Fertility declinein AustraliaFamily Matters No.63 Spring/Summer 2002 Australian Institute of Family Studies14gradual fertility decline from 1.91in 1990 to 1.73 in 2001. Overall,the fertility rate has halved from3.55 to 1.73 in a period of just 40 years.Changes in fertility do not occur in isolationfrom other social, cultural and economic changes.Although it is often difficult to establish cause andeffect patterns, the three modern phases of Aus-tralian fertility decline were accompanied byparallel developments. The first period of sharpdecline began at the same time as the oral contra-ceptive pill became widely available. The secondand third periods occurred at the same time as thelabour force participation of married women andmothers increased, when both men and womenbegan to marry later, when cohabitation becamemore common, when school retention ratesimproved and more people went on to tertiaryeducation, and when the nature of the workforcechanged markedly, especially for younger people.The impact of fertility decline depends partly onthe rate of decline and the level to which fertilitydeclines. Sharp and rapid decline will have a differ-ent impact from gradual decline. While institutionsand the economy may adjust to a gradual decline innumbers, it is much more difficult for institutions toadjust to a rapid fertility decline.The other element of fertility decline is the levelto which fertility drops. Australia’s total fertilityrate (TFR) in 2001 was 1.73. The total fertility rateindicates the number of children a woman will haveif she experiences the current age-specific fertilityrates at each age of her reproductive life. Demogra-phers frequently focus on replacement levelfertility, or the estimated number of children awomen would need to have in her lifetime toreplace herself and her partner. They estimate thatin order to maintain a stable population size overlthough fertility declinehas been apparent forsome time, it has onlyrecently been taken seriously by policymakers. However, as the baby boomer gen-eration ages, the workforce shrinks and health careand income support costs increase, the implications ofa low birth rate are becoming more evident. Fertilitydecline also has economic and institutional implica-tions. Population growth has traditionally been a keysource of economic growth, and concern has beenwidely expressed that without steady populationgrowth, economic growth will stall. Fertility declinealso has substantial implications for age based socialinstitutions such as schools and universities wherefewer children will mean reduced demand for places.The purpose of this article is to sketch some of thedimensions of fertility decline in Australia. It willbriefly describe the extent to which fertility levelshave declined and then describe the groups in soci-ety where fertility rates are relatively low. Since thepaper by Castles in this issue of Family Matters dealswith some of the international dimensions of fertilitychange, this article focuses on Australian patterns.FertilitylevelsandtrendsAustralia has experienced almost uninterrupted fer-tility decline since 1961. This decline has occurredin three main phases (Figure 1). The first phasefrom 1961 to the mid-1970s was a period of sharpand dramatic decline when the fertility ratedropped from 3.55 in 1961 to 2.15 in 1975 – adecline of 1.4 children per woman in just 15 years.The second phase was a period of relative stabilityin the 1980s when the fertility rate fluctuatedbetween 1.84 and 1.92. The third and currentphase, which began in the early 1990s, is a period ofAAThedecliningbirthrateisamajorfactorinpopulationageingandalsohasimportantimplicationsforthewayfamiliesfunction,fortheeconomyandsocietymoregenerally,andforissuesofenvironmentalsustainability.A d e m o g r a p h i c c o n t e x tDAVID DE VAUSFertility declinein Australia
15Australian Institute of Family Studies Family Matters No.63 Spring/Summer 2002the longer term each woman needs to have, on aver-age, 2.1 children. Fertility levels below this rate willlead, in the longer term, to population decline1.Australia’s fertility rate fell below this replacementlevel in 1976 for the first time in the 20th centuryand has continued to decline thereafter.Australia’s below replacement fertility level ishardly unique among the developed economies (seeCastles elsewhere in this issue). Of all the OECDcountries, only Turkey and Mexico have abovereplacement fertility, and the Unitied States, with afertility level of 1.9 (ABS 2002a), has the next high-est fertility level. Indeed, with a fertility level of1.73, Australia has a relatively high fertility