Australian Bureau of Statistics
1301.0 - Year Book Australia, 1994
Previous ISSUE Released at 11:30 AM (CANBERRA TIME) 01/01/1994
|Page tools: Print Page RSS Search this Product|
HOUSEHOLD AND FAMILY TRENDS IN AUSTRALIA
In 1991, 20.0 per cent of households were lone person households, 4.5 per cent were group households and, as would be expected, the large majority (75.4%) were family households. Some of the main characteristics of these three types of household are described below.
LONE PERSON HOUSEHOLDS
The proportion of lone person households has doubled in the past 30 years from 10 per cent in 1961 (Hugo, 1986). Before that, the proportion of lone person households had risen only slowly as evidenced by the fact that these households made up about eight per cent of all households in Victoria in 1901 (McDonald and Quiggin, 1985).
Several trends including lower marriage rates and increased divorce rates contribute to the increase in the proportion of one-person households. At the 1986 Census, 36 per cent of lone persons had never married and 22 per cent were divorced or separated (ABS, 1988). However, the most important determinant of the increase in one-person households is the ageing of the population. In June 1992, 56 per cent of all persons living alone were aged 55 years and over. Reflecting the earlier death of men compared to women, 70 per cent of these older people living alone were women (ABS, 1992a). Of relevance to policies related to housing, a majority (55%,) of persons living alone at the 1991 Census lived in separate houses (ABS, 1993).
In contrast to lone person households, group households were dominated by young persons.
In June 1992, 76 per cent of all persons living in group households were aged less than 35 years. Furthermore, at the 1986 Census, 76 per cent of those in group households had never married (ABS, 1988).
Family households are households containing one or more families. The ABS definition of a family for its statistical collections is as follows:
The ABS applies this definition for the Census of Population and Housing and most other statistical collections of data on families for reasons of statistical methodology, and to assist the collection of accurate data on what can be a complicated subject. However, the following qualifications about this approach need to be understood as background to the descriptions which follow on the various types of families to be found in family households.
There is a great deal of evidence that Australians in general do not define their families in this way. In particular, studies suggest that individuals very often regard their families as extending across households. For example, a child whose parents have separated usually considers the non-resident parent as part of his or her family. Indeed, most people see their parents or their children as being members of their family irrespective of where they live. Evidence also suggests that the flows of support (financial, practical and emotional) between households within the same family are significant and very common (McDonald, 1992: Millward, 1992).
Furthermore, people define their families differently for different purposes. The family members who attend at weddings are different to the family which gets together at Christmas time and this is different again from the family which shares financial interdependency. The growing proportion of people who do not live in family households does not mean necessarily that these persons are not members of families or that they are isolated from other family members. The elderly widow living on her own is often in close contact with her children and flows of support between her and her children are the usual circumstance.
The restriction of the ABS definition of family to only persons who live in the same household means that these very important dimensions of family life and family structure are missed, which can lead to the false impression that the extended family is not important to Australians. However, because the main source of data on families is the ABS, the remainder of the paper deals with statistics according to this definition, despite the limitations of this approach.
TYPES OF FAMILIES
As mentioned above, 75.4 per cent of all households in Australia in 1991 were family households. In 1991, only 1.0 per cent of all family households contained more than one family. Thus, family households (comprising one or more families) and household families (single family households) are near to being the same thing.
Diagram 1. AUSTRALIAN FAMILIES, JUNE 1992
Source:ABS, 1991 Census of Population and Housing (1993), Table B34.
The breakdown of household families into types is shown in Diagram 1.
Diagram I shows a total of 4,298,710 families in Australia, 85.3 per cent of which were couple families, 12.8 per cent were one parent families and 1.9 per cent were other families. Eighty-eight per cent of sole parents were women (ABS, 1992b).
The diagram shows that 52 per cent of couple families had dependent children. A further 37 per cent of couple families consisted of a couple with no offspring present. The remaining 11 per cent of couple families had other, that is, non-dependent offspring only.
The ABS defines a dependent child as a family member under 15 years of age or aged 15 to 24 attending an educational institution full time (ABS, 1992a). In today' s environment, other children aged 15 to 24 years and living with their parents, particularly those not in paid employment, may be equally if not more dependent on their parents than full-time students.
As mentioned in the previous paragraph, 37 per cent of all couple families consisted of the couple alone with no offspring present. These couple only families made up 23 per cent of all Australian households in 1991 and are the principal reason behind the fact that the most common household size is two persons. In terms of age, many of these couples are older couples whose children have all left home, the so-called empty-nest phase of life. For example, in 1986, for 69 per cent of couple only families, the woman was aged 35 years and over and had children (ABS, 1989). This is a growing group in the community because of increased life expectancy and the earlier ages at which parents of this generation had their children.
