Productive and reproductive choices: report of a pilot survey of Urban working women in Karachi.

Kazi, Shahnaz ; Sathar, Zeba A.

INTRODUCTION

The primary objective of the survey was to collect information on women's productive and reproductive choices within the broader context of the conditions in their household. Since women bear the main responsibility of domestic work and child care, their employment outside the house is closely interrelated with household decisions pertaining to fertility, division of domestic duties, etc. Hence, to clearly understand the linkages between work status and domestic roles of women, it is necessary to collect data on both demographic and socio-economic aspects of the household.

Hitherto, data on fertility and labour force participation have been collected separately. (1) The Population, Labour Force and Migration Survey (1979) is an exception; the purpose of the multi-schedule survey was to combine together the separate pieces of information obtained on income and expenditure, labour force, migration, and fertility. This survey represents a big step in the right direction; the philosophy behind it was that reproductive behaviour cannot be studied in isolation. However, the labour force and income and expenditure modules were the same as those generally used for the annual surveys conducted by the Central Statistical Organization and do not include any extra quality information on female work patterns, women's contribution to household income, etc.

The pilot survey undertaken in Karachi is a further step in the direction of gaining combined information on the household's productive and reproductive choices, but with an emphasis on women's role in decision making and the object of evaluating the true impact that employment has on their status and fertility. The questionnaire collected information from the respondents on their educational and socio-economic background, work history, details of current occupation, marital history, fertility, and their aspirations and expectations with respect to their role in household decision making, ideal family size, education of children, old age support, etc. At the household level, data will be collected on the material conditions of the households, earnings, education and occupation of other family members, and division of domestic responsibilities within the household.

The discussion begins with a description of the sampling procedure used and a preliminary sketch of some background characteristics of the respondents. The analysis that follows is divided into two sections. The first section addresses the following important questions related to women's productive activities: What motivates women to enter the labour force? How essential is their contribution to the economic survival of the household? What characteristics distinguish women who are sporadic entrants into the labour force from those with a long-term commitment to work? How do women resolve the conflict between their work and domestic responsibilities? The second section is concerned with the reproductive choices of working women. The survey data are used to explore the relationship between women's work and such demographic variables as fertility, age at marriage, and contraception adoption.

SAMPLING PROCEDURE

One of the major obstacles in selecting a systematic sample of working women is lack of a sampling frame. Though data on labour force participation are collected periodically by the Central Statistical Office through Labour Force Surveys, only 5-10 percent of women are reported as working in the households selected. This does not then give us a complete sampling frame of all working women. The census could, in theory, provide a listing of households whose women work, but in practice this task would prove to be tedious and not very worth while, as the 1981 Census reports the lowest ever participation rate (3 percent) for women. The problem in enumerating female workers is also compounded by the lack of admission, particularly by other household members, that women in the family work outside their homes. As it is, the nature of women's work is such that it is likely to go unrecorded as labour force participation because of the informal nature of most of the jobs that women take up, e.g. stitching at home, weaving, making paper bags, midwifery, vending food items, domestic service, etc.

Thus, anyone who embarks on a full-fledged survey of working women may find it very difficult to sample women working in the informal sector. Snowball technique, or sampling of areas which contain concentrations of women working in informal occupations, would be based on the idea of locating these women either in their workplaces (such as markets etc.) or at their homes. On the other hand, it would be easier to select a systematic sample of working women in the formal sectors through procurement of lists of personnel from factories, government offices etc., and thus put together a comprehensive sampling frame.

Our sample of the pilot survey was selected along these lines, but unsystematically. Our intention was not only to sample but to identify commonly found occupations amongst women to help us further in our complete survey later. The aim was to select more than 60 percent of women from low-paid occupations and the remainder from professional groups and other higher-income occupations.

BACKGROUND CHARACTERISTICS OF THE RESPONDENTS

The sample of 110 working women included respondents between the ages of 19 and 50. Out of these, 93 were currently married, six were separated or divorced, and eleven were widowed. Most of the ethnic groups living in Karachi, including Sindhis, Punjabis, Memons, Baluchis, Pathans and migrants from India, were represented in the sample.

Information on the education, earnings and occupation of the women is presented in Tables 1 and 2. The results indicate that the sample comprises three distinct socio-economic groups. On the one hand, there are highly educated women employed in high-status jobs, working as doctors, teachers, bankers and administrative personnel. Their mean earnings are much above the average, ranging from Rs 2145 for teachers to as high as Rs 5816 for doctors.