levelamong the OECD nations.NumberofchildrenTotal fertility rates provide a picture of average fertil-ity. However, averages do not allow us to answerimportant questions about the components of fertil-ity decline. How much of the decline is because morewomen are remaining childless? How much isbecause women who are still having children areavoiding having large families? Is the decline becausesingle-child families are becoming popular?Identifying the components of fertility decline canhelp focus on some of the factors than might liebehind the decline. The reasons for remaining child-less are likely to be very different from those that leada women to have three rather than five children.Accordingly, any action designed to have an impacton fertility levels needs to be targeted at the particu-lar components that contribute to fertility decline.ChildlessnessChildlessness may be voluntary or involuntary,although the line between these two routes to child-lessness can be rather blurred. It is estimated thatapproximately 7 per cent of couples of reproductiveage are infertile (ABS 2002b). The rate at whichcouples are infertile increases with age – especiallythat of the woman.Levels of childlessness are estimated in differentways. One way is to base estimates on womenwho have completed their childbearing. This isachieved by observing the levels of childlessnessamong women aged 45-49 years. However, theseestimates do not reflect the patterns of women cur-rently of reproductive age and thus they provide an0.00.51.01.52.02.53.03.54.0DepressionIncreasing labour forceparticipation of womenIncreasing access tooral contraceptionBaby BoomReplacement levelWorld War II200020051995199019851980197519701965196019551950194519401935193019251920Total fertility rates, Australia, 1920-2001Figure 1YearTotalfertilityrateSource: ABS Social trends, 1996 and ABS Births, 2001 Australia, ABS cat no.3301.0. Thetotal fertility rate represetns the number of children a woman would bear during her lifetime ifshe experienced current age-specific fertility rates at each age of her reproductive life.05101520251901–05 1911–15 1921–25 1931–35 1941–45 1952–573035Proportion of females who were childless at age45-49 years(a)Figure 2Year of birthPercentNote: (a) Based on data from Censuses of Population and Housing.Source: Rowland, D. T., 1998. The prevalance of childlessness in cohorts of older women,Australasian Journal on Ageing, vol. 17, no. 1, pp.18–23.
Family Matters No.63 Spring/Summer 2002 Australian Institute of Family Studies16Figure 2 reports the levels of childlessness amongwomen aged 45-49 who were born between 1901 to19572as reported by Rowland (1998). This figureshows high levels of childlessness (31 per cent)among women born at the beginning of the century,who experienced the Great Depression in theirearly thirties and who had completed their child-bearing by 1950-1955. These rates of childlessnessdeclined steadily and reached their lowest levelamong women born between 1930 and 1945 ofwhom less than 10 per cent remained childless.Rates of childlessness among women born since theWorld War II have gradually increased and, as indi-cated above, are projected to increase yet further. Ifthese projections are correct, the rates of childless-ness will return to levels close to those evidentamong women born a century earlier.LargerfamiliesThe propensity of women to have larger families isan important component of the fertility rate. Thedecreasing number of women having large familieshas played a major role in fertility decline. The fallingoff in the total fertility rate since 1960 has beenaccompanied by a reduction in the number of womenhaving five or more children. In 1960, almost 12 percent of children born were the fifth or subsequentchild born to the mother. Since then this percentagehas steadily fallen away so that by 1996 just 2.3 percent of births were fifth or higher order births.One reason why the Australian total fertilityrate is not as low as that in many parts of Europe isthat Australian women are still more inclined thanEuropean women to have more than two children(McDonald 1998). McDonald estimates that if theAustralian women who have more than two chil-dren reduced their fertility to just two (like theirtypical European counterparts) the total fertilityrate would drop to 1.4 (McDonald 1998).The importance of these relatively fertile womenfor the overall fertility rate is reflected in McDonald’s(1998) estimate that today’s younger women whohave more than two children will account for half thechildren who are born in the future. In 1996, 37 percent of women aged 40 had more than two