Informalisation of women's work: consequence for fertility and child schooling in urban Pakistan *.

Kazi, Shahnaz ; Sathar, Zeba A.

INTRODUCTION

Female employment is considered an important means of lowering fertility through ways such as raising the age at marriage, through influencing desired family size and also through better knowledge and use of contraceptives. Increasing female labour force participation is frequently recommended as a critical policy measure for reducing the birth rate. However the significant inverse relationship between employment and fertility found for developed countries is weak or absent in the case of developing countries [Rodriguez and Cleland (1980)]. More recent evidence indicates that it is not so much employment per se but type of employment which is a critical determinant of reproductive behaviour [United Nations (1985)]. It has been shown that while high status professional jobs are associated with greater influence on women's domestic autonomy and fertility, low paying jobs lead to an increasing burden of work with entirely different implications for fertility and other household related behaviour.

In the context of Pakistan, despite two decades of industrial growth and development, official data sources show stagnant and low levels of female labour force participation rates (LFPR) in urban Pakistan. The LFPR for urban women ranged between 3 and 5 percent for the period between 1971 and 1988. Data collection methods of government agencies are known to greatly underestimate female labour force participation (FLFP) particularly in rural areas and in the urban informal sector where the distinction between productive and domestic activities tends to be ambivalent. Evidence from micro surveys indicates, on the contrary, an increasing influx of women in the urban labour market, particularly in the informal sector [Sathar and Kazi (1988); Shaheed and Mumtaz (1981); Bilquees and Hamid (1989)]. A large number are shown to be working in home-based piece-rate employment while domestic service mainly as sweepers, washerwomen, maids, etc. is another major avenue of employment for poor females in urban areas with low levels of education and skills.

In the manufacturing sector, findings from special surveys of industrial workers, indicate a very low representation of women workers in the regular workforce and a tendency to relegate women to temporary, casual or contract work [Hafeez (1983); Khan (1986); PILER (1990)]. The informalistion of women's work reduces labour costs since temporary workers are outside the purview of labour legislation and are thereby not entitled to maternity leave and other benefits. Subcontracting in addition, not only allows the firms to circumvent labour legislation but also offers overhead cost advantages.

Evidence from micro surveys has not only highlighted the trend of growing informal employment among urban women but has also provided interesting insights into the distinct demographic and social outcomes associated with variations in the type of employment. It has been shown that low paying, low status jobs in the informal sector have entirely different implications for fertility and other household related behaviour [Sathar and Kazi (1989)].

Until now information on women's employment in the informal sector and its linkages with their reproductive roles has been limited to micro studies. However data from the Pakistan Integrated Household Survey (PIHS) undertaken jointly by the World Bank and the Federal Bureau of Statistics in 1990-91 presents a unique opportunity to verify whether the trends and linkages seen in small, purposively sampled surveys, are supported by nationally representative data where special efforts was made to capture female employment. The PIHS utilised special modules on time use and family enterprise to ensure that women working as unpaid family helpers or women undertaking income-earning activities at home were not missed out in the survey. In addition, female enumerators were asked to ascertain from the female respondents themselves rather than from male household members to ensure representation of women who perform paid work in their homes and those in family-based employment. Data was also collected on other details of women's economic activities such as on earnings, hours worked and work experience together with detailed marriage and reproductive histories. At the household level, education and occupation of all family members was also collected.

The (PIHS) sample comprises 4711 households and 5251 evermarried women aged 15-49. Due to deliberate over-representation of urban areas the sample contains 2513 evermarried women aged 15-49 in the urban areas on whom the analysis is based. (1)

The present study uses the PIHS data to investigate two sets of questions. First, it will address issues related to women's productive activities. What are the employment patterns of women in the urban sector? What characteristics distinguish women working in the informal sector from formal sector workers, in particular, what are the differences in education, earnings, hours of work and length of work experience? The second part of the study explores the differentials in outcomes such as fertility (actual and desired), contraceptive knowledge and adoption and children's schooling by employment status of evermarried women.

METHODOLOGY

For the purpose of this study, the sample of economically active women is divided into two broad categories--formal sector workers and informal sector workers. The distinction between the formal and informal sector is based on the differences in working conditions. The formal sector is characterised by an organised and protected labour market where working conditions are safeguarded by certain laws and regulations which ensure a degree of legal protection to workers. Informal sector activities, on the other hand, are not regulated by any contractual agreement and are outside the purview of any labour legislation. On the basis of this criteria, in the context of Pakistan, the formal sector includes all public sector establishments as well as private sector establishments in the field of industry and credit which comprise 5 workers or more. Establishments which employ more than 5 members come under the purview of labour legislation and are usually considered part of the formal sector. The informal sector workers are further subdivided into three sub-categories; informal sector workers who are employed outside the home, unpaid helpers in family enterprises and women who undertake income-earning activities at home in their own capacity. Women in the formal sector are also subdivided (at least initially) into two groups; the white collar workers (managerial, clerical, professionals and related workers) and factory workers. The analysis that follows will describe these broad categories of employment and their association with fertility behaviour and child schooling. The discussion of linkages between economic and demographic behaviour will, wherever possible, include a comparison with the group who is not involved in any income earning activities.