ONE PARENT FAMILIES
Among families with dependent children, 16.6 per cent were one parent families in 1991. This compares with 9.2 per cent in 1974, 13.2 per cent in 1981 and 14.6 per cent in 1986. Thus, the proportion of one parent families has been gradually rising over the past 20 years primarily as a result of the increase in the rate of marriage breakdown and, to a lesser extent, to the increase in births to women who are not married. Sixty-one per cent of sole parents in 1986 were divorced or separated (ABS, 1991). It is worth keeping in mind, however, that one parent families were equally common 100 years ago; in Victoria, in 1891, 16.7 per cent of all families with dependent children were one parent families (McDonald and Quiggin, 1985). The proportion of one parent families headed by men was higher in 1891 (38%) reflecting high levels of maternal mortality at the time.
Further details on one parent families are contained in Australia's One Parent Families (ABS, 1991). This publication indicates the generalised disadvantage of one parent families. For example, one parent families were much more likely than couple families with children to have low incomes; 63 per cent of one parent families were in the lowest 20 per cent of family incomes compared to only 12 per cent of couple families with children. One parent families were also much more likely to be in rented accommodation than couple families; 54 per cent of one parent families were renting compared to 20 per cent of couple families with children. Furthermore, 14 per cent of one parent families lived with another family compared to only 3 per cent of couple families with children. Sole parent mothers were less likely to be employed than mothers in couple families and sole parent families were less likely to have access to a car; 21 per cent of sole mothers families did not have a car compared to 2.6 per cent of couple families with children.
Reflecting ethnic differences in divorce rates and rates of birth to unmarried women, the proportion of mothers with dependent children who were sole parents was much lower among mothers born in southern European countries than it was for those born in Australia; in 1986, 3.5 per cent of mothers born in Greece were sole parents compared to 5.5 per cent of those born in Italy, 12.8 per cent of those born in the United Kingdom and 14 per cent of mothers born in Australia.
MARRIED AND DE FACTO COUPLES
There has been greater acceptance of de facto relationships in recent times. The percentage of all couples who are in a de facto relationship remains relatively low but is increasing. Among all couple families at the time of the 1991 Census, 91.8 per cent were married and 8.2 per cent were de facto couples (ABS, 1993). At the 1986 Census, 5.7 per cent of couples were de facto. Among couple families with dependent children, the proportion living de facto was even lower (6.5%). De facto couples are more prominent at younger ages; 62 per cent of couples where the woman was aged 15 to 19 and 25 per cent where she was aged 20 to 24 were de facto couples at the time of the 1986 Census (ABS, 1989). However, most of these young de facto couples do not have children.
NATURAL PARENTS AND STEP-PARENTS
For the first time, at the 1986 Census, the ABS introduced a question on household relationships which allowed step-parents and stepchildren to be identified. Among couple families with dependent children, 6.8 per cent contained stepchildren (ABS, 1989). Half of all de facto couples with dependent children had stepchildren. From the perspective of children about five per cent of children aged less than 15 years were living with a step-parent.
As expected, the chance of living with both natural parents declines as the child gets older. In Western Australia in 1986, 91 per cent of one-year old children lived with both their natural parents but as children became older, this percentage fell to 85 per cent at age 6, 80 per cent at age 12 and 77 per cent at age 15 (Government of Western Australia, 1988). Thus, while children may experience a variety of family types, about three-quarters still continue to live most of their childhood with both their natural parents.
TRANSITIONS IN FAMILIES
Families change mainly through the forming of couples through marriage or de facto relationships, births, deaths, relationship breakdown, children leaving home, and the taking-in of relatives for reasons of support. All of these processes have been subject to substantial changes in the past 20 years. In turn, these changes have led to the variations in household and family types already described. In addition, changes in the labour force and education activities of family members have had profound effects on family functioning and family structure over the past 30 years. The following sections review the main changes in these processes.
The best point to begin a discussion of family transitions is the point at which young people leave their parents' home. This is a convenient starting point because in the vast majority of cases, leaving the parental home precedes the formation of the person's first co-residential relationship. Them is a definitional problem, however, because it is common today for the young people to leave home and then to return again as their life circumstances change. Here, two measures are used, the time at which young people first leave home and the proportion of young people who are living with their parents at a point in time.