children.This percentage is projected to decline to 27 per cent.Isthefertilitydeclinereallyjustafertilitydelay?Part of the reason for the fertility decline is thatwomen are starting families later. The sharp fertilitydecline from 1960-1975, followed by a period ofrelative stability in the 1980s and then gradualinadequate guide to future levels of childlessness. Toestimate future levels of childlessness demographersuse marriage and fertility patterns of women in theirreproductive years. These estimates vary somewhatdepending on the methods of projections.The first method of estimating levels of childless-ness indicates that in 1996 10.5 per cent of womenaged 45-49 were childless. The second method,which estimates likely patterns of future childless-ness among younger women, project much higherlevels of childlessness. The precise projections rangefrom at least 20 per cent of women remaining child-less if the year 2000 fertility patterns prevail (Merloand Rowland 2000) through to 28 per cent based onthe midwives 1996 data collection (AIHW 1998).The ABS projects that about a quarter of all womenstill in their reproductive years (age 15-45) willremain childless, and that about a third of women inVictoria and the ACT will remain childless (ABS2002). Bryson, Strazzari and Brown (1999) estimatefrom the Longitudinal Study of Women’s Health thatabout 20 per cent of women aged 18-23 who intendto have children will end up childless.Number of children ever born for womenaged 45-69Table 1Age Mean N45-49 2.3 620650-54 2.4 470955-59 2.7 373560-64 2.8 318765-69 2.7 3197Source: ABS 1996 Census 1 per cent Confidentialised Unit Record File.NB women in older age groups have been excluded as survivalfactors increasingly complicate the interpretation of figures for olderage groups.02468141210161920 1940 1960 1980 20001820Fifth or higher order births 1935-1996Figure 3YearPercentSource: .Fertilitybehaviourmustbeunderstoodwithinasocialandculturalcontext.
17Australian Institute of Family Studies Family Matters No.63 Spring/Summer 2002decline again in the 1990s partly reflects this pattern.Carmichael and McDonald (1999) note that theperiod of relative stability in the 1980s was becausethe women who had delayed having children in theirtwenties (and thus contributed to the fertilitydecline) had babies in their thirties. This delayed fer-tility masked the fact that there was continuingfertility decline among women in their twenties.Two sets of data bear directly on the question offertility decline versus fertility delay. The first set ofinformation is the average family size of womenwho have completed their families. Table 1 reportsthe average number of children born to womenaged 45 and over. It shows that the average numberof children to which women give birth is declining.Those women who had most recently completedtheir childbearing period (women aged 45-49) hadan average of 2.3 children3. Older women whotherefore completed their families in earlier yearshad a higher average number of children. Fertilityfor women aged 50-54 was 2.4 and 2.7 and 2.8 forthe older age groups.The difference in the number of children of theseyounger and older women represents true fertilitydecline that is not attributable to childbearingdelays4- in this case there is a fertility decline of 0.5of a child per woman in just 15 years (from 2.8 forthose aged 60-64 to 2.3 for those aged 45-49).The other set of evidence that relates to the issueof fertility decline versus fertility delay are the agerelated fertility rates. Figure 4 displays the age spe-cific fertility rates for three younger groups ofwomen – those aged 15-19, 20-24 and 25-29 years.In all three age groups the fertility rate has declinedsince 1960. The sharpest fertility declines wereamong those in their early twenties. In 1960, 220per 1,000 women aged 20-24 gave birth to a child inthat year. By 2000 this rate had fallen to a quarter ofthe 1960 level, with just 58 women per 1,000 in thisage group giving birth in 2001. The decline amongwomen in their late twenties was also sharp but notas dramatic. In 1960, 216 per 1,000 women in theirlate twenties gave birth. By 2001, this had halved to104 births per 1,000 women.The pattern of fertility decline for women aged30 or over is quite different from that for youngerwomen (Figure 5). From 1960 to 1975 the fertilityrate of these older women declined, but at amuch more subdued rate than among youngerwomen. After 1975 the fertility rate of these olderwomen reversed and began to increase – albeit at agentle rate. This pattern suggests that some of thefertility decline in the 1960-1975 period was dueto delaying