FEMALE EMPLOYMENT PATTERNS IN URBAN AREAS: FINDINGS BASED ON PIHS DATA 1990-91

The PIHS unofficial estimate of urban female labour force participation rate of 17 percent, which includes a considerable proportion of women in the informal sector, places female employment in the urban areas at a much higher level than other recent estimates. The Labour Force Survey for the same year yielded an estimate of urban labour force participation rates for females of 8.5 percent. Evidence from the Pakistan Integrated Household Survey 1991 supports the prominence of informal sector employment in urban Pakistan. The proportion of women who are most likely to be enumerated as economically active in labour force surveys, those in the formal sector, comprise less than one-fifth of those whose economic activity falls by our definitions into the informal sector.

In keeping with the pattern visible in micro studies, the findings from the PIHS indicate that the overwhelming majority of economically active women between the ages of 15 and 49 work in their homes either as unpaid family helpers in the family business or they undertake income-earning activities at home in their individual capacities and it is precisely these women who are likely to get excluded from official labour statistics

The data indicate distinct differences between women working in the formal and informal sector. Formal sector workers in the sample include women who are employed mainly as teachers, doctors, and 'lower level' formal sector comprises production workers in large factories. On the other end of the spectrum are women who are classified as informal sector workers outside the home who are mainly employed in low status domestic service jobs such as maids, sweepers, cooks and washerwomen. Their earnings, although much lower than their formal sector counterparts, are considerably higher at an average of 609 Rupees than the average monthly remuneration received by home-based workers (Rupees 240).

There are also wide variations in the hours worked whereby women who work outside the home work longer and more regular hours. The mean hours worked per week is highest for production workers, about the same for other formal sector workers and informal sector workers outside the home, while the shortest hours are worked by home based workers.

A comparison of background characteristics of women working in the formal and informal sectors and those not engaging in any income-generating activities shows that women who worked in the formal sector had much higher levels of schooling as compared to those in the informal sector. While on average, workers in the formal sector had completed 11.4 years of schooling, the mean years of schooling for informal workers was 6.5, lower than the average for non-working women. 77 percent of women who worked in the informal sector had received no education as compared to 62 percent of women who were not working and 28 percent of women in the formal sector.

In general, husbands of formal sector workers also tend to be engaged in the same sector. A significantly larger proportion of employed women whether in the formal or informal sector are divorced, separated or with unemployed husbands than is the case for non-working women. Living arrangements seem to vary also by employment status of women. Non-working women were more likely to be living in a non-nuclear household and had a substantially larger household size than those working in the formal sector. Women in the informal sector were placed in between these two groups.

IMPLICATIONS OF TRENDS IN WOMEN'S EMPLOYMENT PATTERNS ON FERTILITY AND CHILD SCHOOLING

Employment is expected to have an impact on reproductive behaviour especially through delaying marriage in the case of pre-nuptial employment and also through increasing access to contraception because of increased mobility outside the home. Women employed in the formal sector marry 3.4 years later than those working in the informal sector and 2.6 years later than non-working women. Exposure outside the home, regardless of whether it is in the formal or informal sector, positively influences knowledge about contraceptives. The highest ever use of contraceptives is among women working in the formal sector followed by those women working in the informal sector outside the home. Lowest contraceptive use is reported by women not engaged in economic activity followed by home-based informal sector workers. Interestingly, there is not much variation in what women state as their ideal number of sons and daughters. When responding to the question of "how many sons and daughters would you like to have if you were starting your family today?" There seems to be a clear convergence of opinion. Four children, more sons than daughters, are considered ideal regardless of women's employment status.

Formal sector employment in high status jobs as professional and white collar workers or even as factory workers in large establishments is associated with much lower achieved fertility. Informal sector employment is not. In fact, fertility of women engaged in informal sector activity inside and outside the home have roughly the same high levels of fertility which are higher than the fertility of women not engaged in income generation. This finding may be partly due to be the effect of differentials in levels of education by employment status. For this purpose we applied a multivariate model controlling for the effects of women's education and also a proxy for economic status of the household, by including husband's occupational groupings. Interestingly, employment in such cases was still found to retain an independent effect on fertility of women even though its impact was reduced, employment in the formal sector was associated with significantly lower fertility than in the informal sector. Fertility for the latter group was once again higher than for non-working women. Given that the majority of women consider four as their ideal family size, it is clear that only women in the formal sector achieve this while others, particularly women in the informal sector, are having children far in excess of their ideal.

The higher levels of fertility of informal sector workers outside the home is surprising given that this group has higher knowledge and ever use of contraceptive methods. It seems to suggest that women in this sector are likely to be pushed into seeking employment because of a large family and already have a number of children well in 'excess' of their ideal family size.

