Stats NZ has a new website.

For new releases go to

As we transition to our new site, you'll still find some Stats NZ information here on this archive site.

  • Share this page to Facebook
  • Share this page to Twitter
  • Share this page to Google+
Ethnic differentials in induced abortions in New Zealand - article

Key Statistics - article, September 2001, p. 9-14

Takes a brief look at existing ethnic differentials in induced abortions in New Zealand. Uses births/abortions data for 1997-2000. The analysis covers characteristics of abortion parents with attention drawn to limitations of demographic data used.

Ethnic Differentials in Induced Abortions in New Zealand1


In recent years, induced abortion has grown significantly in importance as a factor in regulating and influencing fertility despite the availability of a wide range of contraceptive methods. Two decades ago, in New Zealand, one in 10 known pregnancies were terminated by an abortion. By the turn of the millennium, the figure had doubled to one in five. Demographic literature suggests that the propensity to terminate a pregnancy is relatively high among never married women, among women who had had no previous children or no previous abortion, as well as among those who belong to an ethnic minority group (Sceats, 1986; Statistics New Zealand, 2001; Bankole et al, 1999 ). An understanding of these differentials and the underlying reasons for recourse to abortion, is essential for identifying the population sub-groups, which should be targeted for family planning education and greater access to contraceptive methods.

This article takes a brief look at the existing ethnic differentials in induced abortion in New Zealand, using births/abortions data for 1997-2000. The analysis covers selected characteristics of abortion patients, notably their age, marital status, previous number of children as well as previous abortion experience. Attention is also drawn to the general limitations of the basic demographic data involved.

Data sources and limitations

Inter-ethnic comparisons in socio-demographic phenomena are often fraught with definitional and other statistical problem (Petersen, 1997). An added issue here is the lack of apparent consistency between the ethnic-specific abortions (the numerator) and births (the denominator), used in calculating the abortion ratio. Lacking reliable data to derive more refined indices, the abortion ratio was the demographic measure used for comparing the abortion experiences of the various ethnic groups.

The ethnic-specific statistics on live births and abortions are compiled from two independent sources - the birth registration form and the notification of abortion form. Both sources follow the broader concept of self-identification, but use different questions to collect information on the ethnicity of the respondent. For births, the ethnic question on the child’s mother is consistent with that used in the 1996 Census personal questionnaire. It asks which ethnic group(s) the mother belongs to, and makes provisions for multiple responses (Khawaja, 1997). In the case of abortions, the ethnic inquiry is less comprehensive. In fact, the practice may vary from institution (clinic/hospital) to institution. These reservations are relevant to interpreting the ethnic differentials in the incidence of induced abortion.

While the analysis relates broadly to the concept of ethnic self-identification, the abortions and the abortion ratios, focus primarily on the ‘sole’ ethnic group, ie women who specified that they belong solely to a given ethnic group. In all, four ethnic groups, European, Māori, Pacific and Asian populations are considered.

All statistics used in this paper are drawn from Demographic Trends 2000, the Hot Off The Press published on 11 June 2001 entitled ‘Abortions: Year ended December 2000’ or are based on special unpublished tabulations.

About ethnic populations

Recent censuses have indicated a significant change in the ethnic structure of the New Zealand population, with ethnic minorities (Māori, Pacific and Asian people) increasing their share of the total population. At the 1996 Census, the indigenous Māori made up 15.0 percent of all New Zealanders, the Pacific group 5.0 percent, and the Asian population, who are comprised largely of recent immigrants, another 4.6 percent. The majority of New Zealanders (three in four) belonged predominantly to the European group.

These various ethnic groups differ in their growth patterns, spatial distribution, reproductive norms and socio-economic make-up. The Māori and Pacific populations, for example, have higher fertility, higher growth rates and a more youthful age structure than the European or Asian populations. Because of their higher fertility, and a larger proportion of women in prime childbearing ages, they also account for a much larger share of live births than the resident population - one-third versus one-quarter (Khawaja, 1997).

General trend in induced abortions

Looking at the overall trend first, the number of induced abortions has increased by 171 percent in the last two decades, from less than 6,000 in 1980 to over 16,000 in 2000. While the general trend has been upward, there have been small drops in the annual number of abortions on three occasions, viz 1985, 1992 and 1998.

The general abortion rate (number of abortions per 1,000 women aged 15-44 years) more than doubled over the same period, from 8.5 to 19.0 per 1,000. Exact circumstances for this continuing rise are not known, but in some countries the recent increases have been associated with the publicity about possible risks of third-generation oral contraceptives (Wood et al, 1997).

Internationally, our abortion rate well exceeds that for the Netherlands, Germany and Finland - all with rates below 11 per 1,000, but is lower than that of Australia and the United States - both above an estimated 22 per 1,000 (Henshaw et al, 1999; Statistics New Zealand, 2001).