On the other end of the spectrum are women who can be loosely termed as working in the informal sector. They are mostly uneducated and work as low-status workers such as casual labourers, home workers, domestic servants, sweepers, vendors and those in miscellaneous occupations. The meagre incomes in this group range from an average of Rs 380 per month for potters and Rs 447 for home workers to Rs 735 for sweepers. Working conditions in the informal sector are not regulated by any contract and the workers do not have access to the benefits of formal employment such as fixed wages, security of employment, etc. The level of earnings is generally below that in the formal sector. However, since entry into the formal sector is restricted by educational and other requirements, the poor and uneducated females have no alternative but to take up these low-income activities. (2)

In between these two groups is the subgroup of women belonging to lower middle class backgrounds, who also work in the formal sector but in lower-status occupations as compared with those of the professional group. These women include nurses, telephone operators, lady health visitors and factory workers. Their level of education is below that of the first group but most of them have at least completed their matriculation. Average incomes for this category of workers range from Rs 1137 for factory employees to Rs 1882 for telephone operators.

There are wide variations in the standards of living of these occupation groups. Housing conditions are most inadequate for the category of miscellaneous workers: in the majority of their cases, house construction is semi-pucca, water has to be fetched from outside, and wood is the main fuel. However, electricity is available in nearly 60 percent of these households, while more than 77 percent also have a latrine inside the house. They possess few consumer goods, although entertainment items, such as television and radio, seem to have high priority, reflected in the fact that nearly 50 percent of these households possess a black-and-white TV set. Sewing machines and electric irons are other items which are owned by some households in this group. However, appliances like pressure cookers and food mixers, which facilitate cooking, are very rare.

Empirical results in this section indicate that the occupational statuses of women reflect differences in education, income and standard of living of households, thus corresponding closely to differences in socioeconomic status. Work options, reasons for work, and other household dynamics related to economic activities are likely to vary markedly across women from different socio-economic groups. The subsequent analysis will explicitly address these class differences in work choices, relying mainly on occupational classification as an index of socioeconomic position.

WORKING WOMEN: PRODUCTIVE CHOICES

Reasons for Joining the Labour Force

More than 70 percent of the respondents gave financial pressure as a reason for their entering the labour force. A large proportion of these women were forced to earn a living due to a sudden deterioration in their economic position, brought about by such factors as loss or illness of husband, separation, or divorce. Work for these women is the outcome of economic need, which leaves them with no alternative but to take up whatever job they can find to support their families. For them it is not a choice between career and household responsibilities.

Women for whom the important reason for work was pursuit of a career or job satisfaction were highly educated and employed in high-status, renumerative jobs. Only 20 out of the sample of 110 women worked for non-economic reasons. Among these, 75 percent had at least a B.A. degree, 65 percent were earning Rs 2500 or more a month, and 85 percent were employed as professionals, such as doctors, teachers, etc.

Not surprisingly, within the uneducated subgroup of women employed in dead-end, tedious jobs on low wages, there were few instances of women who found any fulfilment in their work. Thus, out of a total of 64 women working as factory workers, domestic servants, and other miscellaneous workers in the informal sector, there were only two women working out of choice.

Work before Marriage

The most important employment spell is that of work before marriage. It seems to be an indicator of a long-term commitment to work and a career orientation. Women who worked before marriage comprised nearly 51 percent of the sample. There was a very strong positive association between work before marriage and level of education and occupation. A much larger proportion of educated women in professional jobs worked before marriage as compared with their uneducated or less educated counterparts. Nearly 90 percent of the professionals (doctors, bankers, etc.) worked before marriage, while in the category of nurses, lady health visitors and telephone operators 77.8 percent of the women worked before marriage. However, in the subgroup of those in lower-income occupations, such as domestic servants, miscellaneous workers and factory workers, the majority of women started work only after their marriage.

The career orientation of women who worked before marriage is further evident from the fact that 80.4 percent of these women continued to work after marriage without any break. The common view that women leave the labour force temporarily in the childbearing phase does not seem to apply to high-level professional women in the sample. Among the category of doctors, bankers, teachers, etc., 78.6 percent continued to work without any interruption after marriage or after the birth of their children, while in the cases of lower-grade professionals, only 66.7 percent of the women had an uninterrupted work history. Professionals, especially in the former group, have considerable prospects for upward mobility. In these cases, gaps in work experience would have an adverse effect on the pace of progress and would prove to be a deterrent to the workers' ability to move on to better positions in their careers.

