Patterns of Female Employment in Urban India

Economic and Political WeeklyDecember 2, 20065001Given this caveat, this paper assesses the impact and analysesthe nature of the changes that have taken place in the patternof female employment in urban India in the context of the reformprocess. The study is based primarily on secondary employmentdata pertaining to the four major rounds (38th, 43rd, 50th and55th rounds) of the NSS surveys, i e, the quinquennial surveysconducted in 1983, 1987-88, 1993-94 and 1999-2000 respec-tively, by the NSSO. The compound annual average rates ofgrowth of employment (to be used for the entire study) have beencalculated from the absolute employment figures for the years1983, 1987-88, 1993-94 and 1999-2000. The absolute employ-ment figures have been calculated by extrapolating the censuspopulation figures of 1981, 1991 and 2001 for the years corres-ponding to the NSS rounds and plotting the respective NSSestimates on the extrapolated figures.The rate of employment generation all over India decreasedover the 1990s both in the rural as well as in urban areas. Inthe urban sector there has been a decline in the annual averagerate of growth of employment from 3.3 per cent to 2.8 per cent.While the urban employment growth rate for males shows a sharpdecline in the later phase of the 1990s, the same figures for thefemales show stagnation (Table 4). Placed in this context, thispaper analyses whether it has been possible for women workersto experience an improvement in their employment conditions.The urban female employment in India has always been per-sistently and substantially lower than that in the rural areas. Thisis evident from the work participation rates (WPRs) for the agegroup 15-59 provided by the NSS (Table2).1 In the urban areas,while the male WPR varied between 50 and 52 per cent withno clear-cut indication of an increase, the female WPR, althoughsomewhat stagnant till 1993-94, has a declining trend. This isinteresting in the sense that it suggests that the process offeminisation of employment in India is not borne out by the dataon the WPRs of urban females. In fact, the lower WPR of theurban females as compared to the rural females is to some extentan indication of the inability of the industrial development processesin the urban areas to create sufficient employment opportunitiesfor women.Patterns of Employment ofUrban Women in Two Major SectorsManufacturingThe NSS estimates in the four quinquennial rounds show thatmanufacturing and services constitute a major share of employedfemales in urban India. The share of females employed inmanufacturing was 24 per cent and that in services was 34.4 percent in 1999-2000 (NSS Employment-Unemployment Survey:Key Results, various rounds).The percentage share of women employed in manufacturinghas been significant in the urban sector. The disaggregations ofthe manufacturing sector into the two divisions provided by theNSS reveals that division (1) has substantially the greater shareof employment for urban females than the other division. By thedefinition used by the NSS, manufacturing (1) consists of agro-based industries that include textiles, garments, leather and leather-products, beverages and food products, tobacco, paper and paperproducts, etc, and therefore traditionally employs most of thewomen workers. But the data in Table 3 suggests that the shareof the urban females employed in manufacturing (1) has declinedsteadily and sharply over the period under consideration, i e, the38th, 43rd, 50th and 55th rounds of the NSS quinquennialsurveys. The other sector, i e, manufacturing (2) that comprisesall non-agro based industries – specifically chemicals, rubber andplastics, basic metals, non-metallic mineral products and manu-facturing of other electrical and electronic as well as transportequipment – have experienced an increase in the share ofemployed urban females. A point that can be noted here is thatthe decrease in the share of the employed in the manufacturingsector as a whole has been more by the usual principal status(UPS) than by the usual principal and subsidiary status (UPSS),although by a very small extent.To understand the importance of this discrepancy in the datafor the two different activity status it becomes important firstto note the definitions of both. The NSS data on employmentdistinguishes between the “principal” and “subsidiary” status ofan activity based on whether the person is “usually” engagedin the activity. Thus a person is classified as UPS worker ac-cording to the activity on which the person has spent a relativelylonger time of the preceding 365 days in the reference period.The activities pursued by a person are classified into threebroadcategories, namely, (i) working or employed, (ii) seekingor available for work (unemployed), and (iii) not in the labourforce.A “non-worker”, that is one whose status has been alreadyascertained on the basis of UPS, is someone whose major partof time in the preceding year was spent as either unemployedor not in the labour force. However, such a person could stillbe involved in some economic activity in a subsidiary capacityin which s/he is then referred to as a “subsidiary status worker”.Table 1: Annual Average Rate of Growthof All-India Urban Employment(in Per cent)YearMaleFemalePersons1983 to 1987-883.14.43.31987-88 to 1993-942.94.43.11993-94 to 1999-20002.54.42.8Source: Calculated from Census reports, 1981, 1991, 2001 and NSSEmployment-Unemployment Survey: Key Results, various rounds.Table 3: Share of Manufacturing Sector in Urban FemaleEmployment: All-India UPSUPSSYearManufacturingManufacturing(1)(2)Total(1)(2)Total198321.14.926.022.04.826.81987-8820.76.226.921.45.727.11993-9418.84.823.619.05.124.11999-200017.06.323.217.66.424.0Source: NSS Employment-Unemployment Survey: Key Results, various roundsTable 2: Work Participation Rates according to UPSS: All-IndiaRoundsYearRuralUrbanMaleFemaleMaleFemale38th198354.734.051.215.143rd1987-8853.932.350.615.250th1993-9455.332.852.015.455th1999-200053.129.951.813.9Source: Employment and Unemployment in India, 1999-2000. Key Results,NSSO.

