among couples, the female partner's aspirations about having a child are more influential in shaping the expectations of the male partner than vice versa (Qu, Weston, & Parker, 2003).¹⁸

Finally, analysis of HILDA suggests that, among cohabiting couples, where couples disagree about wanting a child, the chance of separation increases if it is the male partner who wants a child, while the chance of marriage increases if it is the female partner who wants a child (Weston, Qu, & de Vaus, 2005).

Clearly, the research suggests that the inability to find a suitable partner or the experience of a relationship breakdown can prevent individuals from achieving their ideal number of children. Such practical circumstances can blur the distinction between voluntary and involuntary childlessness.

8. Conclusion

In the first instance, this paper outlined fertility trends, their implications for Australia, and key government policies directed towards supporting Australian families. Secondly, the results from a major national survey that explored contributing factors influencing fertility-making decisions were discussed.

From the research and evidence presented, it suggests that macro-level factors associated with higher fertility rates across countries include a strong economy and relatively low unemployment rates, a higher level of wage replacement during maternity leave, higher levels of government payments to families that reduce the costs of children, a higher rate of female employment (particularly maternal employment), higher use of non-parental child care, and a higher proportion of female employment being part-time.

An important message to be taken from the research is that a number of OECD countries with very different characteristics, economic and social policies have fertility rates that are around 1.7 to 1.8. This demonstrates that there is no unique set of specific policies associated with the achievement of such fertility levels in OECD countries.

Generally, however, it is fair to say that policies that lower the direct and indirect costs of raising children to families and allow women to combine paid employment with childrearing are likely to boost fertility rates. Policy changes in Australia over the last 25 years have certainly reduced the direct costs of childrearing through expanding the family payment system to include families with relatively high incomes. The recently introduced maternity payment has also provided significant financial support to families following the birth of a child.

In order to enhance fertility rates, it is important that families have support as the child grows up, not just when the child is a pre-schooler. Potential parents need to feel confident that any children they may have are in with a good chance of experiencing a secure lifestyle throughout their childhood and into adulthood (where "security" covers both material and psycho-social concerns). Parents also need to feel confident that they can manage to raise their children while also enjoying opportunities for personal development, beyond the world of family life.

Although Australia's fertility rate has been below the level of replacement for about three decades, most people of childbearing age want to have children and few consider no children or one child their ideal number of children. Close to two-thirds expect to achieve their ideal family size. On the other hand, a substantial minority of men and women do not expect to have their ideal number of children. Some people made downward revisions of their preferred family size over time and such revision was related to: advancing age, lack of a partner and relationship breakdown. Reasons for upward revisions included: "growing up", finding a partner and feeling secure in this relationship.

A fundamental challenge for policy makers, would seem to be that much of this terrain is in the realm of the private, and governments are naturally reluctant to intrude into people's relationship issues - even where there may be serious long-lasting consequences when a population is unable to replace itself. Nevertheless, governments have been willing to support programs that aim to enhance relationships, such as relationship counselling programs. There is evidence that premarital education is associated with a lower risk of marriage breakdown, but only a minority of couples participate in such programs, and those who do may possess characteristics that favour strong relationships.

Despite Australia's economic prosperity, people remain concerned about their capacity to create and maintain a family environment in which children can be nurtured and supported financially and emotionally. Such concerns, real or perceived, reflect macro-level trends in economic and employment security and in relationship formation pathways and their stability, as well as micro-level concerns about people's personal capacity to be good parents. Strategies directed towards helping people achieve their childbearing aspirations need to tackle their sense of security in each of these three domains. The data suggest that these strategies should target forces both at the macro- and micro-levels - for neither level acts in isolation.

In conclusion, governments need to use a combination of approaches that is based on the recognition that a low fertility rate is not due to a "lack of wanting children". If Australia is to boost its fertility rate - or at least maintain the current level - the message that raising children has an intrinsic richness and is an enjoyable part of life needs to be conveyed widely. To be effective, however, such a message must reflect reality. Couples need to have personal resources such as a secure income stream, a loving and stable relationship, and the skills and confidence to be parents. They also require access to community resources, including family-friendly workplaces and the confidence that they have a strong, continuing commitment from the community. It is of the utmost importance that parents do not feel alone in raising the next generation of citizens.