The life-cycle employment pattern is entirely different for women in the static, low-income occupations of domestic servants, miscellaneous workers in the informal sector, and factory workers. The majority of women in this group do not work before marriage, and they do not take up a job till at least six years after marriage, or after the birth of at least one or two children. Thus, whereas the average age at which women in professional jobs start work is 20.9 years, in the cases of factory workers and other miscellaneous workers in the informal sector, the corresponding ages are 24.4 years and 23 years respectively. These women do not have any prior commitment to their work role but are forced by economic circumstances to seek employment in order to support their growing families. Since they have little or no education and no work experience, they have to take up low-income activities in the informal sector. The luckier ones manage to get employment in large factories where wages are much higher, although working hours are also longer. Contrary to the general view that women withdraw from the labour force to take care of small children, an increase in family size compels women from the lower strata to seek employment to satisfy the minimum needs of their children. The late entry into the labour market of women employed in the informal sector has also been noted for India [3].

Contribution of Working Women to Household Income

The working women's contribution to household income is markedly higher in the poorer strata. In the lowest-income households, women's earnings, on average, constitute more than half of total household income, whereas for households with total income of Rs 7000 or more women's average contribution to total income falls to 29 percent.

The importance of women's contribution to family income in households of various income groups is brought out in Table 3. Deduction of women's earnings has the greatest impact on the top and bottom ends of the income scale. When household income is estimated net of women's wages, the percentage of households in the lowest income bracket increases dramatically from 24.5 to more than 46 of the total. In six cases women are the sole earners. At the other end of the scale, the percentage of households in the highest income bracket falls from 23.6 to 13.6 of the total. In these cases, women's earnings, although not essential for survival, enable the households to maintain a higher standard of living.

Child Care and Division of Other Household Tasks

The extent to which women's participation in the labour force conflicts with their household duties and child care is a question of great importance. The view that female employment is determined by her responsibility to domestic work and child care is greatly influenced by the experience of Western industrialized nations, where reduction in fertility and availability of time-saving housework technology is believed to have led to an increase in women's employment. Work activities need not interfere with domestic responsibilities if there are other relatives or servants available to look after children and do the housework. This is rare in developed countries but is quite common in Third World nations. Evidence from less developed countries further suggests that, for most women, employment does not reflect a choice between domestic responsibilities and work, but is the outcome of economic necessity which compels women to work to supplement their family income [6; 1].

The results of the survey indicated that in the large majority of cases women's decision to take up a job was the result of financial pressure and did not involve any element of choice. The respondents were asked whether they would give up their jobs if there was an increase in their domestic responsibilities. In only ten cases the answer was in the affirmative, while the large majority stated that they could not afford to give up work and would simply have to make some adjustments on the domestic front.

The most frequent child care surrogates were the respondent's eldest daughter, mother and mother-in-law, in that order of importance. The mother of the respondent looked after the children, not only in cases where she was living with the respondent, but also when she lived separately. In 11 out of the 23 cases in which the respondent's mother was entrusted with child care, she was living separately and the children were dropped at her house on the respondent's way to work. There were only two instances in which the children were cared for by hired help. Even where households can afford to employ domestic servants to look after children, they are left under the supervision of some female relative.

With respect to other domestic tasks, such as washing clothes, cleaning and cooking, the working women's burden of domestic duties depended largely on the presence of other female members in the house. Male members of the household very rarely participated in domestic work and their contribution was limited to shopping. Full-time domestic servants were employed by very high income group only. Thus, 56.3 percent of the females who did all the housework themselves were living in households with no other adult females (of 15 years or more), while 91.2 percent of the females who had no domestic responsibilities belonged to households with at least one other adult female member.

In the households where the respondent had no domestic responsibilities, the chores of cooking, washing and cleaning were largely taken care of by other female relatives or servants, depending on the income level of the respondent. Whereas servants were most frequently in charge of various household tasks of respondents in the highest income bracket, among the lower income groups it was the eldest daughter who in most cases relieved the mother of her domestic responsibilities. In poorer households, the eldest daughter is an important source of support for the mother. She takes on housework and child care, and in a large number of cases this is done at the expense of her education. While the younger children continue to attend school, the eldest daughter drops out and devotes herself entirely to domestic responsibilities. In 26.3 percent of the households in the sample, the eldest daughter was not attending school or college but was in charge of domestic tasks. The age of this group of girls, whose only activity was housework, ranged from 11 years to 21 years.

