REVIEW OF WOMEN’S STUDIESoctober 25, 2008 EPW Economic & Political Weekly48Exclusions From and Inclusions In ‘Development’: Implications for ‘Engendering Development’Padmini SwaminathanUsing official reports and supplemented with field notes, this paper examines the quantum and nature of exclusion from and inclusion in “development”. Data reveal that for women, education and employment move in opposite directions, literacy among women workers is lower than literacy among non-working women and development is not closing the gender gap. The bulk of women “included” in development disproportionately constitute the unorganised workers; the latter are those in the unorganised sector plus those working in the formal sector as informal workers. Given such a macro context, the paper is sceptical of the feminist agenda of engendering development. The author is extremely grateful to Mary E John, without whose perseverance this paper would not have been written.Padmini Swaminathan (padminis@mids.ac.in) is at the Madras Institute of Development Studies, Chennai.Almost a decade ago, I had engaged extensively with anal-ysis of secondary data, mainly Census 1991, to explicate how an otherwise “economically and socially developed” state such as Tamil Nadu perpetuated development-induced “vio-lence” on its citizens, particularly women, adolescent girls and girl children, with the severity of this violence increasing for dalit girls and women [Swaminathan 2002]. A quick recap of the main arguments and data-related findings of that paper is in order here, not only to enable an assessment of the situation a decade later, but to also deliberate on the implications of the macro sce-nario – as captured by our official data systems – on continuing efforts to make development more inclusive, beyond “engender-ing” development itself. A crucial argument was that the interpretation of statistical correlation as causation has contributed in no small measure to policies that aim at helping and enhancing women’s status, while being completely devoid of the contexts and circumstances that account for women’s/households’ observed behaviour on the ground. In addition, the analyses of official data for the decade 1981-91 in Tamil Nadu revealed the following:– Between 1981 and 1991, work participation rates (WPrs) increased for women across Tamil Nadu; however, further disaggregation revealed that where femaleWPRs had increased, this was accounted for largely by increases in the WPRs of girls and adolescents. – Data also revealed significant differences in patterns ofWPRs for male/female children and adolescent populations of the state. Between 1981 and 1991 for the state as a whole, the relative fall in male child and adolescent WPRs was significant (almost 10-12 percentage points) when compared to the insignificant fall inWPRs of female child/adolescent population (1-2 percentage points only). Again, in those districts where employment for women showed increases in the census occupation categories of “household” and “other than household”, a substantial section of thoseemployed were actually in the adolescent age category of15-19years.– The juxtaposition of employment and education data brought out interesting but disturbing patterns. While at one level, the gender gap (females per 1,000 males) increased significantly when data were computed for different levels of literacy, what data also corroborated was the fact that districts with higher levels of adolescent girls in employment were also the ones where signifi-cant proportions of adolescent girls were “not attending schools”.– These trends contributed significantly to the poor levels of formal education in Tamil Nadu’s female workforce.– The disaggregation of employment and education data by sexand social class revealed yet another facet of the complex

REVIEW OF WOMEN’S STUDIESEconomic & Political Weekly EPW october 25, 200849problematic, namely, a polarised population in terms of gender and caste. Thus, far more girls and scheduled caste (SC) girls in particular were “working” when compared to non-SCs; far more women and SC women in particular were in farm rather than in non-farm employment; far more women and SC women in par-ticular were illiterate when compared to non-SCs. In addition, data also revealed that the proportion of women workers re-turned as illiterate was not only significant when compared to men, but that, unlike in the case of men workers, women not in employment were more educated than women in employment.– In conclusion, we had noted that the caste and gender inequi-ties were structural in nature and could not be rectified merely by addressing individuals and/or individual households. Our public policies, however, are not geared towards tackling the sys-temic nature of the class, gender and caste-based discrimination outlined above. Unless such structural inequities were con-fronted, their entrenched nature would render them even more resistant to change as time went by.A decade later, is there a change in the way we view our record of development? And, do we still characterise our development as one that continues to inflict “violence” on the poor in general and women in particular? My own submission is that not only does “development violence” continue but the violence itself has become more overt with severe consequences for the poor and poor women in particular. After 60 years of independence, our Plans are still struggling to make development inclusive, with the Eleventh Five-Year Plan (FYP) approach document being entitled, ‘Towards More Inclusive Growth’ – an indirect acknowledgement that the previous 10FYPs have failed the “inclusiveness” test. This paper is organised as follows. In Section 1, we carry for-ward our analysis of secondary data to understand “change” since 1991. It needs to be stated at the outset that for reasons not officially communicated and/or made public, the 2001 Census data are being disseminated in bits and pieces, because of which it has not been possible to undertake the kind of disaggregated analysis that was possible with the 1991 data.In Section 2, we discuss the phenomenon of the inability of theeconomy to generate employment with an emphasis on the nature of the disjunction between employment and education. Section 3 discusses the terms on which women and women workers in particular find themselves “included” in “development” based on the findings of the Report of the National Commission for Enterprises in the Unorganised Sector (NCEUS) [GOI 2008]. Section 4 provides highlights from a field based study undertaken in Tamil Nadu in order to underscore the contexts and circum-stances that force households to make gendered choices, but which do not form the basis of policies and programmes aimed at women and girls. Section 5 recapitulates our understanding of gendered inclusions/exclusions from development and outlines the implications for our efforts to “engender” development.The repeated allusion to development violence is aimed at making explicit two aspects of what passes for economic growth/development in the country: one, outright exclusions from devel-opment as revealed through, for example, proportions of the population not formally literate, unemployed and so on. Two, an equally crucial aspect of development violence that has to do with the terms by which “inclusion in development” takes place. 