Graph, Abortions and Abortion Rates.

Abortions by ethnicity

European women accounted for over half (8,965 or 56 percent) of all abortions in the year 2000 and Māori women another 17 percent (Table 1). Pacific and Asian women who together make up just under 10 percent of all New Zealand women accounted for 22 percent of all abortions in 2000. Pacific women averaged slightly higher levels of abortion during the years 1997-99 (1,716 per annum) than their Asian counterparts (1,544). However this situation reversed in 2000, as Asian women experienced a 16 percent jump in induced abortion, while Pacific women recorded a 5 percent fall.

The abortion ratios for various ethnic groups during 1997-2000 are compared in Table 1. The abortion ratio is defined as the number of abortions per 1,000 known pregnancies (live births, stillbirths and abortions combined). In all four years, Asian women had the highest abortion ratio and European women the lowest. In 2000, about one in three known pregnancies of Asian women were terminated by an abortion, compared with one in five for European women. The Asian ratio (311 per 1,000) was 48 percent higher than for the latter (210 per 1,000). With ratios of around 255 per 1,000 known pregnancies, Māori and Pacific women occupied an intermediate position, although their abortion ratios were at least 14 percent above the national average of 220 abortions per 1,000 known pregnancies.

Table, Abortions by Ethnicity.

The year 2000 also saw a minor transformation in the ethnic rankings, with Māori women recording a marginally higher abortion ratio than the Pacific women. This resulted from opposing trends in their ratios.

Abortions by age

Analysis by age and ethnicity shows that the Asian women seeking abortions are considerably older than their counterparts. In 2000, those aged 30-44 years accounted for over two-fifths (42 percent) of all Asian abortion patients (Table 2). This was almost twice the corresponding figure (23 percent) for Māori women, who commence childbearing at an earlier age. The European and Pacific women occupied intermediate positions with figures of 31 and 33 percent respectively. On the other hand, 77 percent of the Māori, but only 58 percent of the Asian abortion patients were aged under 30 years.

Table, Distribution of Abortions by Age and Ethnicity.

The relative incidence of abortion varies significantly with age and follows a ‘V’ shape curve (Table 3). The national abortion ratios in Table 3 (last column) indicate that in just under half the cases a pregnant teenager is likely to seek an abortion. For those aged 20-24 years the corresponding figure is one in three. High ratios at younger reproductive ages are not unexpected. As Bankole et al (1999) point out: “ Where women have opportunities for education, employment and career development, younger and unmarried women are the most likely to want to postpone marriage or childbearing, and to obtain an abortion when a pregnancy occurs”. Significantly, the ratio drops to just one in eight for the 30-34 age group. Thereafter, the propensity to abort a pregnancy increases with age.

Table, Abortion Rates by Age and Ethnicity.

Ethnic differentials in the incidence of abortion are quite marked. In 2000, about 75 percent or three in every four pregnant Asian teenagers aborted their pregnancy. The European teenagers had the next highest figure (59 percent), while Māori teenagers had the lowest ratio (32 percent).

Similarly, about half of the pregnant Asian women aged 20-24 years, but only one-quarter of their Māori counterparts terminated their pregnancies. A majority decided to continue the pregnancy. Beyond age 25 years, European women have the lowest abortion ratios. These differences may be attributed partly to the differences in childbearing patterns and in the use of contraceptives. European women have a later childbearing pattern and a pregnancy in their 30s reflects a deliberate decision to have a child and is less likely to be an unwanted pregnancy.

Abortions by marital status

The majority of abortions in New Zealand in any year are to women who have never been married. During the 1990s, they made up a slowly increasing proportion of all women having abortions - up from 59 percent in 1990 to 60 percent in 2000 (Statistics New Zealand, 2001). There has been a parallel, but more pronounced rise among women living in de facto unions - up from 7 percent in 1990 to 13 percent in 2000, and a corresponding fall for married and separated women, down 5 percentage points to 25 percent in 2000. These changes are associated partly with broad shifts in childbearing and family formation norms.

Not all ethnic groups, however, follow the national pattern (Figure 2). About 50 percent of Asian women obtaining an abortion in 2000 were married, compared with only 8 percent of Māori women and 14 percent of European women. The higher proportion of married women among Asian abortion patients partly reflects the marital status distribution of their population. At the 1996 Census, nearly three in five Asian women were legally married compared with only one in three Māori women.

Graph, Distribution of Abortions by Marital Status and Ethnicity.