births. The effect of this is evident in theincreased fertility of older women after 1975 andthe subsequent slowing rate of overall fertilitydecline.However, the fertility decline cannot simply beexplained by fertility delays. The increase in fertil-ity among women aged 30 and over is nowhere nearenough to compensate for the rate of decline amongyounger women. The fact that between 1960 and2001 the fertility rates have dropped in all agegroups attests to the fact that a substantial part ofthe change in fertility rates is due to fertility declinerather than just fertility delay.Whohasthebabies?Fertility behaviour must be understood within asocial and cultural context. Not all women want orhave children, and the number of children womenboth want and have varies somewhat depending onthe social context in which they live. Fertility changealso must be understood within the wider context ofsocial and economic changes such as increasing edu-cational participation of women, extended periods oftraining for both men and women, the changingnature and increased instability of relationships,greater female labour force participation and chang-ing individual and cultural values and aspirations.One part of understanding fertility patterns is toidentify which groups of women are having no chil-dren, and those who are having more than theAge specific fertility rates, 2001Figure 602040608010012040-4435-3930-3425-2920-2415-19Age groupSource: ABS (2002) Births, Australia, 2001.Fertilityrate02550751001751501252001960 1965 19751970 19851980 1990 1995 2000 200522525030–34 35–39 40–44Age specific fertility rates, women aged 30-44, 1960-2001Figure 5YearAgespecificfertilityrateSource: ABS (2002) Births, Australia, 2001 and ABS Historical Statistics Table 39.02550751001751501252001960 1965 19751970 19851980 1990 1995 2000 2005225250 25–2920–2415–19Age specific fertility rates, women aged 15-29, 1960-2001Figure 4YearAgespecificfertilityrateSource: ABS (2002) Births, Australia, 2001 and ABS Historical Statistics Table 39.
Family Matters No.63 Spring/Summer 2002 Australian Institute of Family Studies18age-related patterns represent a substantial changein the prevailing patterns in 1960 at the peak of thebaby boom (see Figures 4 and 5).An adequate understanding of fertility needs toconsider the factors that drive the fertility changes inthese different age groups. Although some factors willaccount for some of the fertility decline across all agegroups (for example, access to contraception andabortion), other factors will apply to just some agegroups. A fruitful way of understanding fertility declineis to focus on the age groups in which the decline hasbeen greatest, both numerically and proportionally.It is among younger women where parallelchanges are most likely to affect their fertility. Thesewomen are staying at school longer, going on toobtain post-school qualifications, delaying bothpartnering and marriage, and establishing them-selves in a career. Of course, it is not just thecircumstances of young women that affects theirfertility. Finding a suitable partner and father willplay an important role. Despite the fact that 31 percent of children are born to unmarried mothers(ABS 2002a), only 11.5 per cent were to singlewomen who were not in a relationship (Nassar andSullivan 2001). To the extent that young men arenot wanting to settle down or are struggling to estab-lish themselves in secure employment they arelikely to be reluctant to become fathers. The evi-dence certainly shows that fewer young men andwomen are partnering (Birrell and Rapson 1998). Ofthose who are partnering an increasing proportionare cohabiting. For a variety of reasons cohabitingcouples are less likely than married couples to havechildren and thus the rise in cohabitation will bothdelay marriage and delay and subdue fertility.MaritalstatusIt is hardly surprising that Table 2 shows that mar-ried women have more children than those who havenever married. Nor is it surprising that marriedwomen aged 45-49 have more children on averagethan similarly aged divorced or widowed women.However, it is instructive to note that women whohave never married have nevertheless had, on aver-age, 0.4 children. This translates to 22 per cent ofnever married women having had at least one child.The same table also indicates that women aged45-49 who are currently in de facto marriages havefewer children than those in registered marriages(1.9 compared with 2.5 children). This reflects thefact that 20 per cent of women aged 45-49 in defacto relationships have no children at all. Thelower fertility of women in this age bracket who livein de