In the 1950s and 1960s, leaving the parental home was associated closely with other observable changes in people's lives such as marriage, employment or education. For example, among women born around 1950, 49 per cent first left home for marriage while 18 per cent left for work or study. Only a small percentage left home simply to be independent (11%). The age at which they left home, therefore, tended to correspond with the timing of these other transitions, in particular, marriage. The shift to earlier marriages between 1940 and 1970, therefore, meant that young people left home at earlier ages during this period. For those born around 1950, the average age at first departure from the parental home was 21.5 years for men and 20.2 years for women (Young, 1987; Glezer, 1993).
In the 1970s, the reasons for leaving home changed but the age at first leaving home remained about the same. The percentage leaving home to marry dropped by about 10 percentage points for both men and women, but this drop was almost exactly matched by the increase in those who left home to live together in a de facto relationship. The 1970s saw a sharp rise, however, in the proportion who said that they had left home in order to be independent. Indeed, for men, independence became more significant than marriage as a reason for leaving home. Glezer (1993) has shown, for example, that 18 per cent of men born between 1947 and 1955 had left home to be independent compared to 29 per cent among men born from 1955 to 1963.
During the 1980s, the work and education environment changed dramatically for young people. Retention to Year 12 at secondary school rose from 34 per cent in 1980 to 64 per cent in 1990 (DEET 1991). Involvement in tertiary education also increased rapidly. Along with these changes, age at marriage continued to increase.
The changes in the lives of young people in Australia in the past 30 years are reflected in statistics relating to their involvement in full-time education. The proportion of 15 to 19 year olds who were at school or full-time students increased between 1961 and 1992 from 28 per cent to 62 per cent for males and from 24 per cent to 69 per cent for females (ABS, 1967 and ABS, 1992c). Over the same period, involvement in full-time education increased for 20 to 24 year-olds from 4 per cent to 14 per cent for males and from 1 per cent to 16 per cent for females. Many of today's young people combine part-time work with full-time education. For example, in August 1992, 47 per cent of full-time tertiary students aged less than 25 years and 31 per cent of school students aged 15 to 19 years were in the labour force. As a consequence of the changes in the 1980s, young people aged less than 25 years are more likely to be at home than was the case in the past (Table 2).
TABLE 2 PERCENTAGE OF 15-19 AND 20-24 YEAR OLDS LIVING WITH THEIR PARENTS 1979, 1986 AND 1992(a)
Source: ABS, Labour Force Status and Other Characteristics of Families, 1979, 1986 and 1992 (6224.0).
Beyond living with their parents, many young people aged 15 to 24 years who are not living with their parents may still be, to a greater or lesser degree, financially dependent on parents (Hartley, 1989). Thus, the increase in the dependency of these young people is understated by the percentages shown in Table 2. A recent study of 23 year-olds by Hartley (forthcoming) indicated that some of those who were still at home with their parents were somewhat apologetic about this, indicating that they were exceeding the social norm. The increased financial dependency of young people upon their parents together with their desire for at least as much independence as previous generations is one of the tensions facing families today.
THE FORMING OF RELATIONSHIPS
The next step in the sequence of family transitions is the formation of a new couple relationship. The three main forms of couple relationships now prominent in Australia are marriage, de facto or living together relationships (heterosexual) and same-sex relationships. There are no reliable statistics available on the incidence of same-sex couples because no Australian survey has addressed this question directly. However, surveys providing indirect estimates would suggest that this group represents less than one per cent of all couples (see, the example, Glezer, 1993).
As mentioned above, de facto relationships account for about eight per cent of all couples. De facto relationships are much more prevalent among couples aged less than 25 years compared to older ages, but many of these young relationships are preliminary to marriage; in a sense, they are trial marriages. Glezer (1993) has shown that 44 per cent of persons aged 27 to 35 years in 1990 who had married had lived together before their marriage. Thus, while other forms of relationship are becoming more common and are more socially acceptable, couple relationships in Australia are still dominated by marriage.
The percentage of Australian women who never marry during their lifetime has been on a long-term swing from being high (14-17%) for generations born in the latter half of the nineteenth century, dropping to low levels (4-6%) for those born between 1920 and 1950, but again returning to the formerly high levels of the past for those born since 1950. A recent prediction suggests that the proportion of Australian women not married by the time they reach age 35 will exceed 20 per cent by the year 2000, a level higher than at any other time in Australian history (McDonald, 1991). Proportions of men who never marry follow similar trends to those shown for women, although, in general, the proportion of men never marrying has been higher than that for women (Carmichael, 1988).
The ages at which Australians marry have been subject to the same long-term swing as the proportions who never marry. A useful indicator of age at marriage for women is the proportion aged 20 to 24 who have ever married. For men, age group 25 to 29 provides the best indication of the trend. These proportions are shown in Table 3.
