Migration and fertility in Pakistan.

Kiani, M. Framurz K.

The phenomena of migration and fertility has not drawn much attention in Pakistan. A few studies based on census data, the National Impact Survey 1968-69 and the Population Growth Survey 1968-71 showed higher crude birth rates for rural areas than for urban areas. However, recent studies showed higher urban fertility in Pakistan.

In view of the increasing rate of rural migration to the urban areas it is important to understand the contribution of migration to population growth. The main objective of the present study is to explore differentials in fertility between migrants and non-migrants along with the socio-economic and proximate determinants and to investigate whether these show statistically significant variation in fertility between migrants and non-migrants.

The primary source of data for this study is the Population Labour Force and Migration Survey (PLM) 1979-80. Migration and fertility questionnaires were merged to determine the migration status of husband and wife to relate it to the fertility of women. A total of 335 female migrants in urban areas and 480 in rural areas were identified.

Some basic characteristics of migrant women, covered by the survey, showed that in the sample population migrant females were younger in age with higher age at marriage and shorter duration of marriage. Migrant females were relatively better educated while their husbands have even better education than non-migrants. The differentials by mean children ever born suggest that urban migrant females have lower mean children ever born compared to non-migrants.

However, in the rural areas the differentials were not substantial. This shows that non-migrant females are more fertile while migrant females tend to retard fertility and prefer a smaller family size. Migrant females showed a slight increase in age at marriage compared to non-migrants both in urban and rural areas. The younger cohorts of females aged less than 25 were married at an early age compared with older cohorts above age 35. Since the fertility module comprises only of ever married females, therefore, younger females show less increase in age at marriage. The duration of marriage also showed that females who had been married less than 20 years have fewer children ever born in the urban areas as compared to females who had been married longer than 20 years. The rural females have not shown substantial differences in fertility due to duration of marriage. The lower levels of fertility in urban female migrants is due to the disruptions caused by migration. In rural areas short distance migration and less educated females may be attributed for lesser effect of disruptions on fertility.

In terms of education, migrant females with 1-8 years of schooling showed substantially less mean children ever born compared to non-migrants both in urban and rural areas. Less children ever born for migrants may be due to their changes in aspirations or the sense of self-efficiency on the basis of which couples decide to migrate. Thus, education affects the attitude and contributes to lower fertility. Husband's economic status being a more meaningful factor when related to the wife's economic status, showed that migrants in each occupation were having less mean children ever born than non-migrants both in urban and rural areas. Migrants engaged both in white and blue collar occupations demonstrate substantially smaller family size than non-migrants. The lower level of fertility among white and blue collar occupations may be due to more educated people engaged in these occupations and these have greater aspirations for goal achievements and thus represent more mobile segments of the population.

The proximate determinants included in the study were duration of breastfeeding, attained and desired family size, knowledge of family planning and contraceptive use.

Duration of breast-feeding showed that urban migrant females who breast-fed more than 24 months demonstrate higher proportions in the 25 and above age groups. In the rural areas migrant females who also breast-fed for more than 24 months show similar patterns as the urban migrant females. The differentials in fertility suggest that migrant females both in urban and rural areas who tend to prolong the duration of breast-feeding includes either selective females or that the migration process itself affects their attitude or behaviour to prolong their durations of breast-feeding. The attained and desired family size vary between migrants and non-migrants. The younger migrants in both urban and rural areas showed desire for smaller family size than non-migrants. Knowledge of family planning methods among migrant females is greater than non-migrants. This suggests that migrant females were more exposed to family planning methods and were better informed. The use of contraceptive methods was also analysed. Even though the number of migrant users of contraceptive methods were very small both in urban and rural areas, the overall use of contraceptive methods was higher for migrants.

To assess whether any variable has influenced migrant fertility more strongly, multiple classification analysis was performed in two parts for urban and rural areas separately. The results in part 1 of the analysis showed that education, migration status and province of destination emerged as major determinants of fertility in urban areas while province of destination, migration status and partners occupations showed significant effect on fertility for rural areas. The above determinants showed a negative impact on fertility both in urban and rural areas while province of destination, migration status and partners occupations showed significant effect on fertility for rural areas.

In part two of the analysis, the relative role of some determinants of fertility have been examined for migrants and non-migrants separately. Length of breastfeeding showed a significant negative association with children ever born. This shows that migrant females with two or more years of breast-feeding had lowest fertility even though the number of cases were very small. The other determinants did not show any significant association with fertility for migrants. However, differences across categories showed considerable variation. The overall analysis suggests that urban migrants who were younger, better educated, and come from distant areas have more of an effect on fertility in urban areas, but rural migrants have a lesser impact on fertility.

In conclusion, the above findings do indicate that migrants show a greater desire for controlling fertility than non-migrants. Since this study is based on a relatively small sample, the conclusions may be interpreted with care.

M. FRAMURZ K. KIANI, The author is Research Demographer at the Pakistan Institute of Development Economics, Islamabad. This is an abridged version of the paper presented in the Fourth Annual General Meeting of the PSDE.