Mother's employment status is also expected to have an impact on child schooling. Women who are more educated and working in higher status jobs would be expected to have more children in school because of their financially stronger position and more modern attitude to work and education. They would also be expected to have less differences in the preference for schooling of boys and girls, and as a result the level of discrimination in schooling would be expected to be lower. In poorer families where women work outside the home a likely outcome of mother's employment is the withdrawal of children, particularly of girl children from school to perform household chores or look after younger siblings while their mothers are at work. This could have serious consequences for the generational reproduction of disadvantages in access to schooling and thereby to formal sector or higher level occupations. The findings based on the Pakistan Integrated Household Survey generally support our expectations. Chances of currently attending school are much higher for children of mothers working in the formal sector. Mothers' employment in the informal sector is associated with much lower chances of children being in school and greater chances of their dropping out. Gender discrimination in sending children to school seemed lower for those women working outside the home, both in the formal and informal sector, despite considerable differences in the levels of children's schooling across the two sectors. The gender gap in schooling which is widest for children aged 15-19, is more pronounced amongst women working in the informal sector at home and nonworking women. To some extent this might be reflective of households that have a stronger adherence to restricting female movement outside the home which affects the employment status of mothers and schooling of girl children. However, it ought to be noted that the proportions of boys and girls attending school, though less inequitable by gender than for home-based workers, are lowest for informal sector workers outside the home.

The proportion of children who drop out (based on those who ever attended school) is notable only after ages 10. Drop out proportions are, once again, highest for women working in the informal sector outside the home and lowest for women in the formal sector. Nevertheless, even in this latter group 44 percent of girls aged 15-19 had dropped out, a much higher proportion than boys of the same age. Though drop-out rates to a large extent must reflect economic differences across employment status of mothers, they are also influenced by the commitment to send and keep children in school on the part of parents as weighed against the needs of their labour and time in the household. Though women in the informal sector are likely to be from the poorest households, at the same time this translates into a weak commitment to provide education for their children. Such a trend has grave consequences for the intergenerational transfer of poverty. The best chances of schooling, especially for girls, 'are among children of mothers working in the formal sector. Curiously, non-working women, either because of their own employment status (which would presumably not require any child care surrogates) or because of their better economic conditions, are more likely to educate their children than women in the informal sector, inside and outside the home.

As mentioned earlier the findings on child schooling are at a very preliminary stage and a multivariate analysis controlling for the effect of income and education is required before any conclusive statement can be made of the independent effect of female employment on child schooling.

The implications of these findings for future employment trends and for women's role in development are far reaching. This Study is merely a first step towards verifying the growth of informal sector activity amongst urban Pakistani women and substantiating some of its demographic and economic implications. Future research in this area will demand that the informal sector be given its due attention.

Hence development even when associated with rising levels of female labour force participation may not guarantee any improvement in women's position or the beneficial effects related to women's employment if as the evidence for a number of developing countries indicates it leads to the growing concentration of women in the informal sector [Standing (1989)]. Inspite of the upward trend in the absorption of women workers in informal employment and its distinctly different demographic and social impact, research on household responses to women's employment rarely incorporates distinctions by sectors or type of employment.

REFERENCES

Bilquees, Faiz, and Shahnaz Harold (1989) A Socio-economic Profile of Poor Women in Katchi-Abadis. Islamabad: Pakistan Institute of Development Economics.

Hafeez, Sabeeha (1983) Women in Industry. Islamabad: Women's Division.

Khan, N. S. (1986) Women's Investment in the Industrial Sector in Punjab. Lahore: Applied Socio-Economic Research.

PILER (1990) Women in the Industrial Labour Force. Unpublished Report.

Rodriguez, German, and John Cleland (1980) Socio-economic Determinants of Fertility in Twenty Countries. World Fertility Survey Conference 1980, Voorburg, Netherlands.

Sathar, Zeba, and Shahnaz Kazi (1988) Productive and Reproductive Choices of Metropolitan Women: Report of a Survey in Karachi. Islamabad: Pakistan Institute of Development Economics.

Sathar, Zeba, and Shahnaz Kazi (1989) Female Employment and Fertility: Further Investigation of an Ambivalent Relationship. The Pakistan Development Review 28:3.

Shaheed, Fareeda, and Khawar Mumtaz (1981) Invisible Workers: Piece Labour Amongst Women in Lahore. Islamabad: Women's Division.

Standing, Guy (1989) Global Feminization through Flexible Labour. World Development 17:7.

United Nations (1985) Women's Employment and Fertility. New York: United Nations Population Studies. No. 96.

* Owing to unavoidable circumstances, the discussants comments on this paper have not been received.

(1) Raising factors have to be applied to the sample in order to make it nationally representative. Most tables presented provide both the sample and 'raised' number of cases.

Shahnaz Kazi and Zeba A. Sathar are Chiefs of Research at the Pakistan Institute of Development Economics, Islamabad.