TABLE 3 PERCENTAGE EVER MARRIED WOMEN AGED 20-24 YEARS AND MEN AGED 25-29 YEARS, 1891-1991
Up to the 1930s, about 30 per cent of women aged 20 to 24 had married. Beginning with the Second World War and continuing through the post-war period to 1971. age at first marriage fell substantially in Australia, so that by 1971, the percentage of 20 to 24 year-old women who had married had risen to 64 per cent. A similar trend to a much earlier age at marriage is also evident for men in the same period; just over 40 per cent of men aged 25 to 29 had married before the war, but the percentage had reached 74 per cent by 1971. Since the early 1970s, the trend to early marriage has very rapidly gone into reverse, so that, by 1991, the percentage of women married at young ages was lower than it has ever been.
The remarkable turnaround in age at first marriage is further evidenced by the decline of teenage marriages. Fully 33 per cent of women turning 20 in 1972 had married in their teen years (McDonald, 1991); by 1991, this percentage had fallen to under five per cent (ABS, 1992d).
It is important to realise, however, that age at marriage in a large proportion of cases does not indicate the age at which the couple first began to live together. As mentioned above, almost half of couples marrying for the first time have lived together beforehand and, indeed to a considerable extent social attitudes now favour this approach. About half the respondents in the 1988-89 National Social Science Survey indicated that they would recommend young people live together with a steady partner and then marry. Very few recommended that couples live together without ever marrying (Evans, 1991). That is, at present, the approach of young Australians is to marry (80% of young people intend to marry), but to delay marriage to much older ages than the ages at which their parents married. In the meantime, many will live in de facto relationships. The vast majority of young people following this pattern do not have children until they have married Glezer 1993).
FAMILIES AND THE LABOUR FORCE
Changes in rates of participation in education of persons aged less than 25 years have already been described; it was evident that those changes were highly associated with new patterns of leaving home and forming new relationships. Another major trend affecting family behaviour, particularly after a relationship is formed, is the increasing participation of married women in the labour force (which comprises those in, or actively seeking, paid employment).
Table 4 shows a gradual and persistent increase in the participation of married women in the labour force, with the sharpest rise occurring between 1961 and 1971 when restrictions on the employment of married women in the public sector were lifted. By 1991, two-thirds of married women in the peak labour force ages of 25 to 54 years were in the labour force. Studies of working lifetimes indicate that most of the remaining one-third of married women are in paid employment at some time after their marriage. Thus, where marriage once spelt an end to the working careers of most women, married women today define their lives not only in terms of home and children but also in terms of paid employment.
TABLE 4 LABOUR FORCE PARTICIPATION RATES OF MARRIED WOMEN, BY AGE GROUP 1933-91
Among all women in the peak labour force ages of 25 to 54 years, participation in the labour force increased from 37 per cent in 1966 to 67 per cent in 1992. However, over the same period, among those employed, the percentage working full time dropped from 70 per cent to 57 per cent. Thus, the growth of the female labour force has been associated with the trend towards part-time jobs. The overall proportion of persons working part time rose from 9 per cent in 1966 to 23 per cent in 1992. Nevertheless, as just pointed out, a majority (57%) of employed women aged 25 to 54 were working full time in 1992. For women with dependent children, however, the percentage working full time among those employed was lower (41.7% for women in couple families and 53.9% in one parent families) (ABS, 1992a).
For women with dependent children, the degree of participation in the labour force is related to the age of the youngest child (Table 5). In couple families, the participation rate of women with a child aged less than 5 years is still below 50 per cent. Once the child reaches school age, however, participation rates for these women jump to about 70 per cent. The percentage working full time also rises slowly as the youngest child ages. participation rates for sole parents are lower than those for women in married couple families when the youngest child is aged less than 15 years. However, if a sole parent is employed, she is more likely to be working full time than women in couple families irrespective of the age of the youngest child. Sole parents whose youngest dependent child is aged 15 to 24 years have by far the greatest participation in the labour force of all mothers distinguished in the table. It can be speculated that the involvement of sole mothers in the labour force is influenced by the social security system. Loss of social security benefits and costs of child care make it more difficult for sole parents to work part time, so they are more likely than women with children in couple families to work full time or not to work at all. However, once their youngest child turns 16 years, they are no longer eligible for the sole parent pension. so their labour force attachment is much greater.