References

• Access Economics. (2001). Population ageing and the economy. Canberra: Commonwealth Department of Health and Aged Care.
• Adkins, D. (2003). The role of institutional context in European regional fertility patterns. Paper presented at the 2003 Annual Meeting of the Population Association of America, Boston.
• Allan, G., Hawker, S., & Crow, G. (2001). Family diversity and family change in Britain and Western Europe. Journal of Family Issues, 22(7), 819-836.
• Andersson, G. (2005). A study on policies and practices in selected countries that encourage childbirth: The case of Sweden (Working Paper WP 2005-005). Rostock: Max Planck Institute for Demographic Research.
• Australian Bureau of Statistics. (1997). Australian demographic trends 1997 (Cat. No. 3102.0). Canberra: Author.
• Australian Bureau of Statistics. (2001a). Marriages and divorces Australia 2000 (Cat. No. 1310.00). Canberra: Author.
• Australian Bureau of Statistics. (2001b). Year book Australia 2001 (Cat. No. 1301.0). Canberra: Author.
• Australian Bureau of Statistics. (2002). Australian social trends 2002 (Cat. No. 4102.0). Canberra: Author.
• Australian Bureau of Statistics. (2003). Marriages and divorces Australia 2002 (Cat. No. 1310.00). Canberra: Author.
• Australian Bureau of Statistics. (2004). Employee earnings, benefits and trade union membership (Cat. No. 6310.0). Canberra: Author.
• Australian Bureau of Statistics. (2005). Population projections Australia 2004-2101 (Cat. No. 3222.0). Canberra: Author.
• Australian Bureau of Statistics. (2006a). Australian historical population statistics 2006 (Cat. No. 3105.0.65.001). Canberra: Author.
• Australian Bureau of Statistics. (2006b). Year book Australia 2006 (Cat. No. 1301.0). Canberra: Author.
• Australian Bureau of Statistics. (various years). Births Australia (Cat. No. 3301.0). Canberra: Author.
• Baxter, J. (2002). Patterns of change and stability in the gender division of household labour in Australia 1986-1997. Journal of Sociology, 38(4), 399-424.
• Birrell, B., Rapson, V., & Hourigan, C. (2004). Men + women apart: Partnering in Australia. Melbourne: Australian Family Association and Centre for Population and Urban Research.
• Bittman, M. (2002). The effects of time pressures on mothers' health. Paper presented at the Seminar Series, Australian Institute of Family Studies, 19 September, Melbourne
• Bittman, M. (2004). Parenting and employment: What time use surveys show. In N. Folbre & M. Bittman (Eds.), Family time: The social organisation of care (pp. 69-89). London: Routledge.
• Castle, F. (2002). Three facts about fertility: Cross-national lessons for the current debate. Family Matters, 63, 22-27.
• Centrelink. (2007). A guide to Australian government payments. Canberra: Author.
• Coleman, D. A. (1999). Reproduction and survival in an unknown world: What drives today's industrial population, and to what future? (NIDI Hofstee Lecture Series No.15). The Hague: Netherlands Interdisciplinary Demographic Institute.
• Crittenden, A. (2001). The price of motherhood: Why the most important job in the world is still the least undervalued. New York: Henry Holt.
• d'Addio, A., & d'Ercole, M. (2005). Trends and determinants of fertility rates in OECD countries: The role of policies (OECD Social, Employment and Migration Working Papers No. 27). Paris: OECD.
• Encel, S. (2002). Voodoo demographics: The "threat" of an ageing population. Sydney: Social Policy Research Centre, University of New South Wales.
• Evans, M., & Kelley, J. (1999). Small families or large? Australia in international perspective. Australian Social Monitor, 2(1), 13-19.
• Gauthier, A., & Hatzius, J. (1997). Family benefits and fertility: An econometric analysis. Population Studies, 51(3), 295-306.
• Grant, J., Hoorens, S., Sivadasan, S., van het Loo, M., DaVanzo, J., Hale, L. et al. (2004). Low fertility and population ageing: Causes, consequences, and policy options. Santa Monica: Rand Corporation.
• Halford, W. K., & Simons, M. (2005). Couple relationship education in Australia. Family Process, 44(2), 147-159.
• Hugo, G. (2001). A century of population change in Australia. In Year book Australia 2001 (Cat. No. 1301.0). Canberra: Australian Bureau of Statistics.
• Jackson, N. (2007). Population ageing in a nutshell. People and Place, 15(2), 12-21.
• Kagitcibasi, C. (1997). The value of children: A key to gender issues. International Pediatric Association Journal (INCH), 9(1). Retrieved 11 January 2008, from http://www.healthnet.org.np/pediatric/kag.html
• Kippen, R. (2004). Declines in first and second birth rates and their effect on levels of fertility. People and Place, 12(1), 27-36.
• Kogel, T. (2006). An explanation of the positive correlation between fertility and female employment across Western European countries. Loughborough: Department of Economics, Loughborough University.
• Kohler, H., Billari, F. C., & Ortega, J. A. (2001). Towards a theory of lowest-low fertility (MPIDR Working Paper WP 2001-032). Rostock: Max Planck Institute for Demographic Research.
• Kohlmann, A. (2002). Fertility intentions in a cross-cultural view: The value of children reconsidered (Working Paper WP 2002-002). Rostock: Max Planck Institute for Demographic Research.
• Laroque, G., & Salaníe, B. (2005). Does fertility respond to financial incentives? (Discussion Paper 5007). London: Centre for Economic Policy Research.
• Laws, P. J. & Sullivan, E. A. (2005). Australia's mothers and babies 2003 (Perinatal Statistics Series No. 16). Sydney: Australian Institute of Health and Welfare, Perinatal Statistics Unit.
• Mackay, H. (1997). Generations: Baby boomers, their parents and their children. Sydney: Macmillan.
• Martin, J. (2003, February). The ultimate vote of confidence: Fertility rates and economic conditions in Australia, 1975-2002. Paper presented to the 8th Australian Institute of Family Studies Conference, "Steps forward for families: Research, practice and policy", Melbourne.
• McDonald, P. (2000a). Low fertility in Australia: Evidence, causes and policy responses. People and Place, 8(2), 6-20.