WORKING WOMEN: REPRODUCTIVE CHOICES

Fertility Differentials

As already pointed out, our primary aim, in conducting this survey, was to study the association between women's work and their reproductive behaviour. Previous studies have ignored the impact of work participation on fertility in the context of Pakistan [10]. However, the different reasons why women may be working and other behavioural factors and not the mere fact that women work may be more relevant in terms of fertility differentials [4]. We need to distinguish between those women who work out of choice and those who do so out of necessity. Women of the first type tend to work in higher-status occupations continuously or with few breaks and are quite likely to have started working before marriage and having children later, probably out of some consideration for their jobs. Women of the second type are more or less pushed into working out of necessity and for them it is a negative rather than a positive choice. These women may have entered their lower-status, lower-paid occupations after death or sickness of the major earner, or widowhood or separation and/or when the burden of supporting many children could not be borne by just one earner in the family. These women, then, are unlikely to have an altered attitude towards reproduction as a result of their getting paid employment.

Evidence from the previous section indicates that professional-clerical women and the group of lower middle class professionals are more likely to have a "career orientation", to remain in continuous employment, and to have engaged in pre-nuptial employment. It is interesting, then, to note that women who work in professional and clerical occupations have much lower fertility, particularly as compared with women who work as domestic servants etc. and also as compared with artisans such as potters, tailors, etc. Similarly, there is a negative association between the income a woman earns and her fertility behaviour. The mean parity of women who earn less than Rs 500 per month is 5.8 children, while that of women who earn more than Rs 2500 is 2.3 children. Fertility differentials by education are even stronger than those by respondent's income. Uneducated women, on average, had a mean parity of 6.3 children as compared with that of 2.3 children for women with at least a B.A. degree.

Another interesting source of fertility differentials was ethnic affiliation of the women. It is particularly important because despite living in Karachi, the largest metropolis of Pakistan and a melting pot of many different groups, the ethnic groups still adhere to their own values. Differentials might, to some extent, be attributable to differences in economic classes of the ethnic groups. The Agha Khanis, Biharis and Bengalis had the lowest fertility, followed by the Urdu-speaking community from India. Amongst the groups sampled, the Kutchis, who are immigrants from the Rann of Kutch, have the highest fertility.

Age at Marriage

Recently, Pakistan has recorded considerable rises in the average ages at marriage, particularly for women among whom the proportions marrying at ages 15-19 years have fallen sharply [5]. Age at marriage not only marks the entry into a sexual union and the beginning of exposure to childbearing but may also be an important gauge of women's status, since the older the woman is when she marries, the greater the chances she has of attending school, taking up pre-marital employment, and having a more equal relationship with her husband. It is, therefore, important to look at the differentials in average ages at marriage found in the survey as they have important implications for measurement of women's status and for fertility.

In the survey, generally speaking, the economic status and the likely measures of women's status are related positively with ages at marriage. The association of age at marriage with educational level is strongly positive: the average age at marriage of women who are uneducated is 15.3 years, while women with less than 10 years of schooling marry at an average age of 18.2 years, and those with 10-12 years schooling marry, on average, at 21 years. Not surprisingly, the highest age at marriage is found in women who are in professional-clerical jobs, followed by professionals of slightly lower economic status, e.g. nurses, telephone operators, etc., while domestic servants and artisans such as potters, home tailors, etc., marry at ages 16.1 and 15.3, respectively. Factory workers have an intermediate, though generally low, age at marriage of 17.7 years.

Across ethnic groups, the differentials were more or less like those found in fertility: the Agha Khani--Bihari--Bengali group married last while the Kutchi women married earliest. Once more, spousal age difference was related negatively to age at marriage: women who marry men more than 20 years older than themselves do so, on average, at age 15, whereas those who marry men closer in age to themselves marry over 6 years later at 21.5 years.

Another interesting finding was that the average age of those who marry a cousin or someone in the biradari is 18.2 years while for those who marry an "outsider" (a non-relative basically) it is almost 21 years. This reflects the pattern that cousin marriages are greatly preferred by families, and if a suitable cousin or relative can be found, marriage occurs earlier than in the cases in which the search for a suitable partner extends beyond the biradari. However, one likely reason for the choice of a non-relative partner may be that the girl is educated, works before marriage, and has a broader horizon from which to choose her marriage partner.