TABLE 5 WOMEN WITH DEPENDANT CHILDREN: LABOUR FORCE PARTICIPATION RATES AND PERCENTAGE EMPLOYED FULL TIME, ACCORDING TO THE AGE OF YOUNGEST CHILD AND FAMILY TYPE, JUNE 1992
In June 1992, there were almost 2 million couples with dependent children in Australia. The percentage where both parents were in the labour force was 59.1 per cent, but the percentage with both parents employed was 53.3 per cent. Thus, where both parents were in the labour force, either partner was unemployed in 9.8 per cent of these families. Where both the partners were employed, they were both working full time in 39.3 per cent of cases. Just over 400,000 dependent children representing 10 per cent of all dependent children in couple families had neither parent employed. On the other hand, 52 per cent of children in couple families had both parents employed and 19 per cent had both parents employed full time (ABS, 1992a).
Also in June 1992, there were 412,100 one parent families. The parent was in the labour force in 54.7 per cent of these families and employed in 45.7 per cent. Thus, among sole parents in the labour force, the parent was unemployed in 16 per cent of cases. Where the sole parent was employed, 60 per cent were employed full time. About 379,000 dependent children representing 56 per cent of children living in one parent families did not have an employed parent (ABS, 1992a).
Overall, therefore, among 4.7 million dependent children aged less than 25 years in June 1992, 16.4 per cent had two parents employed full time, 60.3 per cent had one parent employed full time, 6.7 per cent had an employed parent but no parent employed full time and 16.7 per cent had no employed parent. These statistics indicate the diversity of family types and family employment circumstances now prevalent in Australia. Decisions about family formation are taken in the context of these circumstances (ABS, 1992a).
In June 1992, the family status of employed men and women was almost exactly the same: 42 per cent had a dependent child, 25 per cent had a spouse but no dependent child and 33 per cent had neither a spouse nor a dependent child.
There has also been a long-term downward swing in the proportion of Australian women who remain childless. For Australian women born between 1861 and 1913, 20 per cent or more did not have children. The peak level was around 25 per cent for women born in the 1870s. The mothers of the baby boom generation changed this so that for women born in the 1930s and 1940s, the rate of childlessness fell to 8-10 per cent, only marginally above the expected rate of childlessness due to physiological reasons alone. The trend, however, has been rapidly reversing for women born from 1950 onwards, such that a return to percentages childless in excess of 20 per cent can be predicted for women born in the late 1960s (McDonald, 1984).
There is a long-run trend in the average number of children born in a lifetime to women who marry. The most significant feature of this trend is the decline in the average number of births per woman from 5.1 for women born in 1861-66 to 2.6 for women born in 1903-08. This was the era in which control over fertility became widely established. Reflecting the onset of the baby boom, however, completed family size for women who had married increased again reaching a peak of 3.2 children per woman for those born from 1928-38. Since then family sizes have declined again and can be predicted to be only just above an average of two children per married woman for those born in the first part of the 1960s.
The recent decline in average family size is accompanied by a decline in the percentage of women who are having three or more children. About 60 per cent of the generation of married women who had peak fertility during the baby boom (those born in the late 1920s and the 1930s) had three or more children. For those born in the late 1940s, this percentage had fallen to 42 per cent (ABS, 1992e). On the basis of birth rates applying in 1988, 32 per cent of all women (not just married women) will have three or more children, 24 per cent will have two, 24 per cent will have one and 20 per cent will have none (McDonald, 1990).
The discussion so far has related to the number of children born to women over their lifetime. This measure is limited by the fact that it can only be obtained when women have reached about 40 years of age. Another approach to the measurement of birth rates uses the total fertility rate (TFR).
The TFR indicates the average number of births that would be born over a lifetime to a hypothetical group of women if they were to experience the age specific birth rates (that is, births per thousand women in specific age groups) applying in a given year. The so-called baby boom, a period of unusually high birth rates extended from 1946 to 1971 (TFR greater than 2.8 births per woman). At its peak in 1961, the TFR reached 3.6 births per woman. The boom collapsed in the early part of the 1970s with the TFR falling below 2.0 in 1978. Since 1979, it has fluctuated in a narrow band between 1.84 and 1.92 births per woman. At this level, the TFR is below a level consistent with long-term replacement of the population (just over two births per woman).
Nevertheless, the total fertility rate has been higher in Australia in the 1980s than in almost all other western industrialised countries. Very recently, however, the TFR has risen in several countries, such as the Scandinavian countries and the United States (Table 6). Most demographers interpret this recent rise in TFR in developed countries as a temporary rise due to the occurrence of births which were delayed in the mid-1980s.
Birth rates for younger aged mothers have declined sharply in the past 20 years (Table 7). In 1971, births to mothers under the age of 25 years contributed 40.1 per cent of the total fertility rate compared with only 26.1 per cent in 1991. The mean age of mothers at the birth of their children thus rose from 27.0 to 28.5 years.