1 Understanding Inclusion/Exclusion from Macro-DataAnalysis of available1 secondary data on employment and educa-tion for the decade 1991-2001 (Census) and post-2001 (National Sample Survey) reveals a complex picture of exclusion and inclu-sion at the macro level. In 2001, 63 per cent of the male population has been returned as “literate” compared to 45 per cent of the fe-male population nationally. The equivalent figures for rural India are 59 per cent males as against 38 per cent females, and for urban India 75 per cent and 63 per cent. In other words, for the country as a whole and after 60-odd years of independence, 37 per cent males and 55 per cent females suffer exclusion from formal literacy. The analyses of data relating to those returned as literate provide a pic-ture of the “nature and level of inclusion”: for the country as a whole, 66 per cent of literate males have not gone beyond “matric/secondary” against 72 per cent for literate females; only 7 per cent of literate males and 5 per cent of literate females figure in the category, “graduate and above other than technical degree”, while the “technical degree” category has slightly less than 2 per cent of the literate males against 0.6 per cent of literate females (Table 1). The analysis of literacy levels of the “worker” population for 2001 reveals a disturbing contrast between male and female workers. While almost 71 per cent of male workers are literate (against a 63 per cent literacy among males in the general popu-lation), only 36 per cent female workers are literate (against 45 per cent general female literacy). In other words, while develop-ment may have increased the work participation rate for females, it has not translated into a greater proportion of literate women becoming workers as has happened in the case of males. What is Table 1: Literacy among General Population and among Workers – India and Tamil Nadu(2001, in %) General Population Worker PopulationTotal/Rural/Urban Illiterate Literate Illiterate Literate Males Females Males Females MalesFemalesMalesFemalesIndia Total 36.76 54.85 63.24 45.15 29.50 64.30 70.50 35.70Rural 41.42 61.67 58.58 38.33 34.71 67.86 65.29 32.14Urban 24.96 36.68 75.04 63.32 15.91 39.79 84.09 60.21Tamil Nadu Total 27.36 42.86 72.64 57.14 21.33 53.50 78.67 46.50Rural 32.49 51.21 67.51 48.79 27.93 59.33 72.07 40.67Urban 20.8832.20 79.12 67.80 12.50 37.23 87.50 62.77Source: Census 2001. The Indian Association for Women’s Studies held its Silver Jubilee Conference in Lucknow, February 7-11, 2008 under the theme “Feminism, Education and the Transformation of Knowledge: Processes and Institutions”. The occasion of the conference represented a special opportunity to confront the challenges facing the women’s movement and women’s studies today, after several decades of intervention within and beyond institutions of higher education. The IAWS is grateful to EPW and Maithreyi Krishnaraj for agreeing to bring out revised papers from one of the plenaries of the conference “Fields of Knowledge”. The papers address the critical subject areas of development, history, literature, science and law, all of which have had distinct relationships with gender and feminism. In so doing, they demonstrate both the kinds of critique that feminism has made possible and a transformed understanding of the issues at stake.

REVIEW OF WOMEN’S STUDIESoctober 25, 2008 EPW Economic & Political Weekly50also of concern is that this national picture of relatively greater illiteracy among female workers is repeated in socially and eco-nomically developed states such as Tamil Nadu (Table 1). Feminist and women’s studies scholarship on the relatively larger numbers of women workers crowded in low paying jobs and in tasks desig-nated as unskilled explain why employment and education are moving in opposite directions for women.Using census data we have computed a sex ratio measure to understand the distance between the genders in literacy levels (Table 2). For the country as a whole, for every 1,000 literate males in the general population, 666 females are literate. But among workers, for every 1,000 literate male workers, only 234 female workers are literate. As we disaggregate the literate popu-lation figures, the starkness of the distance between the genders in the general population and among workers becomes sharper. Thus, in the category “graduate and above other than technical degree”, we have 490 females per 1,000 males in the general category; but in the worker population category this figure drops to 165. In the “technical degree or diploma equal to degree or post-graduate” category, the equivalent figures are 388 females in the general population and only 264 females among workers. A point to be noted here is the proportion of population returned as “non-workers” and the proportion of “literate non-workers” among males and females (Table 3, p 52). For the country as a whole, 48 per cent males have been returned as non-workers against 74 per cent females. Urban areas have a larger proportion of population returned as non-workers when compared to rural areas. However, unlike men, the rates of literacy for female non-workers are higher than the rates of literacy for females in the population in general, and in both rural and urban areas. The same is true for states such as Tamil Nadu, indicating and empha-sising that social and economic development are not necessarily gender just. Despite the fact that states like Tamil Nadu reveal “higher than national level” work participation rates for females, these rates are still far below the rates obtaining for males; further, rising levels of female literacy are not reflected in the literacy levels of the worker population. Development in the last decade then, has not been able to reverse the opposite directions in which literacy and employment have been moving for women at the national level, including in states whose economic and social indicators of development are well above the average.The 61st round of the National Sample Survey (NSS) on Employ-ment and