10 8 6 4 2 0

5.6 6.0 6.1 6.5 7.6 8.1

-		-		-
	GDP		Tertiary

8.6

10.00

5.4

5.00

0.82 0.55

0.61 0.37

0.00 1983 to 1993-94 1993-94 to 1999-00 Year

Economic and Political WeeklyDecember 2, 20065005data is that the increases have been more by the UPSS than byonly taking the UPS, which then implies that even in the servicesector there has been an increase in the subsidiary activities. Anincrease in the subsidiary activity in this sector might have takenthe form of increased domestic service among urban women. Thiscan be better substantiated as we look at the data relating to theshare of female workers by the type of contract (given in Tables9aand 9b). The data pertaining to the type of contract shows thatthe sector is predominated by regular employment. This is in itselfa positive sign but there has been a drop in the share of employedof urban females in this category by approximately 3 per centby both the type of activity since 1983. It is worth mentioningthat the increase in the regular employment between the 50thand the 55th round in the public administration, personal,communityand other services have been significantly more bythe UPSS. This indicates an increase in a certain kind of regularsubsidiary activity for urban women. This kind of subsidiaryactivity might have the form of domestic services, which maybeextremely low paid even though they have a regular basis ofpay-ment.In fact, micro studies have shown a rising incidence ofdomestic services in urban areas. In that sense, regularemploymentmight not be an indication of better job options for the women.In the service sector, there has been a drastic decline in therate of growth of female employment (Table 8). The declineshave taken place between 1993 and 2000. This is interesting aswe find that within this period the rate of growth of output inthis sector was above 8 per cent. It implies that the output growthrate in this sector was not translated into employment growthopportunities. We find that the employment elasticity of outputgrowth for the urban females to be declining in this sector. Infact, the employment elasticity of output growth for the urbanworkers as a whole shows all the more decline in this sector2(Figure 2).Again from Table 7, it is evident that although there has beena rise in the share of employed in the banking and real estatesectors, yet the percentage of women involved in the sector isquite insignificant. The data show that there have been impressiverate of growth of employment for urban females till 1993-94,but between 1993 and 2000 the rate of growth of female em-ployment in this sector has dropped drastically, more so by theUPS (Table 8).A quick look at the female employment trends in the BPOs andthe ITeS: The banking and real estate sectors comprise financialintermediaries, insurance, hardware and software relatedservices,and the IT-enabled services (IteS) and business process outsourcing(BPO) services. In the wake of liberalisation in the Indian contextit becomes important to mention briefly the rapid rise of the exportsector related to software and information technology enabledservices (ITeS). This sector in the 1990s has been hyped a lotregarding its growth potential and its employment