facto relationships may indicate that men andwomen in de facto relationships want fewer chil-dren. This may be because they are less keen onhaving children or because they are less confidentin the stability of the relationship, or because decid-ing not to formally marry and not to have childrenis a deliberate choice about the way in which suchcouples wish to live.SocialandeconomicdisadvantageFertility is lower among women who are in relativelyadvantaged social and economic positions. This pat-tern can be seen from a number of sets of evidence.normal number of children. While identifyingwhich types of women have few children and whichhave large families does not tell us what the causesof low or high fertility might be, it can neverthelesspoint us in the direction of factors that may encour-age or discourage fertility.This section focuses on fertility differencesaccording to the age, country of birth and indige-nous status of mothers, social and economicdisadvantage and where they live.AgeAge-specific fertility rates provide a useful snapshotof the ages at which women are most likely tohave children. Figure 6 shows that childbearing isconcentrated within a fairly narrow age band. In2001, fertility rates were highest among women intheir late twenties and early thirties. The year 2001Number of children ever born to women aged 45-49 bymarital status 1996Table 2No One 4 orMean child child more N% % %Formal marital statusMarried 2.5 5.8 9.2 15.1 4538Widowed 2.3 9.0 13.8 13.1 145Separated/divorced 2.3 8.6 14.1 14.1 1169Never married 0.4 78.0 10.2 3.1 354Social marital statusMarried 2.5 5.5 9.3 15.2 4386Defacto 1.9 19.7 15.2 12.7 244Not married 1.9 23.2 12.9 11.6 1391Source: ABS 1996 Census 1 per cent Confidentialised Unit Record File.Those recoded in 1 per cent sample file as having 4 or more children are treated as, onaverage, having had 4.5 children.Number of children ever born to women aged 45-49 bymothers education, 1996Table 3No One 4 orMean child child more N% % %No qualifications 2.4 8.8 9.4 16.2 3837Vocational 2.3 10.1 11.6 9.4 464Diploma 2.2 11.8 11.8 12.7 584Degree or higher 2.0 18.0 12.2 7.8 817Source: ABS 1996 Census 1 per cent Confidentialised Unit Record File.Those recoded in 1 per cent sample file as having 4 or more children are treated as, onaverage, having had 4.5 children.220.127.116.11.82.01.91st 2nd 4th3rd 6th5thDocile Most disadvantagedLeast disadvantaged7th 8th 9th 10th2.1Fertility by social and economic disadvantage of regionFigure 7TotalfertilityrateSource: ABS (2000) Births, Australia, 1999.
19Australian Institute of Family Studies Family Matters No.63 Spring/Summer 2002First, the higher a woman’s education the fewerchildren she has. Table 3 shows the distinctive fer-tility levels of women with a university degree,especially compared to those with no post-schoolqualification. Women aged 45-49 with a degree weretwice as likely than those with no qualification to bechildless (18 per cent compared with 8.8 per cent).Conversely, women with a degree were only half aslikely as those without a qualification to have fouror more children (7.8 per cent compared with 16.3per cent).This link between education and number of chil-dren is probably a two-way link. Having morechildren may reflect a poorer range of choices forthose women with lower education. It also probablyreflects the fact that having children at a youngerage makes it more difficult to complete or continuewith one’s education.Further evidence of a link between fertility andsocial and economic disadvantage comes from thehigher fertility levels in the more socially and eco-nomically disadvantaged areas of the country. Usingthe ABS (1998b) classification of Social and Eco-nomic Disadvantage (SEIFA index), areas can begrouped into ten deciles. These decile groups areranked from the 10 per cent of regions that are themost disadvantaged through to the 10 per cent thatare the least disadvantaged (Figure 7).This figure shows that the least disadvantagedareas have the lowest fertility rate (under 1.6). Thefertility rate steadily increases as the region becomesmore socially and economically disadvantaged. Thefertility in the second most disadvantaged regions(ninth decile group) is over 2.The higher fertility rate in the more disadvantagedregions should not be attributed only to the higherlevels of social disadvantage in these areas. Areashave different population profiles, age and ethnic mixand the like, and these factors, as well as levels ofsocial and economic disadvantage, may contribute tothe higher fertility levels in the poorer regions.EthnicityandindigenousstatusEthnic and racial background