Unemployment Situation in India covering the period July 2004-June 2005 provides another facet of the disproportion-ately greater education-employment divergence for women in the country. The NSS provides data on what it calls the education level specific worker population ratio, namely, the number of persons of age 15 years and above who are usually employed in a particular education category per 1,000 persons in that education category (Table 4, p 53). Table 4 also provides comparative information from earlier rounds. What we see are very low levels of the educated em-ployed among women; thus, for instance, in rural India, while 851 males per 1,000 males in the category “graduate and above” are employed, the equivalent figure for females is only 345. For urban India, the respective figures for thesameeducationalcategories are 795 and 290.Table 2: Gender Gap (Females Per 1,000 Males) in Literacy: General Population, Workers and Non-Workers: India and Tamil Nadu(2001)Total/ Rural/Urban Educational Level General Workers Non- PopulationPopulationWorkersIndia Total Total 933 463 1,436 Illiterate 1,392 1,008 1,664 Literate 666 234 1,253 of which Literate but below matric/secondary 731 274 1,201 Matric/secondary but below graduate 544 138 1,392 Technical diploma or certificate not equal to degree 264 190 429 Graduate and above other than technical degree 490 165 1,632 Technical degree or diploma equal to degree or post-graduate degree 388 264 750 Rural Total 946 559 1,367 Illiterate 1,408 1,093 1,653 Literate 619 275 1,095 of which Literate but below matric/secondary 684 313 1,094 Matric/secondary but below graduate 436 161 1,089 Technical diploma or certificate not equal to degree 262 208 376 Graduate and above other than technical degree 278 113 864 Technical degree or diploma equal to degree or postgraduate degree 274 203 519 Urban Total 900 211 1,606 Illiterate 1,324 529 1,702 Literate 760 151 1,556 of which Literate but below matric/secondary 855 152 1,458 Matric/secondary but below graduate 687 105 1,724 Technical diploma or certificate not equal to degree 266 177 473 Graduate and above other than technical degree 621 198 2,106 Technical degree or diploma equal to degree or postgraduate degree 439 293 838Tamil Nadu Total Total 987 540 1,596 Illiterate 1,547 1,355 1,704 Literate 777 319 1,537 of which Literate but below matric/secondary 848 373 1,454 Matric/secondary but below graduate 695 215 1,698 Technical diploma or certificate not equal to degree 144 80 256 Graduate and above other than technical degree 599 246 1,788 Technical degree or diploma equal to degree or postgraduate degree 507 368 787 Rural Total 992 695 1,421 Illiterate 1,564 1,476 1,654 Literate 717 392 1,272 of which Literate but below matric/secondary 781 441 1,250 Matric/secondary but below graduate 584 268 1,295 Technical diploma or certificate not equal to degree 107 67 162 Graduate and above other than technical degree 407 199 1,022 Technical degree or diploma equal to degree or postgraduate degree 467 390 594 Urban Total 982 333 1,800 Illiterate 1,514 993 1,776 Literate 841 239 1,812 of which Literate but below matric/secondary 941 268 1,711 Matric/secondary but below graduate 780 172 1,983 Technical diploma or certificate not equal to degree 165 86 321 Graduate and above other than technical degree 670 263 2,114 Technical degree or diploma equal to degree or postgraduatedegree 518 362 854Source: Census 2001.

REVIEW OF WOMEN’S STUDIESEconomic & Political Weekly EPW october 25, 200851capacity (measured in terms ofemploymentelasticity)came down to near zero; in the public sector, it has been negative in mostcases.Thus the major source of employment generation and labour absorption is the unorganised sector of the economy whoseemployment weightage is as high as 92 per cent of the total employed labour force [GOI 2001: 34; 2002: 2 and 26]. On examining all major sources of information, the commit-tees found that the rate of unemployment in India has increased significantly in 1993-94 and was above 7.3 per cent in 1999-2000 compared to 6 per cent in 1993-94 on current daily status (CDS) basis.2 The number of unemployed has increased from 20.13 million in 1993-94 to 26.58million in 1999-2000. Nearly 74 per cent of the unemployed are in rural areas, while 60 per cent of the unemployed are educated (higher secondary and above).A point that both the committees stress is the fact that an un-employment rate of 7.3 per cent should not be read as implying that those employed have “decent” jobs. One dimension of this “quality” is the level of income that the employment provides. That a large part of the employment generated in the economy provides very low levels of income is very evident from the fact that whereas unemployment, even according to the most expan-sive measure, namely the CDS measure was only 7.3 per cent, the percentage of population in poverty was as high as 26.1 per cent. Thus being employed need not necessarily enable an individual/household to rise above the poverty line. An equally important issue flagged by both the reports borne out of their exercise of disaggregating unemployment data is that the rate of unemployment is higher among the educated than among those with lower levels of education. Thus, according to one of the reports [GoI 2001], measuring unemployment rates on usual principal and subsidiary status (UPSS) basis, the unemploy-ment rates for the illiterate is as low as 0.2 per cent rising to 1.2 per cent for literates up to primary school and 3.3 per cent for middle school. For the category of “educated labour” (that is, secondary education and above), the rate was 7.1 per cent – morethan three times the unemployment rate on the UPSS basis for the population as a whole (ibid: 26). The starkness of the mis-match increases when the criterion of education is combined with age. While between 1993-94 and 1999-2000, the incidence of unemployment among educated youth, both for general and technical education, declined sharply, these rates for youth are still high at 14.8 per cent of secondary education and 23.7 per cent for all types of technical education (ibid: 28).The high rate of unemployment among educated youth is in many ways the core of the problem because it creates a sense of despair across a wide section of the population including not only the educated youth but also their parents and families. It needs to be emphasised that the aspirations of this group cannot be met just by creating any employment opportuni-ties…The problem of unemployment in this category can only be solved if high quality employment is created (ibid: 28-29, emphasis added). The regional disaggregation of unemployment data raises fur-ther issues of concern. The