creatingcapacities, especially for urban educated females. In fact, severalstudies conducted to this effect reveal the promising nature ofthis sector in terms of offering a combination of employmentand export revenues from hardware, software, BPOs and ITeS.The rise in the output growth rate in the banking and realestatesegment of the service sector in the 1990s is often attributedto the increased activity in this sector. The nature of workinvolvedin these services includes a range of activities like dataentry, data processing, medical transcription, back-office workand increased activities around call centres. A look at theemploymentcreating capacity of this sector reveals that muchof the software-related activities are in English. Hence theemployment generated by it caters to only a very specificurban,metropolitan, English-speaking, rich and mostly uppercaste section of society. In fact, the recent trends of decline inthe rate of educated unemployment in urban areas confirms thisto some extent.The banking and real estate sectors show an increased shareof women employed, although the figure is quite low, and it alsoregisters high rate of growth of employment. The unavailabilityofemployment data pertaining to the software sector prevents anyspecific conclusions from being drawn. But some micro evidencesuggests that there has been a reasonable participation of womenTable 8: Annual Average Rate of Growth of Urban FemaleEmployment in the Service Sector: All-IndiaUPSUPSSIndustry/Year19831987-881993-9419831987-881993-94totototototo1987-881993-941999-20001987-881993-941999-2000Banking andreal estate10.911.76.312.510.77.6Public administratorand other services4.65.02.46.14.82.5Service (1+2)4.95.32.76.35.12.8Source:Calculated from Census Reports 1981, 1991 and 2001 and the NSSEmployment-Unemployment Survey: Key Results, various rounds.Table 9a: Urban Female Workers by Type of Activity in theService Sector according to the UPSRounds38th43rd50th55thYear19831987-881993-941999-2000Self-employedBanking, real estate, etc17.78.517.623.0Public administration and other services9.711.116.113.3Regular and salariedBanking, real estate, etc74.986.682.175.6Public administration and other services76.075.871.172.9CasualBanking, real estate, etc7.44.80.41.4Public administration and other services14.313.112.813.7Source: NSS Employment-Unemployment Survey: Key Results, various rounds.Table 9b: Urban Female Workers by Type of Activity in theService Sector according to the UPSSRounds38th43rd50th55thYear19831987-881993-941999-00Self-employedBanking, real estate, etc19.518.822.426.1Public administration and other services12.521.222.216.8Regular and salariedBanking, real estate, etc73.677.177.272.5Public administration and other services72.165.864.269.6CasualBanking, real estate, etc6.94.10.41.4Public administration and other services15.413.013.613.6Source: NSS Employment-Unemployment Survey: Key Results, various rounds.Table 10: Annual Average Rate of Growth of Urban FemaleEmployment in Trade: All-IndiaYearUPSSUPS1983 to 1987-884.15.61987-88 to 1993-943.52.71993-94 to 1999-0012.110.2Source:Calculated from Census Reports 1981, 1991 and 2001 and the NSSEmployment-Unemployment Survey: Key Results, various rounds.