Also between 1971 and 1991, the percentage of first nuptial births occurring to women aged 30 and over rose from 7.6 per cent to 31.1 per cent while the number of teenage confinements fell from 30,500 to 14,600. The drop in births to teenage women in the past 20 years has been due entirely to the drop in births to married teenagers from 20,500 in 1971 to 2,600 in 1991. This in turn has been due to the near disappearance of the 'shot-gun' wedding. In 1971, about 13,000 teenage brides were pregnant at marriage; in 1991, the figure had fallen to only 1,100. Now about one in ten Australian women have a baby before their 20th birthday compared to about one in four in 1971 (McDonald, 1988; and calculations from ABS, 1992f).
Besides the decline in fertility and the increasing age of mothers, the other significant change in fertility in the past 20 years has been the increase in ex-nuptial births, that is, births occurring to women who are not married. In 1971, ex-nuptial births represented 9.3 per cent of all births but this had risen to 23.0 per cent by 1991, that is, almost one in every four children born are now born to a woman who is not married (ABS, 1992f). The change is not due to an increase in the rate at which women who are not married give birth but to similar rates of birth occurring to the now much higher percentages of women who are not married compared to 1971 (McDonald 1988).
In 1991, paternity was acknowledged on the birth certificates of almost 80 per cent of children whose mothers were not married (ABS, 1992f). Thus, being born to a woman who is not married does not imply that children are cut off from their fathers. In many of these cases of ex-nuptial birth, the child is born to parents who are living together but are not married. Reliable statistics are not available, but it seems that about 50 per cent of all ex-nuptial births occur to women who are in a de facto relationship (Khoo and McDonald, 1988).
THE ENDING OF RELATIONSHIPS
Suppose we consider a group of couples who marry when the wife is 25. The husband is two and a half years older. This is not unlike the present average ages at first marriage. It is also not unlike the average ages at first marriage which applied 100 years ago. Using 1891 death rates, 30 years after the marriage, both partners would still be living in only 46 per cent of cases. Using 1991 death rates, both would still be living in 88 per cent of cases. Forty-five years after the marriage when the wife turns 70, both partners would still be living in just 15 per cent of cases with 1891 death rates but 56 per cent of cases with 1991 death rates (1891 survival rates from McDonald, Ruzicka and Pyne, 1987; 1991 survival rates from ABS, 1992g). This example serves to indicate the dramatic changes in expectation of life together of today's couples compared to those 100 years ago. Some may argue, given this reality, that it is not surprising that divorce rates are higher today than they were 100 years ago.
A century ago, formal divorce rates were very low with perhaps only one per cent or so of marriages ending in divorce within 30 years of the marriage. In the next 70 years, divorce rates rose slowly to a level of about 10 per cent of marriages ending in divorce by the mid-1960s. While the evidence is persuasive rather than conclusive, it has been argued that this rise over 70 years merely reflected an increasing trend towards formalisation by divorce of marriage breakdowns, rather than any real change in the rate of marriage breakdown (James, 1984). Thus, in the hypothetical case set out in the previous paragraph, 100 years ago, we could have expected about 10 per cent of couples surviving 30 years after their marriage to have separated. The percentage still together after 30 years, taking into account both widowhood and separation, would therefore be about 41 per cent.
Today, most marriage breakdowns are formalised by divorce. About 37 per cent of marriages can be expected to end in divorce (Carmichael and McDonald, unpublished). Adding a small fraction for marriage breakdowns which are never formalised by divorce, we can estimate that about 40 per cent of marriages end in separation within 30 years of the marriage. Thus, again considering the hypothetical case described above, taking into account both widowhood and separation, the percentage of today's couples who can expect to be still together after 30 years is 53 per cent. Overall, then, today's married couples are more likely to be still living with their spouse after 30 years than couples 100 years ago.
The example provides another useful basis for discussion because Young (1989) has shown that in Australia in recent times, the last child leaves home on average when the mother is 54 and the father is 57. Thus, we can estimate that over 50 per cent of couples can expect their first marriage to survive to the point where all their children have left home. Others will go into this 'empty nest' phase through repartnering after the ending of their marriage. Thus, a majority of Australians can expect to spend some time living together with their partner with adult children living elsewhere. One hundred years ago, this family phase hardly existed at all because of high mortality, late child-bearing and the tendency of at least one unmarried child to remain at home (McDonald and Quiggin, 1985). The empty nest phase is also not short; the number of widowed women in Australia does not exceed the number of married women until age 73 while the number of widowed men never exceeds the number of married men at any age (ABS, 1992d).