• McDonald, P. (2000b). The "toolbox" of public policies to impact on fertility: A global view. Paper presented at the seminar "Low fertility, families and public policies", organised by the European Observatory on Family Matters, Sevilla.
• McDonald, P. (2001, March). Theory pertaining to low fertility. Paper presented at International Union for the Scientific Study of Population, Working Group on Low Fertility, "International perspectives on low fertility: Trends, theories and policies", Tokyo.
• McDonald, P. (2002). Low fertility: Unifying the theory and the demography. Paper prepared for the meeting of the Population Association of America, "Future of fertility in low fertility countries", Atlanta.
• McDonald, P. (2006a). An assessment of policies that support having children from the perspectives of equity, efficiency and efficacy. In Vienna Yearbook of Population Research (pp. 213-234). Vienna: Austrian Academy of Sciences.
• McDonald, P. (2006b). Low fertility and the state: The efficacy of policy. Population and Development Review, 32(3), 485-510.
• McDonald, P., & Kippen, R. (1999). The impact of immigration on the ageing of Australia's population (DIMA Discussion Paper). Canberra: Department of Immigration and Multicultural Affairs.
• McDonald, P., & Kippen, R. (2000, March). The implications of below replacement fertility for labour supply and international migration, 2000-2050. Paper presented to the Annual Meeting of the Population Association of America, Los Angeles, California.
• McDonald, P., & Temple, J. (2006). Immigration and the supply of complex problem solvers in the Australian economy. Canberra: Australian Government Department of Immigration and Multicultural Affairs.
• Milligan, K. (2005). Subsidizing the stork: New evidence on tax incentives and fertility. Review of Economics and Statistics, 87(3), 539-555.
• Ministerial Taskforce on Child Support. (2005). In the best interests of children: Reforming the Child Support Scheme. Report of the Ministerial Taskforce on Child Support. Canberra: Author.
• Monnier, A. & de Guilbert-Lantoine, C. (1996). La conjuncture démographique: l'Europe et les pays développés d'outre-mer. Population, 16(4), 1005-1030.
• Morehead, A. (2001). Synchronizing time for work and family: Preliminary insights from qualitative research with mothers. Journal of Sociology, 37(4), 355-369.
• Organisation for Economic Co-operation and Development. (2007). Social expenditure database. Paris: Author. Retrieved 11 January 2008, from http://www.oecd.org/els/social/expenditure
• Phillips, K. (2002). Fertility rates, women in the workforce, and economic health. Family Matters, 63, 50-53.
• Population Reference Bureau. (2006). 2006 world population data sheet. Washington, DC: Author
• Productivity Commission (2005). Economic implications of an aging society: Productivity Commission Research Report. Melbourne: Author. Retrieved 11 January 2008, from http://www.pc.gov.au/__data/assets/pdf_file/0020/69401/ageing.pdf
• Qu, L., & Soriano, G. (2004). Forming couple relationships: Adolescents' aspirations and young adults' actualities. Family Matters, 68, 43-49.
• Qu, L., & Weston, R. (2004). Family size: Men's and women's aspirations over the years. Family Matters, 69, 18-23.
• Qu, L., Weston, R., & Kilmartin, C. (2000). Children? No children? The effect of changing personal circumstances on decisions about having children. Family Matters, 57, 14-19.
• Qu, L., Weston, R., & Parker, R. (2003, July). Two to tango? The importance of partnerships and partners' views in shaping fertility aspirations and expectations. Paper presented at the Eighth Australian Social Policy Conference, Social Policy Research Centre, University of New South Wales, Sydney.
• Rowland, D. (2003). An ageing population: Emergence of a new stage of life? In S. Khoo & P. McDonald (Eds.), The transformation of Australia's population: 1970-2030. Sydney: University of New South Wales.
• Ruzicka, L., & Caldwell, J. (1982). Fertility (Country Monograph Series no. 9, Population of Australia, vol. 1). New York: Economic and Social Commission for Australia and the Pacific.
• Saunders, P. (2001). The changing social security policy context: Implications for the CDEP program, the Indigenous welfare economy and the CDEP scheme (CAEPR Research Monograph No. 20). Canberra: Centre for Aboriginal Economic Policy Research, Australian National University.
• Sleebos, J. (2003). Low fertility in OECD countries: Facts and policy responses (OECD Social, Employment and Migration Working Papers No. 15). Paris: OECD.
• Stanley, S., Amato, P., Johnson, C., & Markman, H. (2006). Premarital education, marital quality, and marital stability: Findings from a large, random household survey. Journal of Family Psychology, 20(1), 117-126.
• Stanton, D. (2002). Director's report: The "fertility crisis". Family Matters, 63, 2-3.
• Thomson, E. (1997). Couple childbearing desires, intentions, and births. Demography, 34(3), 343-354.
• Treasury. (2002). Intergenerational report 2003-03 (2002-03 Budget Paper No. 5). Canberra: Author.
• Weston, R., de Vaus, D., & Qu, L. (2003, November). Partnership formation and stability. Paper presented at the (USA) National Conference on Family Relations, Vancouver.
• Weston, R., & Qu, L. (2001). Men's and women's reasons for not having children. Family Matters, 58, 10-15.
• Weston, R., Qu, L., & de Vaus, D. (2005, September). Pathways from cohabitation. Paper presented at the Household, Income and Labour Dynamics in Australia (HILDA) Survey Conference, Melbourne.
• Weston, R., Qu, L., Parker, R., & Alexander, M. (2004). It's not for lack of wanting kids: A report on the Fertility Decision Making Project (Research Report No. 11). Melbourne: Australian Institute of Family Studies.
• Weston, R., Qu, L., & Soriano, G. (2003). Australia's ageing yet diverse population. Family Matters, 66, 6-13.
• Whiteford, P., & Angenent, G. (2002). The Australian system of social protection: An overview (Occasional Paper No. 6, 2nd ed.). Canberra: Department of Family and Community Services.