unemployment rate (on the CDS basis) 2 Failure to Generate EmploymentThe economic surveys presented to Parliament year after year just before the annual budget do not evaluate the effects of economic development and growth on the more material aspects of the well being of people, such as for example the quantum and nature of employment generated by economic growth. In an inte-resting exercise, Dev and Mooij (2002) have analysed the budget speeches of finance ministers starting from 1988-89. Among other things they find that all the 12 budget speeches analysed by them were generally silent about employment creation in the regular economy. While they found separate sections devoted to agriculture and industry, there was hardly any mention of employment and labour market policies, except of specific anti-poverty employment programmes. It is ironical that the first time that the labour market is specifically dis-cussed in a budget, it is to make the point that lay-offs, retrenchments and closures should be made easier… It would be no exaggeration to state that, as far as one can judge from these budget speeches, India in the 1990s had no employment policy… A new term brought into the budget speeches in 1999 is empowerment… This empowerment has nothing to do with changing power relations or a redistribution of productive assets (ibid: 52). Further, according to the authors, the post-1998 budgets reveal the implicit interpretation of poverty by the government. It [poverty] is a residual interpretation – in contrast to a relational one. Poverty is seen as something that can disappear with a capital injection. A relational interpretation, on the other hand, would hold that poverty is the result of social and economic relations: the poor are poor as a result of their position within the social and economic structure” (ibid: 52). What the above discussion reveals is that the disjunction be-tween development and employment generation is builtintothe basic conceptualisation of economic growth in the annual budgets of the government. In a different way the same point is reinforced in theReport of the Task Force on Employment Opportunities (2001) and the Reportofthe Special Group on Targeting 10 Million Employment Opportunities (2002). Both reports emphasise the fact that there has been a steady decline in the job creating capacity of the economy, which has accelerated since 1993-94. This decline needs to be noted in the contextofadecelerationin population growth, a significant de-celeration in labour force growth, but a significant acceleration in grossdomestic product (GDP) growth between1982-83and 1992-94andbetween1993-94 and 1999-2000 from 5.2 per cent to 6.7percentperannum. The reports point out that thegrowthin GDP is accompanied by a significant decline in labourintensity in production in almost all sectors, barring a few, andalsointhe aggregate; the labour intensity (measured as employmentelas-ticitytovalueadded)declined from 0.52 to 0.16 [GOI 2002: 24]. Asaresult of the decline in the labour intensity of production, employment growth fell to 1.07 per cent per annum (between 1993-94 and 1999-2000) from 2.7 per cent per annum in the past (that is, between 1983 and 1993-94) in spite of acceleration in the growth of domestic product from 5.2 per cent (between 1983 and 1994-94) to 6.7 per cent (between1993-94and1999-2000).This in turn means that the capacity of job creation per unit of out-put went down about three times compared to that in the 1980s and early 1990s. The organised sector’s employment generating We are grateful to Mary E John for help in putting together this edition of the Review of Women’s Studies – Ed

REVIEW OF WOMEN’S STUDIESoctober 25, 2008 EPW Economic & Political Weekly52is higher in high literacy states, almost 21 per cent in Kerala; next is West Bengal with 15 per cent followed by Tamil Nadu at 12 per cent oftheirrespective labourforce.Further in each of these states, the incidence of unemployment among the youth is even higher and more so for women. Among young women in Kerala, the unemployment rate is as high as 46 per cent, in West Bengal it is 39 per cent and nearly as high in Tamil Nadu. Once again this high incidence of unemployment needs to be juxtaposed against the positive aspect of increasing levels of education discernible among the younger age groups of the population.3The statistics discussed above relate to general education, which is not the same thing as possession of “marketa-ble skills”. At the same time it is not easy to quantify the level of skills in the labour force since such data are not readily available. However, in 1993-94 the National Sample Survey Organisa-tion of India conducted a survey where information on the possession of 30 different marketable skills by persons in the labour force was sought. The results of this component of the survey revealed that hardly 10 per cent of male workers and 6 per cent of female workers in rural areas possessed spe-cific marketable skills. The urban areas returned relatively better figures but still abysmally low by any yardstick – 19.6 per cent for male workers and 11 per cent for female workers.Both the committees emphasise the painful fact that the level of vocational4 skills in the labour force in India com-parespoorly with the position in other countries. Only 5 per cent of the Indian labour force in the age category 20-24 has vocational skills whereas the percentage in industrial countries is much higher, varying between 60 per cent and 80 per cent. It may be argued that in developing countries like India, economically productive skills are ac-quired not only in formal training/ education institutions but also through the family. But it also needs to be stressed that currently the traditional artisan classes are among the econom-ically poorest in the country. Table 3: Levels of Education : General Population, Workers and Non-Workers: India-Tamil Nadu(2001, in %)Total/Rural/Urban Educational Level General Population Workers Population Non-Workers Males Females MalesFemales MalesFemalesIndia Total Total 51.74 48.26 68.37 31.63 41.05 58.95 Illiterate 36.7654.8529.564.344.5251.59 Literate 63.24 45.15 70.535.7 55.48 48.41 ofwhich Literate but below matric/secondary 65.55 71.91 57.74 67.43 76.18 73.05 Matric/secondary but below graduate 22.6 18.46 26.52 15.6 17.27 19.19 Technical diploma or certificate not equal to degree 0.86 0.34 1.03 0.84 0.63 0.22 Graduate and above other than technical degree 6.5 4.78 8.79 6.21 3.4 4.42 Technical degree or diploma equal to degree or postgraduatedegree 1.080.631.41.580.650.39 Rural Total 51.39 48.61 64.15 35.85 42.25 57.75 Illiterate 41.4261.6734.7167.8648.7258.92 Literate 58.5838.3365.2932.1451.2841.08 