are also associatedwith family size (Table 4). Women aged 45-49 whowere born in New Zealand, Western Europe andAsia (except Vietnam) have the smallest familieswith an average of about 2.1 children. New Zealandborn women are especially likely to have no chil-dren at all (21 per cent).The highest fertility was among women born inthe Middle East and North Africa where the averagewoman had 2.7 children. Middle Eastern womenwere also the most likely to have large families – 28per cent had four or more children.Australian born women, together with those bornin the United Kingdom and Southern Europe had fer-tility levels between these extremes. Australian bornwomen aged 45-48 had an average of 2.3 children(higher than projected for women still in their repro-ductive years) and around 10 per cent were childless.Indigenous Australian women, on average, havelarger families than non-indigenous women (Table4). In 1996, indigenous women aged 45-49 had hadan average of 3.1 children5compared with the 2.3children borne by other women.Indigenous women had distinctive patterns of fer-tility. Just 6 per cent had no children (compared with10.5 per cent for non-indigenous women). However,the sharpest difference between indigenous andother women was the percentage that had four ormore children. While 14 per cent of non-indigenouswomen had four or more children, over a third (36per cent) of indigenous women had this many chil-dren. However, higher infant mortality rates amongNumber of children ever born to women aged 45-49 bymothers country of birth and indigenous status 1996Table 4One 4 orMean None child more N% % %Mother’s country of birthMiddle east/ 2.7 9.4 8.3 28.1 96North AfricaSouthern Europe 2.4 7.3 8.2 12.8 439Vietnam 2.4 13.0 16.7 25.9 54Australia 2.3 10.7 9.0 14.8 3959United Kingdom 2.2 9.7 11.7 12.6 657Other Asia 2.1 10.9 18.3 10.1 257Western Europe 2.1 8.1 20.5 11.8 161New Zealand 2.0 21.2 12.7 16.1 118Mother’s indigenous statusIndigenous 3.1 6.0 4.0 36.0 50Non indigenous 2.3 10.5 10.4 14.0 6112Source: ABS 1996 Census 1 per cent Confidentialised Unit Record File.Those recoded in 1 per cent sample file as having 4 or more children are treated as, onaverage, having had 4.5 children.Fertility rate by location in stateFigure 80.00.51.01.52.02.5NTTasWASAQldVicNSWRestCapitalStateSource: ABS (2000) Births, Australia, 1999.TotalfertilityrateFertility rate by remotenessFigure 91.01.21.41.61.82.02.22.4Very remoteRemoteModeratelyaccessible(outer regional)Accessible(innerregional)Highlyaccessible(major cities)Accessibility of regionSource: ABS (2002) Births, Australia, 2001.Fertilityrate
Family Matters No.63 Spring/Summer 2002 Australian Institute of Family Studies20among partners, as well as fertility difficulties due todelaying having children, may mean that some menand women do not have as many children as theyideally want (Qu, Weston and Kilmartin 2000; Quand Weston 2001).The World Values Survey of 1995-1997 askedAustralian men and women how many childrenthey had and how many they had wanted. Table 5reports the match between the ideal number andthe actual number of children among men andwomen aged 45 and over.The table shows that the patterns for men andwomen are remarkably similar. Just on 50 per cent ofthose aged 45 and over had their desired number ofchildren; overall men and women of this age grouphad smaller families than they originally wanted,with younger men and women (those aged 45-54)being the most likely to have their desired number ofchildren. Two-thirds of women in this age group hadtheir desired number of children compared to just on40 per cent of women aged 65 or over. Almost a quar-ter of men and women aged 45-54 had fewer childrenthan they ideally wanted, and only 11 per cent ofwomen and 14 per cent of men aged 45-54 had morechildren than they wanted.In the past, considerable effort has been given tohelping women control their fertility and limit thenumber of children they have. The observation thatalmost a quarter of women who have completedtheir childbearing have fewer children than theyideally want points to the need to understand thebarriers to achieving these fertility goals.ConcludingcommentsAustralian fertility has reached an historical lowpoint. Some people are alarmed at this low anddeclining fertility rate because of its social and eco-nomic implications. Others welcome it because ofthe link between population pressures and environ-mental degradation.indigenous children will mean that there will not besuch a stark gap between indigenous and non-indige-nous women in their number of living children.LocationWhere people live is linked to their