The modem rise in divorce rates in Australia began in the late 1960s and accelerated during the 1970s. Having reached a level equivalent to between 35 and 40 per cent of marriages ending in divorce, the rate of divorce levelled off during the 1980s. The trend in the number of divorces between 1951 and 1991 shows an upturn in the late 1960s leading to the very large number of divorces occurring in 1976, the first year of operation of the Family Law Act 1975. The 'no fault' provisions of the new Act were attractive to many people who had been separated for some time and also enabled many people to divorce sooner than they would have done under the previous law. These factors led to a heaping of divorces into one year, 1976. By the end of the 1970s, this temporary peak had passed. A secondary peak occurred around 1982-83, perhaps reflecting the economic recession of the early 1980s. To the extent that divorces are affected by recessions, we can expect the divorce rate to rise again in the early 1990s and there was some evidence of such a rise in 1991.
The other major effect that the new law has had is in bringing divorces forward within a marriage. Whereas under the previous law, a couple had to have a five-year separation before 'no fault' provisions applied, with the new law this waiting period was reduced to just one year. Almost immediately the new law was introduced, the proportion of marriages ending in divorce within five years of the marriage rose from about one per cent to eight per cent. Today, about eight per cent of marriages end within 5 years, 17 per cent within 10 years, 28 per cent within 20 years and 37 per cent within 30 years. This progression indicates that the rate of divorce is not greatly affected by the length of the marriage, with about one per cent being added to the percentage divorced for each additional year of marriage. That is, there is no special time in a marriage when divorces are more likely to occur than any other time, except perhaps in the first few years of the marriage. From the perspective of children, this means that about one per cent can expect to experience the divorce of their parents for each year of life. That is, by age 10 about 10 per cent of children have divorced parents; by age 20, the percentage has risen to 20 per cent (Carmichael and McDonald, unpublished).
From an international perspective, Australia's divorce rate is on the low end of those countries with high divorce rates. Its rate of divorce is much lower than that of the United States and the Scandinavian countries and is also lower than that of England and Wales and Canada. On the other hand, Australia's divorce rate is far higher than the rates applying in Italy and Greece (Table 8).
The divorce rate from second marriages is a little higher than that from first marriages; about 38 per cent compared to about 34 per cent in 1986 (McDonald, 1988). Little is known about the breakdown of de facto relationships, but the rate is likely to be quite high. In a sample of de facto couples who had a birth in 1984, 19 per cent were no longer together 18 months after the birth of the child (Khoo and McDonald, 1988).
Prior to the major increase in divorce rates which occurred in the mid-1970s, remarriage rates for divorced persons were relatively high, with almost a quarter of all divorced persons aged between 25 and 40 years marrying in a given year (Table 9). The divorce boom since the 1970s has been accompanied by very much lower rates of remarriage, which continue to decline. In all years, remarriage rates were higher for men than for women.
The decline in remarriage rates after divorce may reflect the increasing availability of other relationship options such as de facto relationships and relationships in which the partners do not live together. An AIFS study in 1987, found that four to six years after divorce from their first marriage, the relationship patterns shown in Table 10 prevailed among a sample of people who had been married between 5 and 14 years and had two children from the marriage. Most had formed new relationships, especially the men, but the new relationship did not necessarily take the form of marriage. The women, who usually had the children of the marriage with them, were twice as likely, as men to have a relationship with a person with whom they did not live.
The International Year of the Family comes at a time when Australians, like people in other industrialised countries, are re-establishing the meaning of family in their lives. While there has always been a variety of family types in Australia, the breadwinner model of family consisting of a father in paid employment, a mother in the home and two to four children has been the dominant cultural image of the family. In the more distant past, the variety of family forms was due more to necessity than choice, for example, sole parent family arising from the death of one of the parents. Over the past 30 years, however, many Australians have chosen to live in family forms which are different to the long-held cultural image of the breadwinner family and their decisions have met with increasing levels of acceptance by the community.
In the re-establishment of the meaning of family, it can be expected that less emphasis will be placed on the structure of families and more upon the ways families work, that is, on the nature and quality of family relationships both within and across households.