ofwhich Literate but below matric/secondary 72.01 79.66 65.04 74.04 81.67 81.61 Matric/secondary but below graduate 19.39 13.67 23.48 13.7 13.73 13.66 Technical diploma or certificate not equal to degree 0.57 0.24 0.66 0.5 0.43 0.15 Graduate and above other than technical degree 3.75 1.69 5.03 2.06 1.97 1.56 Technical degree or diploma equal to degree or postgraduatedegree 0.510.220.670.50.270.13 Urban Total 52.62 47.38 82.55 17.45 38.37 61.63 Illiterate 24.9636.6815.9139.7934.2236.27 Literate 75.04 63.32 84.09 60.21 65.78 63.73 ofwhich Literate but below matric/secondary 52.78 59.42 42.94 43.05 65.66 61.5 Matric/secondary but below graduate 28.94 26.17 32.69 22.58 24.03 26.63 Technical diploma or certificate not equal to degree 1.45 0.51 1.78 2.08 1 0.3 Graduate and above other than technical degree 11.96 9.77 16.41 21.51 6.12 8.28 Technical degree or diploma equal to degree or postgraduatedegree 2.221.282.875.561.370.74Tamil Nadu Total Total 50.32 49.68 64.93 35.07 38.52 61.48 Illiterate 27.3642.8621.3353.535.5637.96 Literate 72.64 57.14 78.67 46.564.44 62.04 ofwhich Literate but below matric/secondary 59.59 65.04 53.52 62.48 69.66 65.92 Matric/secondary but below graduate 22.23 19.88 24.09 16.22 19.13 21.14 Technical diploma or certificate not equal to degree 1.68 0.31 1.71 0.43 1.63 0.27 Graduate and above other than technical degree 4.75 3.67 5.87 4.53 2.89 3.37 Technical degree or diploma equal to degree or postgraduatedegree 1.350.881.441.661.190.61 Rural Total 50.2 49.8 59 41 41.3 58.7 Illiterate 32.4951.2127.9359.3339.0845.48 Literate 67.51 48.79 72.07 40.67 60.92 54.52 ofwhich Literate but below matric/secondary 66.98 73.01 61.57 69.33 76.23 74.94 Matric/secondary but below graduate 18.69 15.22 20.52 14.01 15.56 15.85 Technical diploma or certificate not equal to degree 1.16 0.17 1.07 0.18 1.31 0.17 Graduate and above other than technical degree 2.48 1.41 2.93 1.49 1.7 1.36 Technical degree or diploma equal to degree or postgraduatedegree 0.590.380.580.580.60.28 Urban Total 50.46 49.54 75.01 24.99 35.71 64.29 Illiterate 20.8832.212.537.2331.4631.02 Literate 79.12 67.887.562.77 68.5468.98 ofwhich Literate but below matric/secondary 51.61 57.71 44.65 50.09 62.83 59.33 Matric/secondary but below graduate 26.04 24.16 28.03 20.21 22.84 25 Technical diploma or certificate not equal to degree 2.25 0.44 2.42 0.87 1.97 0.35 Graduate and above other than technical degree 7.21 5.74 9.12 10.03 4.14 4.83 Technical degree or diploma equal to degree or postgraduatedegree 2.171.342.393.621.80.85Source: Census 2001.

REVIEW OF WOMEN’S STUDIESEconomic & Political Weekly EPW october 25, 200853Table 4: Education Level Specific Worker-Population RatioGeneral Education Level Male Female 1993-94 1999-20002004-05 1993-94 1999-20002004-05 All Workers All Workers All Workers All Workers All Workers All Workers (1) (2) (3) (4) (5) (6) (7)Rural Not literate 918 895 892 540 513 550 Literate and upto primary 909 880 895 416 403 449 Middle 770 768 802 290 290 371 Secondary 728 737 732 258 257 305 Highersecondary 686 713 709 234 206 252Diploma/certificate* – 821 523 Graduate and above 834 836 851 366 310 345 All 864 841 846 486 452 485Urban Not literate 870 839 831 300 271 304 Literate and upto primary 850 830 855 203 177 234 Middle 723 732 760 131 129 161 Secondary 677 668 673 134 124 123 Highersecondary 607 608 608 147 124 129 Diploma/certificate* – 798 – – 486 Graduate and above 818 806 795 301 273 290 All 768 752 763 223 197 227* In NSS 50th and NSS 55th round surveys there was no separate code for diploma/certificate and as such the estimates of persons with general education level “diploma/certificate” could not be obtained separately and they were classified in the equivalent level of general education. In the 61st round survey, persons with general education level `diploma/certificate which were equivalent to below graduate level were identified separately.Source: Employment and Unemployment Situation in India, 2004-05. NSS 61st Round, NSSO, Ministry of Statistics and Programme Implementation, Report No 515, Part I, September 2006, GOI, p 82.3 Nature and Terms of ‘Inclusion in Development’TheReport on the Conditions of Work and Promotion of Liveli-hoods in the Unorganised Sector produced byNCEUS provides a graphic account of the terms in which the bulk of labour in the country “work”. The NCEUS report begins by demarcating the unorganised sector from unorganised workers, and the unorganised sector from the or-ganised sector. Based on these definitions it provides estimates of employment in the unorganised sector and estimates of unorganised workers as a whole. According to the report, as in January 2005, the total employment (principal plus subsidiary) in the Indian economy was 458 million, of which the unorganised sector accounted for 395 million or 86 per cent of total workers in 2004-05. Of the 395 million unorganised sector workers, agriculture accounted for 253 million and the remaining 142 million are employed in the non-agriculture sector. The report estimates that the increase in employment be-tween NSS 55th round (1999-2000) and NSS 61st round (2004-05) – of the order of 8.5 million in the organised sector – was entirely informal in nature, that is, without any job or social security. The share of the agricultural workforce (farmers plus agricultural labourers) in the total workforce has shown a steady decline in the last two decades; nevertheless, the proportion of those employed in agriculture is still large, namely, 56.6 per cent of the workforce as in 2004-05. The share of agricultural workers in the female workforce is very high, almost 73 per cent in 2004-05, when compared to the male agricultural workforce share of 49 per cent. What is disturbing about the numbers dependent on agriculture is what the NCEUS re-port terms as “lopsided structural transformation”, namely, a trans-formation where a declining share of income from the agricultural sector is not accompanied by an equivalent decline in employment in that sector. Other aspects characterising agricultural workers employed in a sector that is not economically dynamic include the fact that the agricultural workforce continues to remain at the bottom of the educational ladder including a considerable gender gap in education. Thus for instance, while 93 per cent of female agricul-tural labourers were either illiterate or had studied only up to primary level, for males the figure was 81 per cent in 2004-05. Further, the SCs and scheduled tribes (STs) are disproportionately represented in the agricultural labour population. While the male-female distribu-tion of agricultural labourers is almost similar for the SC population, the share of females is higher among the STs and