Ideal Family Size

Although many objections have been raised against the utility of asking women how many children they would like to have, the answers convey at least a "normative" response. From answers on ideal family size and ideal family composition, we can at least deduce whether large families are preferred to small ones and whether sons are preferred to daughters in the survey sample.

The majority of women consider an ideal family size to be between 3 and 5 children. Only 12.7 percent of them want 0-2 children and 4.5 percent want 6 or more children. However, 10 percent of women believe that the number of their children is dependent on the will of God. A generally inverse association is found between ideal family size and education, income of respondents and total household incomes. The desired family size is also inversely related to the average number of dead children. Those who believe that the number of their children depends on the will of God have lost 2.3 children, while those who believe in small families have lost one-tenth of that average.

The degree of preference for male children in a society is often a pronatalist influence. Sons are generally highly desirable in Pakistan. This is, presumably, because they are likely to support their parents in old age, whereas daughters in a patrilineal system marry early and move to their husbands' homes [9]. Our sample shows the expected response: a larger number of sons is preferable to a large number of girls. Women who had a large number of daughters seemed weighed down by this fact and were desperate to have male offsprings.

In this light it is indeed ironic that when asked who would look after them in their old age, more women expected daughters (41 percent) than those who expected sons (31 percent) to do so. The remainder gave no definite answer or said that children of both sexes would look after them. The more educated the woman, the greater the likelihood of her response in favour of daughters. This applied to the richest income groups as well. Amongst the ethnic groups, the parents belonging to Agha Khani, Bihari, Bengali and Indian-origin groups were most likely to expect daughters to be their support in old age, while the Kutchi, Punjabi, Sindhi, Baluchi and Pathan parents were more likely to expect sons to assume this responsibility. This finding shows some clear differentials in gender status and gender preferences amongst ethnic groups.

Contraceptive Adoption

In terms of demographic change, the more critical factor in a society is the proportion of married couples who use some form of fertility control. Overall, the pressing problem of high growth rates in Pakistan is largely attributable to persistently high levels of fertility and to the fact that only a minute proportion (6 percent)of women use contraception. Contraceptive-use levels are higher for urban residence and for educated women [8]. This is once more supported in our sample. Since it is an urban sample, we find that 37.3 percent of women use some form of contraception, while 47.3 percent do not use any fertility control, and 15.5 percent report that they are not exposed to childbearing as they are widowed, divorced, etc.

Contraceptive adoption is higher amongst those women who have the highest total household income (53.8 percent) as compared with women amongst the lowest income groups (25.9 percent); there is a similar differential in use between the most educated and the least educated women and between those women who themselves earn the highest income and those who earn the lowest income. As regards the ethnic groups, use levels are highest amongst the Urdu-speaking groups originating from India. The Punjabis have the lowest reported levels of contraceptive use.

Much has been written and said recently about the "unmet" need for contraception. The World Fertility Surveys, conducted in 42 developing countries, identified a considerable gap in many countries between those women who say they want no more children but are not using any form of contraception and those who do not want more children and are also using some form of contraception [11]. In conducting a similar exercise, we identify a similar gap in the survey population. Of the group who say they want no more children, 54.9 percent are using contraceptive methods, while 1.8 percent are unexposed, and 42 percent are using no contraception at all.

CONCLUSIONS

The results from the survey most certainly confirm that there are strong inter-linkages between socio-economic conditions of different households which influence both productive and reproductive choices of women belonging to them. Some common patterns found were that women working in higher-status occupations, with accompanying high levels of education and income levels, were more likely to marry later, to desire smaller families and to use means of contraception to limit their fertility. On the other end of the spectrum are the women, perhaps typical of the larger majority of working women in urban areas, who are forced to take up paid employment in low-status jobs, are likely to be uneducated and earn low incomes. Their contribution to household income is, probably, essential for their survival, especially when they are the major bread-earners (e.g. in cases of widowhood, separation, etc.). Their "production" choices are then based on stringent financial needs and working is not likely to be an experience leading to changes in their attitudes towards family size or use of contraception and subsequently to alterations in fertility-related behaviour.

