Chapter 4

FERTILITY

 

One of the major objectives of the PRHFPS is to estimate recent fertility and its differentials. Information on fertility will help to determine the impact of changes in the use of family planning and changes in the other proximate determinants of fertility. The fertility estimates presented in this chapter are based on the last three births of ever-married women 15-49 years old who were interviewed in the PRHFPS. The birth history of last three live births also obtained information on the single, twin and multiple birth status, sex, date of birth and survival status of each child. Respondents were however, first asked to report the aggregate number of sons and daughters they had ever given birth to in their lifetime referred to as ‘Children Ever Born’. To encourage complete reporting, women were asked separately about children still living at home, those living elsewhere and children who had died.

4.1       Children Ever Born and Surviving

 

Tables 4.1 and 4.2 show the average number of children ever born to all women aged 15-49 years and to currently married women aged 15-49 years respectively. For all women, mean number of children ever born is 2.6 (Table 4.1) and it increases smoothly from 0.10 to 6.74 as the age group increases from 15-19 to 45-49. By her early 20s, the average woman has given birth to nearly one child. Women in their early 30s have an average of 4.3 births, and this figure rises to 6.4 births for women aged 40-44 years. The average mean number of children ever born for currently married women is 4.1 (Table 4.2). The estimates for currently married women are similar to those for all women at older ages. As expected, at younger ages the currently married women record considerably more births than all women.

Table 4.1

 

Mean Number of Children Ever Born to All Women, by Age and Background Characteristics

 

 

Table 4.2

 

Mean Number of Children Ever Born to Currently Married Women by Age

and Background Characteristics

 

The figures show that women in rural areas, and women with no education tend to have more children than those in other categories. There is likelihood of considerable overlap in these groups, since it is the women in rural areas who have the least education. Provincial differences in CEB are somewhat eratic, which urges caution in drawing conclusions.

 

Figure 4.1 shows mean CEBs and mean numbers of surviving children by women’s age. At age 30, the average Pakistani woman has over two surviving children. At age 40, this number is about five children. Clearly the country is still a long way from achieving a stable population in which women, on average have only two surviving children. Figure 4.2 displays the pattern by Urban-Rural. The higher cumulative fertility is visible in rural areas, compared to urban in all age groups.

Figure 4.1

 

Mean Number of Children Ever Born and Surviving, Among All Women, by Age

 

 

Figure 4.2

 

Mean Number of Children Ever Born and Surviving, Among All Women by Age and Residence

 

Urban                                                                    Rural

 

4.2       Trends in Mean Number of Children Ever Born and Surviving

Table 4.3 gives trends in average CEBs for all women since 1975. For the 1994-95 PCPS data set the comparable figures for all women were not available as only households with currently married women were selected. The fall of average CEBs for the young age groups clearly indicate the effect on fertility of the later age at marriage. The main point of reference is the 1975 PFS, 25 years ago. The average CEBs have fallen in each age group in the PRHFPS, 2000-01.

Table 4.3

 

Trends in Mean Number of Children Ever Born to All Women by Age

 

Age	PFS 1975	PDHS 1990-91	PFFPS 1996-97	PRHFPS 2000-01
15-19	0.23	0.16	0.11	0.10
20-24	1.48	0.95	1.02	0.88
25-29	3.09	2.61	2.75	2.41
30-34	4.80	4.29	4.62	4.29
35-39	5.88	5.49	5.58	5.33
40-44	6.88	6.26	6.43	6.40
45-49	6.83	6.42	7.17	6.74
All Ages	4.3	2.6	2.8	2.6

Table 4.4 shows trends in the mean numbers of children ever born to currently married women, as reported by various surveys over the past twenty-five years.

Interpretation of these trends should be seen in the light of differences in survey methods adopted by the successive surveys. For instance, the PCPS surveys did not use a complete birth listing to establish numbers of children, and the PDHS sample appears to have a problem with the omission of children. If these surveys are ignored, the trend over time indicates modest declines between ages 20 and 40. Above this age, there is not much evidence of decline. It appears that fertility decline is relatively recent and has affected younger but not older women, whose peak childbearing years are further back in the past and thus their cumulative fertility in successive surveys has not shown much decline. However, there does not appear to be any delay in childbearing after marriage among women in age group 15-19 and to some extent, 20-24 years. Any overall fertility decline can thus mainly be attributed to the increase in age at marriage.