other backward classes (OBCs). Overall, Muslim and upper caste women are less represented as agricultural labourers than their male counterparts. For the country as a whole, the incidence of child female agricul-tural labourers is higher than for child male agricultural labourers.Discussing at some length the conditions of work of women workers in non-agriculture, the NCEUS report notes the following: In 2004-05 only half of the total non-agricultural workforce and as little as one-third of women workers worked in conventionally designated workplaces. This implied that about 89 million men and only about 10 million women workers had a conventional place of work, either of their own or belonging to their employer. Besides, lack of a clear-cut employer-employee relationship in the case of the self-employed and outworkers/homeworkers, contributes in no small measure to unorganised workers being legally unrecognised as workers and therefore excluded from the purview of existing laws relating to minimum wages and social security, howsoever inadequate the latter may be.TheNCEUS report sums up the situation on women’s conditions of work in non-agriculture as follows: The overall picture that emerges is one of greater disadvantage for women workers in general and those belonging to rural as well asSCs/STs in particular. Apart from such inherited disadvantages as lower social position, a number of other factors also contribute to such a picture. These are their limited asset position, access to resources, and low levels of education and skill. Education and consequently some ability to ac-quire formal skills, could be a moderating force but this aspect presents a dismal picture. About 71 per cent women workers in India are either illiterates or have education only up to the primary level compared to 49 per cent for men. For rural women this is as high as 88 per cent but the highest percentage with such low education are the women work-ers belonging to SCs/STs, that is, 92 per cent [GOI2008: 92]. The macro quantitative picture pieced together by the NCEUS re-port on the nature of exclusion and terms of inclusion in develop-ment needs to be read along with the body of feminist research that has brought out the ways in which women’s gendered identity has been central to the national and international drive towards cheaper labour but higher productivity and profits over the last few decades. In the process this research has also highlighted the strategies through which governments, transnational corporations and even trade unions have managed to continually erode potential and ex-isting entitlements for women workers [Pearson 2004: 607]. Case studies of units in export processing zones (EPZs) across the globe, including in India [Razavi, Pearson and Danloy 2004; Swaminathan 2001] are good examples of this combined onslaught. In a bid to attract foreign investment, the conditions offered to foreign investors/buyers by developing countries ensure that these major employers are absolved of virtually all responsibility for the welfare of their workers, despite the fact that units in EPZs are formal units

REVIEW OF WOMEN’S STUDIESoctober 25, 2008 EPW Economic & Political Weekly54where national regulations stipulating wage levels, non-wage ben-efits, hours of work, safety regulations, etc, ought to apply. Apart from exemptions, these formal units have managed to protect them-selves also legally by not designating their women workers as “workers”, through a simple device of employing their workers through contractors, with only the latter on the employers’ pay roles.Amartya Sen’s arguments that increasing women’s freedom to work outside the home is crucial for increasing their freedom in do-mains such as the home, healthcare, reproductive control, etc, have been critically studied and commented upon by several feminist re-searchers. Feminists have critiqued the facile connections that are made between promoting women’s workforce participation and in-creasing their agency. Without decrying the need for increasing women’s paid employment outside the home, feminists have di-rected attention to a number of factors that disempower women despite wage work. As Koggel (2003) puts it succinctly, “The central question…thus becomes: is Sen’s account sufficiently discerning of the ways in which global forces of power and local systems of op-pression operate and interact in ways that limit women’s freedom and agency even when they have paid work?” (ibid: 167). Another important contribution of feminist analysis of women’s work is the futility of merely suggesting that better data be collected such that, through better data collection techniques and tools, women’s work as economic activity rather than leisure can be netted. The exam-ples of women not being counted as workers, of home-based work increasing women’s wage employment but not contributing to their participation in the public sphere or enhancing their freedom and/or agency in the private sphere clearly illustrate why simply improving the definition of work and the collection of data on la-bour is not enough. In different ways, feminist research on wom-en’s paid and unpaid work concretely demonstrates how econo-mies, production systems and households across the globe continue to be heavily subsidised by programmes, policies and practices that keep intact the phenomenon of devaluing and/or inferiorising women’s work. TheNCEUS reports are the latest among other official Indian documents,5 apart from the census and NSS reports, to recognise, (a) the levels of exclusion from development of sections of the female population; and (b) the peripheral nature of their inclu-sion in development. There is also official recognition of the fact that the situation calls for intervention. Unfortunately, however, calls for intervention as well policies enacted thus far have re-duced and relegated gender issues to the “social sector” leaving macro policies largely untouched. In the next section, we provide a concrete illustration of the manner in which erosion of rural livelihoods dependent on agriculture has progressively contributed in different ways to households being forced to take gendered decisions, leading to discrimination and the exclusion of women from the gains of development. This exclusion, in our opinion, cannot be addressed by welfare measures alone. 