ABS (1967), Census of the Commonwealth of Australia, 30th June 1961, Volume VIII. Australia, Statistician's Report (5429/65), Canberra
ABS (1984), Australian Families, 1982 (4408.0.), Canberra
ABS (1986), The 1986 Census Dictionary (2174.0), Canberra
ABS (1988), 1986 Census of Population and Housing, Format CSD 21 Page, Canberra
ABS (1989), Census 86 - Australian Families and Households (2506.0), Canberra
ABS (1991), Census 86 - Australia's One Parent Families (2511.0), Canberra
ABS (1992a), Labour Force Status and Other Characteristics of Families, June 1992 (6224.0), Canberra
ABS (1992b), Facts on Families, 1992, Canberra
ABS (1992c), The Labour Force Australia, August 1992 (6203.0), Canberra
ABS (1992d), Estimated Resident Population by Marital Status, Age and Sex, Australia, June 1990 and Preliminary June 1991 (3220.0), Canberra
ABS (1992e), Census 86 - Fertility in Australia (2514.0), Canberra
ABS (1992f), Births, Australia 1991 (3301.0), Canberra
ABS (1992g), Deaths, Australia 1991 (3302.0), Canberra
ABS (1993), Census 91 - Basic Community Profile, Australia (2722.0), Canberra
CARMICHAEL, G. (1988), With This Ring: First Marriage Patterns, Trends and Prospects in Australia, Department of Demography, Australian National University and Australian Institute of Family Studies, Canberra
CARMICHAEL, G. and MCDONALD, P. (unpublished), 'The rise and fall of divorce in Australia', Australian Institute of Family Studies, Melbourne
DE GUIBERT-LANTOINE, C. and MONNIER, A. (1992), 'La conjoncture demographique: l'Europe et les pays developpes d'Outre-Mer', Population, July-August
DEPARTMENT OF EDUCATION, EMPLOYMENT AND TRAINING (1991), Retention and participation in Australian schools, 1967 to 1990, Australian Government Publishing Service, Canberra
EVANS, M. (1991), 'Alternative to marriage', National Social Science Survey Report, 2 (5), Canberra
GLEZER, H. (1993), 'To tie or not tie the knot: pathways to family formation', Paper delivered to 4th Australian Family Research Conference, Sydney, Australian Institute of Family Studies, Melbourne
GOVERNMENT OF WESTERN AUSTRALIA. (1988), Families in Focus in Western Australia, Perth
HARTLEY, R. (1989), What Price Independence: A report on a study on young people's incomes and living costs, Youth Affairs Council of Victoria, Melbourne
HARTLEY, R. (forthcoming), 23: Becoming Adult in Australia, Australian Institute of Family Studies, Melbourne
HUGO, G. (1986), Australia's Changing Population: Trends and Implications, Oxford University Press, Melbourne
JAMES, M. (1984), Marriage and Marital Breakdown in Victoria, 1860-1960, PhD thesis, La Trobe University, Melbourne
KHO0, S. and MCDONALD, P. (1988) 'Ex-nuptial births and unmarried cohabitation in Australia', Journal of the Australian Population Association. 5 (2), November, Canberra
MCDONALD, P. (1984), Can the family survive?, Discussion Paper No. 11, Australian Institute of Family Studies, Melbourne
MCDONALD, P. (1988), 'Families in the future: the pursuit of personal autonomy', Family Matters, No. 22, December, Australian Institute of Family Studies. Melbourne
MCDONALD, P. (1990), 'The 1980s: Social and economic change affecting families, Family Matters, No. 26, April, Australian Institute of Family Studies, Melbourne
MCDONALD, P. (1991), 'The shift away from marriage among young Australians', Family Matters, No. 30, December, Australian Institute of Family Studies, Melbourne
MCDONALD, P. (1992), 'Extended family in Australia: The family beyond the household', Family Matters, No. 32, August, Australian Institute of Family Studies, Melbourne
MCDONALD P. and QUIGGIN, P. (1985), Lifecourse transitions in Victoria in the 1880s'. in Grimshaw, P., McConville, C and McEwen, E., Families in Colonial Australia, George Allen and Unwin, Sydney
MCDONALD, P., RUZICKA, L and PYNE, P. (1987), 'Marriage, fertility and mortality', in Vamplew, W., Australians: Historical Statistics, Fairfax, Syme and Weldon, Sydney
MILLWARD, C (1992), 'Keeping in touch: Extended family networks', Family Matters, No. 32, August, Australian Institute of Family Studies, Melbourne
WESTON, R. and KHOO, S. (1993), 'Family reformation: a key adjustment strategy', in Funder, K., Harrison, M. and Weston, R., Settling Down: Pathways of Parents After Divorce, Australian Institute of Family Studies, Melbourne
YOUNG, C (1987), Young People Leaving Home in Australia, Department of Demography, Australian National University and Australian Institute of Family Studies, Canberra
YOUNG, C (1989), 'The effect of children returning home on the precision of the timing of the leaving-home stage', in Grebenik, E., Hohn, C. and Mackensen, R., Later Phases of the Family Life Cycle: Demographic Analysis,. Clarendon Press, Oxford
These documents will be presented in a new window.
This page last updated 18 June 2009