Thus, women from higher socio-economic groups are more likely to have a "career orientation", to have worked before marriage and to take fewer breaks, whereas women from lower echelons are likely to work out of financial need and would prefer to give up working if their financial situation were to improve. The latter group is unlikely to feel liberated from the stereotyped image of females staying at home and looking after children--as in many cases amongst this group such a situation would, perhaps, be ideal.

In terms of the demographic transition theory, the two groups of women present ends of the spectrum from low to high fertility. The first group has adopted different attitudes towards childbearing, and, quite possibly, to marriage as well. They view children as a responsibility, who have to be put through school and nourished well--and since this is likely to be expensive, not only financially but also in terms of time, a small family-size is preferred. At the other end of the spectrum are women who believe that children are a blessing of God and that a large or even limitless number is desirable. For many of the poor, a large family is the main source of prestige and survival. In this regard, sons are much more desirable than daughters because they leave home at relatively early ages in a patrilineal society.

Comments on "Productive and Reproductive Choices: Report of a Pilot Survey of Urban Working Women in Karachi"

As stated by the authors, the objective of the paper was to get a better insight into women's motivations to enter the labour force, the importance of women's contribution to overall household income, women's commitment to work, and the way women resolve conflicts between their occupational and domestic responsibilities. The authors attempted to attain their objective through a purposive sample-survey of 110 working women in Karachi engaged in a variety of occupations. What the authors have presented is only a pilot study, and I am not sure whether the complete survey has been carried out as yet.

The study makes some important contributions to the existing literature on the subject. For example, it does show that for the majority of the women in their sample, work does not represent "a choice between a career and household responsibilities but is the outcome of economic need, where they have to take whatever job they find to support their families". This is a conclusion that I have inferred in my own work and it is good to see it supported empirically.

Also, data on the age at which work is initiated, the respondents' reasons for working, and the wife's contribution to household income are all very useful.

However, there are several points on which the authors seem to lack conceptual clarity and would do well to reconsider them in the full study that they are planning to complete in the future.

For instance, they say that women decide to go to work mainly because of financial need, job satisfaction, and other reasons. Now, I can understand the financial need as a reason for working, but job satisfaction is an aspect of the work situation itself and cannot be considered a part of the motivational background leading to the decision to work. It may be an aspect of the willingness to continue working but it cannot be considered an element in the initiation of work.

Then, I am not sure how the authors measured the completion of the respondents' family size. In one of their tables, they show the percentages of women who worked at various stages of their life cycle: before the birth of their children, before completion of family size, and after completion of family size. Now, how can the authors arrive at women's completed family size from a cross-sectional sample of women when some of the respondents are still in their reproductive ages?

Again, the authors found a positive association between parity and work participation. From this they deduce that work participation emerges from or is determined by a large family size. They may be correct but this conclusion cannot be accepted as valid without the relevant multivariate analysis.

In fact, the authors have drawn most of their conclusions from simple cross tabulations, perhaps because the study under review is a preliminary analysis of the pilot survey. I presume that the authors will be sharpening their analysis through the use of appropriate multivariate techniques before they make final conclusions from the survey. The insufficiency of bivariate analysis is particularly significant in the context of the relationship between the background socio-demographic variables and the mean number of CEB shown by the authors in one of their tables.

Since this is the first report on the pilot survey, the authors have given a superficial and preliminary introduction to many different themes. I hope that they will develop additional in-depth, multivariate analyses of specific topics in the future.

Finally, I want to re-emphasize a couple of the interesting results of the study. One of these is the role of the eldest daughter in sharing household responsibilities. Another one is the focus on ethnic differentials in work participation and fertility, even though one quickly runs into the problem of very small cells when the sample is only 110. In meeting the objectives set out by the authors, they have had only a partial success. I don't think work commitment was measured comprehensively enough, for the motivation for work was not conceptualized adequately. The objective that was more adequately met was the measurement of the contribution of the wife's income to total household income.

Nasra M. Shah

Ministry of Public Health, Kuwait

REFERENCES

[1.] Chai, L. H., and M. Chao. "Fertility and Women's Labour Force Participation in Korea". Recent Empirical Findings on Fertility." Korea, Nigeria, Tunisia, Venezuela, Philippines. Washington, D.C.: Smithsonian Institution Interdisciplinary Communications Program. 1976. (Monograph Series No. 7)

[2.] Mazumdar, D. "The Urban Informal Sector". Worm Development. Vol. 4, No. 8. 1976.