4 Observations from the Field: Social Sector vs Social Content of Macroeconomic Policies The conventional notion that growth powered by the secondary and tertiary sectors would generate sufficient employment so as to absorb surplus labour from the primary sector has not taken place; on the contrary, the primary sector continues to maintain its rank as the largest employer of the rural people even as income from the primary sector has declined over time. In an attempt to understand the nature of risks faced by the poorest among the poor and also their coping mechanisms, we undertook a “liveli-hood assessment survey” in several villages across Tamil Nadu between July and October 2004.6 The relevance of reproducing here the observations from the survey lie in the remarkable manner in which people on the ground, women in particular, in their own words linked their inability to make a transition to a better life because of the disjuncture they perceived between the macroeconomic issues of employment and growth on the one hand, and on the other, the low scale and poor quality of the otherwise “functioning” welfare institutions in their villages.The field visits provided crucial insights into how the vulnera-bility of the rural agricultural households had increased over time. At the outset, it needs to be recorded that across the villages, and among almost all sections of the agricultural labouring popula-tion, the risk and therefore vulnerability due to declining agri-cultural activities (the most important source of livelihood for those with and without land) had increased considerably. The villagers, women included, traced the decline in agricultural activities to a combination of factors: continuous failure of monsoons, depletion of groundwater, change in cropping patterns, changes in institutional patterns that govern agrarian relations, etc, all of which had combined to erode their livelihood base. A direct economic consequence of this combination of factors was the decline in the number of days of employment, hitting land-less agricultural labouring population the hardest. Most villages had no other major activity that could provide alternate employ-ment to agriculture. Landless households among the agriculture-based employment groups were the most vulnerable. They were forced to cope by cutting down the number of meals they took in a day, discontinuing schooling of their children, delays in seeking medical attention for their ailments, defaulting on repayment schedules on their loans, and/or become more indebted, thereby further increasing their vulnerability. The gender question that emerged here was the differential impact that this vulnerability held for men and women: while to some extent men ventured out in search of coolie work, at times even staying out of the village for days together, such options were not available to women. They had neither the resources nor the support system to enable them to make these search trips. At the same time, we realised that a resolution to this gender problem did not lie only (or not even) in enabling women to go out in search of coolie work, but in addressing the larger question of the erosion of the main source of their livelihood.A drop in and/or lack of income had other adverse fallouts. For example, women in one village pointed out that all children were not in school and further that there was considerable dropout at the middle and higher levels for one or several of the following reasons: deteriorating income standards meant that they were forced to pull out their children from classes that did not serve noon-meals and also because they could not meet other school-related expenditure such as travel to the secondary school outside the village, and schoolbooks. In some cases older children had to discontinue schooling in order to share household responsibilities while their

REVIEW OF WOMEN’S STUDIESEconomic & Political Weekly EPW october 25, 200855parents went out in search of work. In quite a few villages,girls’ education was especially constrained by the combination of limited income as well as poor transport. While parents expressed their willingness and desire to educate their daughtersfurther,they could not translate this desire in practice. Hence, beyond the eighth standard, the gender gap in educationallevelsbecomesstark. Another dimension of the gendered nature of the problem was the following: in almost all villages women clearly expressed the point that, while they were happy that some among their village children, girls as well as boys, had managed to study up to the 12th standard, they were very aware of the futility of being “educated” only up to the 12th standard. One, it was pointed out that pursuing education beyond the 12th standard was expensive even in government higher educational institutions and also at times non-accessible (because of non-fulfilment of eligibility criteria by these children). Not all could afford to educate all their children; forced to make a choice, the parents opted to spend on higher education and hostel accommo-dation for their boys rather than for girls. The parents had a reason for this gendered choice:7 opportunities for employment outside of the ones available in the village, namely, agriculture, were nil, unless the educated chose to go to large metros and towns. While in fact a few boys from these villages had found some service jobs in metros like Chennai (like lorry booking, cleaner jobs with transport companies, etc), educated girls were handicapped by lack of social support and economic opportunities, and were therefore confined to their households. Villages close to metros such as Chennai were sourced for adolescent girl labour to work in garment units, particularly in theEPZ; the latter organised pick up and drop services for such labour, but most villagers were reluctant to send their daughters.All sections of the population were extremely anxious about the uncertainty facing their children, educated or otherwise. Their hopes were shattered because of the realisation that their chil-dren had no future either in the “traditional” occupation of agri-culture, or in “modern” occupations, the latter requiring the kinds of qualifications that the village population could, as of now, ill afford. Women were frustrated that their work and earn-ings were not sufficient to enable their children to rise above a certain level of education and/or acquire some professional skill. This in turn implied that they could not get into better paying jobs – a necessary condition for reducing the insecurity of their lives.The institution of self-help groups (SHGs) (a much-touted women empowerment intervention policy of the government) had a pervasive presence in almost all the villages that we visited. However, what came out quite clearly was that the nature of activities and the ability ofSHGs to sustain these activities had a lot to do with the general level of economic activity in the village, the cohesiveness of the village population, plus the institutional support provided by the panchayats. The