fertility levels.Women living in capital cities have a lower fertilityrate than those living elsewhere. Figure 8 indicatesthat in each state women living in capital cities havea fertility rate of between 0.3 and 0.5 children lowerthan elsewhere.These differences between the capital cityand other rates will be due to a variety of factorsincluding a different population mix in differentregions. In the Northern Territory, for example, theindigenous population outside Darwin will con-tribute to the higher fertility in the rest of theTerritory. Education levels, job opportunities andcohabitation rates are also higher in capital citiesand these will undoubtedly be part of the reason forlower rates of capital city fertility.The part played by access to educational,employment and other services is partly capturedby the link between remoteness and fertility. TheABS has constructed the ARIA index to indicate theremoteness of a location from major populationcenters. Using this index, Figure 9 shows a clearlink between remoteness and fertility. In the mostremote areas the total fertility rate is 2.28 com-pared to that in the most accessible locations wherethe total fertility rate is 1.65.DesiredversusactualnumberofchildrenWhile it may once have been the case that manywomen ended up having more children than theywanted, the ready availability of contraceptionshould mean that most men and women are ableto avoid having more children than they want. Moreto the point, the competing demands of work andchildren, relationship breakdown and disagreementMatch between desired and actual number of children for men and women aged 45 and over,Australia, 1995-1997Table 5Men Women45-54 55-64 65+ All 45-54 55-64 65+ All% % % % % % % %Have more than wanted 13.8 23.4 22.2 19.7 10.8 27.5 28.2 21.8Had desired number 61.2 56.4 48.1 54.8 66.2 45.1 39.7 50.7Have fewer than wanted 25.0 20.2 29.6 25.5 23.1 27.5 32.1 27.5N 116 94 135 345 130 102 131 363Source: World Values Survey combined file, using the 1995-1997 wave.Australian fertility has reached anhistoricallowpoint,andsomepeoplearealarmedbecauseofthesocialandeconomicimplications.
21Australian Institute of Family Studies Family Matters No.63 Spring/Summer 2002The policy levers that are available to govern-ments to influence fertility are limited. To theextent that fertility reflects personal value prefer-ences and lifestyle choices governments only have avery limited role to play in influencing values andpreferences. Furthermore, declining fertility levelscan reflect other positive changes in our societyincluding better education and a wider range ofchoices for women. It would be both wrong andunacceptable to reduce opportunities for womenwith a view to increasing fertility.However, preferences can be influenced by the bar-riers that discourage having children. The role ofpolicy should be to enable couples to have the numberof children they choose. This may mean that policywill support proper family planning to limit family sizewhere desired, and to remove barriers and disincen-tives to having the desired number of children.Some barriers to fertility are not within thedomain of policy to influence – for example, there isrelatively little that policy can do to encourage menand women to either want a partner or to find asuitable partner. While relationship education andsupport services can assist people in developing andsustaining relationships there are nevertheless lim-its to the extent to which these programs can alterthe wider patterns of changes in partnering.However, there are some disincentives that socialpolicy can address. These disincentives include thefinancial costs of rearing a child and the opportunitycosts for women in taking time out of the workforce.Industrial relations policies that encourage family-friendly work practices and taxation structures thatdo not penalise parents, especially single incomefamilies, are all part of the mix that may removesome of the barriers to fertility.However, we still know too little about exactlywhat the barriers and disincentives to fertility are.We know that at a macro level changes in fertility areassociated with other macro level social, economicand cultural changes (see Castles in this issue). Butas well as understanding the structural, demographicand social context in which fertility occurs, there is aneed to understand the decisions individuals makeabout having children. While these decisions occurwithin a broad social, economic and cultural con-text, individuals still make decisions.A better understanding of fertility decision makingmust add this subjective dimension in order to gain