[3.] Papola, T. S. "Sex Discrimination in the Urban Labour Market: Some Propositions Based on Indian Evidence". In Richard Anker et al. (eds.), Women's Roles and Population Trends in the Third World. London: Croom Helm. 1982.

[4.] Sathar, Z. "Women's Status and Fertility in Pakistan". Paper presented at Workshop on Women's Status and Fertility Sponsored by Rockefeller Foundation, New York, July 8-11,1985.

[5.] Sathar, Z., and F. Kiani. "Delayed Marriages in Pakistan". Paper presented at the Third Annual General Meeting of the PSDE, Islamabad, August 10-12, 1986.

[6.] Sen, G., and C. Sen. "Women's Domestic Work and Economic Activity". Economic and Political Weekly. Vol. XX, No. 17. 1985.

[7.] Sethuram, S.V. "The Urban Information Sector: Concept, Measurement and Policy". International Labour Review. Vol. 114, No. 3. 1976.

[8.] Shah, N., and M. Shah. "From Non-use to Use: Prospects of Contraceptive Adoption". In I. Alam and B. Dineson (eds.), Fertility in Pakistan: Review of Finding from the Pakistan Fertility Survey. Voorburg, The Netherlands: ISI. 1984.

[9.] Sirageldin, I., and A. Khan. "Son Preference and the Demand for Additional Children in Pakistan". Demography. Vol. 3, No. 4. 1979.

[10.] Syed, S. "Female Status and Fertility in Pakistan". Pakistan Development Review. Vol. XVII, No. 4. 1978.

[11.] Westoff, C., and A. Pebley. "Alternative Measures of Unmet Need for Family Planning in Developing Countries". International Family Planning Perspectives. Vol. 7, No. 4. 1981.

(1) Data on fertility was collected in National Impact Survey 1968-69, Pakistan Fertility Survey 1975 and Contraceptive Prevalence Survey 1984; data on labour force participation is collected through Labour Force Surveys conducted by the Central Statistical Organization.

(2) For the purpose of this paper, the formal-informal dichotomy is used to distinguish some important characteristics of women's employment, such as level of earnings, working conditions criteria for entry etc. A more detailed discussion of the informal sector is available in Sethuram 171 and Mazumdar 12l.

SHAHNAZ KAZLAND ZEBA A. SATHAR *

* The authors are Senior Research Economist and Senior Research Demographer, respectively, at the Pakistan Institute of Development Economics. This is a revised version of the paper submitted at the Third Annual General Meeting of the Pakistan Society of Development Economists.

Table 1
Education Level by Occupation of Respondent (Percentages)

 Pro- Lower Grade Domestic
 fessionals Professionals Servants

No Education -- -- 86.7
Less than Matric -- 16.7 13.3
Matric, F.A. 7.1 50.0 --
B.A. &above 92.9 33.3 --

 100 100 100

 Miscellaneous Factory All
 Workers Workers

No Education 86.4 29.6 36.4
Less than Matric 13.6 29.6 14.5
Matric, F.A. -- 37.0 19.1
B.A. &above -- 3.0 30.0

 100 100 100

Professionals: doctors, bankers, administrative personnel, teachers.

Lower grade professionals: nurses, lady health visitors, telephone
operators.

Miscellaneous workers: casual labourers, home workers, tailors,
potters, weavers, vendors.

Table 2 Average Monthly Earnings by Occupation of Respondent

Occupation Average Monthly Number of
 Income (Rs) Women

Doctor 5817 6
Banker 4033 6
Administrative Personnel 3233 3
Teacher 2145 13
Nurses 1427 7
Lady Health Visitors 1376 3
Telephone Operators/Card
 Punchers 1882 8
Factory Workers 1137 27
Casual Labourers 667 6
Home Workers 447 7
Domestic Servants 510 5
Sweepers 738 10
Potters/Weavers 381 6
Vendors 533 3
Average Earnings 1614 110

Table 3

Distribution of Total Household Income/Total Household
Income Net of Women's Earnings

 Percentages of all Households
 having the Monthly Income in
 the Income Class as

 Income Net
 of Women's
Monthly Income Gross Income Earning

Less than Rs 1750 24.5 46.4
Rs 1750-2999 28.2 18.2
Rs 3000-6999 23.6 21.8
Rs 7000 or more 23.6 13.6