10.9 10.7 16.4 9.5 9.8 10 16.9 9.9 0 2 4 6 8 10 12 14 16 18

1983 1987-88 1993-94 1999-00 Year

Economic and Political WeeklyDecember 2, 20065007into trading activities, mainly in wholesale and the petty retailtrade. This trend becomes clearer as we look at the share of urbanfemale workers by their activity type/occupational status. Thisshare shows that this sector is dominated by self-employment.Self-employed women workers constitute more than 80 per centof the total female employment in this sector (Table 11). Thispredominance suggests that there has been an increased activityaround retailing and wholesale trade. Estimates given bySundaram(2001) reveal that retail trade alone accounted forthemajor share of the increase in absolute terms in this sector.The overall share of retail trade in total employment in theservicesector increased from 20 per cent in 1993-94 to about27 percent in 1999-2000 and the share of retail trade in women’semployment increased from approximately 20 per cent in 1993-94to 24 per cent in 1999-2000. It can therefore be concluded thatin the 1990s the increased activities in the trade sector mainlyrevolved around retail trade. More often, in the case of women,this kind of retailing boils down to street vending and petty sellingof a whole range of items from green vegetables to ‘paan’, beediand cigarette.From the above discussion it is therefore clear that thestagnationand declining shares of women’s employment in manufacturing,the reduced employment opportunities for women workers inboth manufacturing and services, in this period, have all playeda role in pushing or crowding large numbers of women intopettyretail trade, even as many have been pushed out of theworkforce.Feminisation Not EvidentAlthough the main concern of this paper has been to assessthe extent to which the female workforce has been absorbed inexport-related activities, a look at the male employment figuresover this period suggest that the situation had been worse forthe males. While the urban male employment growth rate hadbeen declining steadily over the period (Table 1), the WPRs ofthe urban males have remained somewhat constant. In such asituation, the rate of growth of employment of the urban malesin the service sector experienced a sharp decline (Tables 12a,12b). While the male employment growth rate in manufacturingshows a slight increase according to UPSS, it registers a declinebetween 1983 and 1999-2000 when only UPS is considered. Buta notable point in this is that the employment growth rate forurban males between 1987-94 and 1993-2000 in the manufac-turing sector registers an increase, while the same for urbanfemales in this period has been declining. A comparison ofemployment growth rates of urban males and females in themanufacturing sector thus reveals that the situation for the urbanwomen was worse than that of males in this particular sector.It can also be noted that “feminisation” was not the relevantprocess in India in this period.In the trade sector, one finds from the above estimates thatalthough there have been increases in the annual average rateof growth of urban male employment, the growth rates arenotashigh as compared to their female counterparts. A majorfall in the employment growth rate of the urban males hasbeenobserved in the services sector, where there has been adropin the absolute number of male workers, which is evidentfrom thenegative rate of growth. The reduced employmentgrowthrates for both males and females in the service sectortosome extent can be attributed to the decline in the organisedsector employment in the 1990s as much of organised sectoremployment was based in the service sector.4Apart from the above caveat in the overall employmentsituationof the urban females in India, the NSS estimates alsoshow quite high unemployment rates by the daily status for bothmales and females, the rates being higher for females as has beenshown by the above table.Brief Discussion on the States5Given the above observations, the urban female employmenttrends, when analysed at a state disaggregated level, reveal someimportant facts. While the overall rate of growth of employmentin individual states gives a mixed result and not that of acleardeceleration following the all-India pattern, the sectorwisepictures are somewhat different. The results reveal a lot ofvolatility and discrepancies in the employment situation of theurban females. In the manufacturing sector, there has been adeterioration of employment opportunities for the urbanwomen,which has been mentioned in the earlier part of thepaperand it is reflected in the analysis of individual states asincreased subsidiary activities among the women. In fact, aninteresting observation for the states in the manufacturingsectorhasbeen that the urban women in this sector have beensporadically engaged in economic activities rather than engagedfor a continuous period, which signifies a lack in the avail-abilityof continuous work for urban women in this sector inthelast period. The kind of volatility reflected in the rate ofgrowth of employment of urban females it has been noticedmainly in the states where the share of women employed inmanufacturing havebeen quite low, like Punjab (only about13per cent) or in the states like Karnataka, Madhya Pradeshand Orissa where the shares of employed have dropped sub-stantially over the period.Even the services sector gives similar results of volatility,tremendous fluctuations and disparities in the rate of growth ofemployment in this sector. The share of urban women workersin this sector for almost all the 14 major states has been sub-stantially high. In more than half of the states over 30 per centof female workers are employed in this sector. The rate ofgrowthof employment as well as the share of employedfemaleshasbeen higher by the UPS. But the high degree ofvolatility also indicates instabilities in terms of employmentgeneration in the sector, which is not desirable for an economy.In fact, as we analyse the trends at the state level, it is interestingto note that states like Bihar, Haryana and Punjab, etc, whichshow high degree of fluctuations in the rate of growth ofemployment have ahigher proportion of women workers en-gaged in the tradesector.Also an analysis of open unemployment rates across statesshows that in Bihar and Maharashtra there have been significantincreases in these rates by all three criteria; in Bihar, specificallythe rate has increased from 1 per cent in 1983 to approximately9 per cent in 1999-2000. In Rajasthan and UP, we find that thetrend is similar to the all-India pattern where the rates haveincreased by the usual status but have declined by the other twocriteria [NSSO Reports].Such wide disparity among the states makes it difficult toarriveat any generalised conclusion. However, one canobviouslysee that there have been declines in the rate ofgrowthoffemale employment in half of the states considered,

lli