Table 4.4

 

Trends in Mean Number of Children Ever Born to Currently Married Women Aged 15-49, by Age

 

 

The number of surviving children is the net outcome of fertility and child mortality. Table 4.5 shows trends in mean numbers of surviving children among currently married women. The average numbers of surviving children are somewhat higher in the 1996-97 and 2000-01 surveys than in the previous surveys. It appears likely that the modest declines in marital fertility have been offset by improvements in child survival.

Table 4.5

 

Trends in Mean Number of Living Children to Currently Married Women Aged 15-49, by Age

 

4.3       Age-Specific and Total Fertility Rates

Fertility rates are derived from the birth history for the last three live births including multiple births, which was administered to ever-married women. In order to convert rates based on the reporting of ever-married women to conventional ‘all-women’ fertility rates, never-married women are included in the denominator on the assumption that they have experienced no births.

Table 4.6a & b shows the age-specific fertility rates (ASFR) and total fertility rate (TFR) calculated for four-year periods prior to the survey by province and residence. Four-year periods has been taken for calculation of these rates to reduce sampling error and to minimise the displacement of births.

A total fertility rate at national level has been estimated at 4.8. Fertility peaks at age-group 25-29 (258 births per 1000 women). Across provinces total fertility rate for women aged 15-49 is lowest in Punjab and Sindh (4.7), followed in ascending order by NWFP (5.1) and Balochistan (5.4). As expected TFR is lowest in major urban (3.44) and highest is rural area (5.4). In other urban area TFR is 4.02.

 

Table 4.6a

 

Age-Specific Fertility Rates, Total Fertility Rates for the Four-Year Period

Preceding the Survey, by Province

 

 

Table 4.6b

 

Age-Specific Fertility Rates, Total Fertility Rates, for the four-year Period

Preceding the Survey, by Residence

 

 

4.4       Comparison of PRHFPS Fertility Trend Estimates with Other Sources

The data in Table 4.7 suggest that fertility rates have dropped significantly in the last four years. However, because of possibility of omissions and shifting, the results need to be interpreted with caution. Some births are likely to have been displaced across the 4-year boundary as a result of exaggeration of children’s ages by the respondents. Omission of recent births is also a common feature of such surveys. This appears evidenced by the sex ratio of 1.00 in the age group 0–4 year olds (as shown earlier in Table 2.1). A more detailed analysis of these factors and appropriate methods of adjustment will appear in later detailed report.

The possible displacement of births away from the survey and the likely omission of recent births both have the effect of biasing downwards the estimate of the TFR for the most recent period. Because these data problems have probably affected the estimates of levels from previous surveys more than in the 1996-97 PFFPS or 2000-01 PRHFPS, the estimate calculated for the PRHFP survey is certainly a strong evidence of a continued decline in fertility.

All previous fertility surveys using a birth history have shown the same shifting of births. They all give the impression of a rapid fertility decline immediately prior to the survey which requires careful interpretation of estimates. The true rate of decline is therefore likely to be much smoother than suggested by a superficial examination of the trends in any one survey. The trend is better evaluated looking across surveys, concentrating on results obtained for comparable periods.

Table 4.7

 

Trends in Age Specific and Total Fertility Rates

 

 

 

In general, the rates for each age group, and the total fertility rates follow the same pattern between the surveys, with a common indication of a steady decline in fertility.

Overall a consistent downwards trend is apparent (Table 4.7). The rate of decline in TFR shown is probably generally correct. From the PFS in 1975 to the PRHFPS 2000/01 survey the TFR has fallen by one and half child, equivalent to about 24 percent. Overall it seems safe to conclude that the slow decline in fertility in Pakistan is continuing, and that the current TFR is close to 4.8.

 

 

 

4.5       Differentials in Recent Total and Marital Fertility and Mean Children Ever Born

Table 4.8 shows the preliminary estimates of total fertility rates, total marital fertility rates and children ever born of women aged 40-49 years by province, residence and education. The children ever born shown in table 4.8 indicates average parity of women aged 40-49 years. The total marital fertility rates represent the sum of age-specific rates for currently married women only, from age 15 to 49 years. The levels of TFR may suffer from the biases described previously, but the differentials are likely to be reliable.

As expected, Balochistan has the highest rate, and Punjab and Sindh the lowest. Urban families have at least one and half child less than rural ones, and there is a marked negative effect of education on fertility. Educated women have lower fertility compared to those who have no education.

Table 4.8

 

Differentials in Fertility for 1997-2000