Abortions performed on women living in de facto unions also varies between the ethnic groups. Among Māori women, those living in de facto unions accounted for 21 percent of all abortions in 2000, compared with 13 percent for European women, less than 8 percent for Pacific women and fewer than five percent for Asian women. Once again, a large part of this variation may be explained by women’s living arrangements. At the 1996 Census, 16 percent of Māori women were living in a de facto union compared with only 4 percent of the Asian women. The proportion of European and Pacific women living in de facto unions were almost identical at 9 percent.

Abortions by previous children

A growing proportion of women having abortions have already had one or more live births (Statistics New Zealand, 2001). About 20 percent of women having an abortion in 2000 had one previous live birth, compared with 17 percent in 1990. Similarly, about 34 percent of women have had two or more live births, compared with 32 percent in 1990. By contrast, about 47 percent of the women who had an abortion in 2000 had no children compared with 51 percent in 1990.

An ethnic analysis (Figure 3) shows that 54 percent of European women obtaining an abortion in 2000, had no previous children, compared with 38 and 29 percent respectively of the Pacific and Māori women.

Graph, Distribution of Abortions by Previous Number of Children and Ethnicity.

On the other hand, Māori and Pacific pregnant women are more likely to seek an abortion if they had had at least two previous live births. In 2000, 50 percent of Māori abortion patients had had two or more previous live births - nearly twice the figure for European women (27 percent). The corresponding figure for Pacific women was 44 percent. The Asian figure (30 percent) was closer to the European level.

Because Māori and Pacific women tend to have larger families than their European and Asian counterparts, one would expect these two groups to have a higher representation among abortion patients who had already had at least two children. In fact, a significant proportion (14 percent) of Māori and Pacific women having an abortion in 2000 had had 4 or more children. For Asian and European women the corresponding percentage was fairly low - about 3 percent.

Abortions by previous abortions

Repeat abortions are on the rise. About one in three abortions in 2000 were performed on women who had had at least one previous abortion, up from one in five in 1990 (Statistics New Zealand, 2001). One in 10 women having an abortion in 2000 had had two or more previous abortions, over twice the figure recorded in 1990.

Differences between the ethnic groups in terms of the number of previous abortions are relatively small. For women in each of the ethnic groups, over 60 percent of abortions were first time abortions, the figures ranging from 63 percent for Pacific women to 68 percent for European women. Similarly, about 1 in 4 abortions were performed on women who had had one previous abortion. Because European and Asian women recorded slightly higher proportions amongst women having no previous abortions or only one previous abortion, their proportions amongst the two or more previous abortions were slightly lower than for the Māori and Pacific women.


Unlike in many OECD countries, the abortion rate in New Zealand is continuing its upward trend, despite a wide availability of modern contraceptive methods. Repeat abortions in particular are on the rise. This brief study found significant variations in the abortion experiences of various ethnic groups. Asian women, especially those aged below 25 years, show a relatively high propensity to abort a pregnancy. Further research is needed to identify the underlying reasons for terminating unintended pregnancies.


Bankole, A. Singh, S. and Haas, T. (1999), Characteristics of Women Who Obtain Induced Abortion: A Worldwide Review, 25(2): 68-77.

Henshaw, S., Singh, S. and Haas, T. (1999), Recent Trends in Abortion Rates Worldwide, International Family Planning Perspective, 25(1): 44-48.

Khawaja, M.A. and Parr, A.C.J. (1993), Trends and Differentials in Induced Abortions in New Zealand, 1985-1992, in Ethnicity and Gender, Population Trends and Policy Challenges in the 1990s - Proceedings of a Conference, Wellington, July 1993. Wellington.

Khawaja, M.A. (1997), Growing Ethnic Diversity in New Zealand and its Implications for Measuring Fertility and Mortality. Paper read at the Australian Population Association Conference.

Petersen, W. (1997), Ethnicity Counts. Transaction Publishers. New Jersey.

Sceats, J.E. (1986), Induced Abortion in New Zealand 1976-1983. Government Printer. Wellington.

Statistics New Zealand (2001), Demographic Trends 2000. Wellington.

Statistics New Zealand (2001), Abortions: Year ended December 2000, Hot Off The Press, Wellington.

Statistics New Zealand (2001), Unpublished tabulations.

Wood, R., Bitting, B. and Dunnell, K. (1997), Trends in Conceptions Before and After the 1995 Pill Scare, Population Trends, 89: 5-12.


1 This paper was prepared by Mansoor Khawaja with assistance from Irene Tang and Kirsten Nissen.

Printable version

The downloadable file is in Adobe Acrobat format. If you do not have the Adobe Acrobat Reader you may download the reader to view or print the contents of this file.

  • Share this page to Facebook
  • Share this page to Twitter
  • Share this page to Google+
  • Share this page to Facebook
  • Share this page to Twitter
  • Share this page to Google+