afuller grasp of the factors behind declining fertilityrates. A number of studies in which the AustralianInstitute of Family Studies is currently involved willadd to this knowledge base. These include a jointstudy being conducted with the Office of the Status ofWomen, the HILDA panel survey (see article byWeston and Wooden in this issue), and the AustralianTemperament Project (see article by Smart in thisissue). When we have a better appreciation of howindividuals make (or do not make) their fertility deci-sions we will be in a better position to identify thetypes of policies that will enable men and women toachieve the levels of fertility they desire.Notes1 This is in the situation in which there is no net migration.Migration can help sustain stable population levels whenfertility rates are below 2.1. However, McDonald and Kippen(2000) and others have argued that, when the fertility levelsdecline to those currently seen in Australia and manyEuropean countries, the levels of migration required tomaintain a stable population are either unachievable or maybe politically unacceptable and potentially socially divisive.Furthermore, McDonald and Kippen maintain that immi-gration does not have much impact on the population agestructure.2 Those born in 1957 would have reached the end of theirchildbearing years by 1996 census3 This is the actual number of children born by women in thisage group. It is higher than the projected fertility rate of 1.72which is estimated from current fertility levels across theage groups of women of reproductive age.4 Although some of the actual fertility decline may result frominfertility that results from delaying childbearing.5 This estimate from the 1 per cent sample file is subject tosampling error because of small numbers. The full censusindicates that the actual figure is 3.24 (ABS 2001).ReferencesABS (1998a), Australian Social Trends 1998, AustralianBureau of Statistics, Catalogue No. 4102.0, Canberra.ABS (1998b), 1996 Census of Population and HousingInformation Paper: Socio-economic Indexes for Areas,Australian Bureau of Statistics, Catalogue No. 2039.0,Canberra.ABS (2000a), Australian Social Trends 2000, AustralianBureau of Statistics, Catalogue No. 4102.0, Canberra.ABS (2000b), Births, Australia 1999, Australian Bureau ofStatistics, Catalogue No. 3301.0, Canberra.ABS (2001), Births, Australia 2000, Australian Bureau ofStatistics, Catalogue No. 3301.0, Canberra.ABS (2002a), Births, Australia 2001, Australian Bureau ofStatistics, Catalogue No. 3301.0, Canberra.ABS (2002), Australian Social Trends 2002b, AustralianBureau of Statistics, Catalogue No. 4102.0, Canberra.AIHW (1998), Australia’s Mothers and Babies, 1996, NationalPerinatal Statistics Unit, Australian Institute of Health andWelfare, Sydney,Bryson, L., Strazzari, S. & Brown, W. (1999), “Shaping fami-lies: Women, control and contraception”, Family Matters,no. 53, pp. 31-38.Birrell, B. & Rapson, V. (1998), A Not So Perfect Match: TheGrowing Male/Female Divide, 1986-1996, Centre forPopulation and Urban Research, Monash University,Clayton.Carmichael, G. and McDonald, P (1999), “Fertility trends anddifferentials” , Paper presented to the Transformation ofAustralia’s Population 1970-2030 Conference, AustralianNational University, Canberra.McDonald, P. (1998), “Contemporary fertility patterns inAustralia: First data from the1996 Census”, People andPlace, vol. 6, no. 1, pp. 1-12.McDonald, P. & Kippen, R. (2000), “The implications of belowreplacement fertility for labour supply and internationalmigration, 2000-2050”, Paper presented to the 2000 meet-ing of the Populations Association of America, Los Angeles,March. Internet: http://demography.anu.edu.au/Publications/ConferencePapers/PAA2000/labourpaper.pdfMerlo, A. & Rowland, D. (2000), “The prevalence of childless-ness in Australia”, People and Place, vol. 8, pp. 21-32.Nassar, N. & Sullivan, E. (2001), Australia’s Mothers andBabies 1999, AIHW Catalogue No. PER 19, NationalPerinatal Statistics Unit, Australian Institute of Health andWelfare, Sydney.Rowland, D. (1998), “The prevalence of childlessness incohorts of older women”, Australasian Journal on Ageing,vol. 17, no. 1, pp. 18-23.Qu, L., Weston, R. & Kilmartin, C. (2000), “Children? Nochildren? Effects of changing personal relationships ondecisions about having children”, Family Matters, no. 57,Spring/Summer, pp. 14-19.Weston, R. & Qu, L. (2001), “Men’s and women’s reasons fornot having children”, Family Matters, no. 58, pp. 10-15.David de Vaus is the Senior Research Advisor at the Aus-tralian Institute of Family Studies and Associate Professorof Sociology at La Trobe University.