Lists of tables and figures

List of tables

• Table 1 Total fertility rate and projected population changes, selected countries, 2006-2050
• Table 2 Ideal number of children by age, Australia
• Table 3 Proportion and ranking of factors considered important in having children, by gender
• Table 4 Proportion of men and women who were living with a partner, 1986, 1996 and 2001

List of figures

• Figure 1 Total fertility rate, Australia, 1921-2005
• Figure 2 Total fertility rate, Australia, 2000-2005
• Figure 3 Age-specific birth rates, Australia, 1921-2005
• Figure 4 Age and sex structure of Australia's population, observed (1911, 1961, 2004) and projected (2051)
• Figure 5 Proportion of population aged under 15 and aged 65 or over (1901-2004), and projected proportion (to 2100)
• Figure 6 Family spending in cash, services and tax measures, percentage of GDP, 2003
• Figure 7 Family spending in cash, services and tax measures, percentage of GDP, 1980-2003
• Figure 8 Men: current, expected and preferred number of children
• Figure 9 Women: current, expected and preferred number of children
• figure 10 Men at age 20 who wanted to have a child: number of children wanted now (2004) and then (at age 20) by current age
• Figure 11 Women at age 20 who wanted to have a child: number of children wanted now (2004) and then (at age 20) by current age
• Figure 12 Age-specific first marriage rate, women