women in almost all the villages were also aware of the limits ofSHG-sponsored activities and realised that while their SHGs could tackle short-term contin-gencies and support them in times of crises, the SHGs were no alter-native to large-scale employment, good infrastructure and function-ing public institutions – in all of which the state has a primary role to play. During the discussions on the quantum, nature and adequacy of the government’s welfare policies aimed at different categories of people such as the poor, women, disabled, widows, destitutes, girl children, pregnant women, etc, the women were very emphatic that, while through their local bodies and/or SHGs they did source the different programmes and very often also ensured that the targeted population benefited from the welfare schemes, these ad hoc “schemes” did not address the two hard questions that they posed: one, restoration of their livelihoods, and two, what would enable their children to make a transition to a better life since their present levels of education had failed to do so. These questions made us re-examine the rationale of our “social sector/welfare policies” and its relationship to “economic development”. 5 Implications for ‘Engendering Development’ AgendaThe picture of gender gaps in achievement/outcomes as sketched above through analyses of secondary data relating to employ-ment and education has more often than not led to the institution of “welfare” measures ostensibly aimed at eliminating such dis-crimination and in also “empowering” the discriminated. The persistence of poverty and vulnerability despite the “social sec-tor” and despite any number of studies documenting the linkages and connections between macroeconomic policies and changes on the ground, has contributed very little to the way discrimina-tion is perceived at the policy level and/or in the manner data are gathered. There is in fact, at the policy or macroeconomic level, very little use for nuanced analysis (of the type provided in the case study from Tamil Nadu) of the underlying causes contribut-ing to gaps/discrimination, vulnerability and disempowerment. The official sanction that was accorded to conduct a time use survey [GoI 2000] to net the range of activities as well as the time spent on these activities by different members of selected house-holds across seven states of the country remains an exception and till date has not been repeated, despite official assurances to the contrary. For us, the data thrown up by the survey considerably substantiated and enriched the concept of the gender division of labour, a term centre staged by feminists to highlight, among other things, the assignment to women in particular the tasks of daily and generational reproduction. The term has, to put it mildly, revolutionised the concept of “work”, forced researchers, policymakers and data collecting agencies to rethink the categories of “productive” and “unproductive” work, and take cognisance of themultiple and simultaneous nature of much of women’s work particularly in rural areas. Gender analysis has not only enriched the definitions of activities but is also instrumental in foreground-ing the issue of the relationship between women’s participation in production and their status in the household. The significance of the recognition of the multiplicity and simultaneity of much of women’s work lies in the fact that it provides to a large extent an explanation of how households persist in the face of their impoverishment and diminishing income. The empirical picture presented in the case study above, in our opinion, goes further. It also engages with the question of why and how households are forced to take gendered decisions in their attempt to cope with the progressive and secular decline in the sources of their livelihoods. In the west, the discussion initiated by feminist economists in particular around what has come to be subsumed under “care labour” has in fact contributed significantly to three policy issues:

REVIEW OF WOMEN’S STUDIESoctober 25, 2008 EPW Economic & Political Weekly56pay equity, the valuation of non-market work and greater support for parental work [Folbre 1995: 74]. The inconclusive but ongoing nature of debates on these issues notwithstanding, feminism has challenged the patriarchal family, and helped establish new rights for women and children. At the same time feminists are aware that policies to recognise, compensate and provide paid leave for “care work” superimposed atop gendered labour market structures and gendered notions of care giving are likely to re-produce gender gaps in employment and the gender division of care work [Morgan and Zippel 2003: 73]. In developing societies and particularly in the rural areas of these societies, where the concentration of poor households is the highest, the notion of in-stitutional assistance for care work hardly exists. On the contrary, the decline in public investment in rural areas in general, the low investment in infrastructure relevant to alleviating household drudgery (such as fuel, water and sanitation), and the bias to-wards facilitating infrastructure for private investment (such as investment in roads but not in public transport) – on top of the decline in sources of livelihood means that the burden of keeping body and soul together falls disproportionately on women, the prime caregivers in society. While planned economic development of the Indian nation these last six decades has excluded larger numbers of women among others and/or included others in a peripheral manner, there are across the country, individuals and organisations, including some state agencies, who are involved in operationalising schemes that are making a difference to the lives of people on the ground, including women. The insights from some of these efforts at engendering development [Swaminathan and Jeyaranjan 2008] clearly point towards the range of measures that need to be taken simultaneously when addressing the intertwined issues of pov-erty, caste, class and gender. By the same token, such initiatives also reveal the risks, dangers and sacrifices that are inherent when the effort is not merely towards redistribution but towards restructuring of social, political and bureaucratic institutions and ways of functioning. Therefore scaling up such interventions to cover larger geographical areas and numbers of the marginal-ised immediately attracts bureaucratic and social opposition. While pockets of inclusive development models exist, the models themselves cannot be replicated through political or administra-tive fiat. Neither will a programme of political mobilisation and action alone suffice, since their sustainability needs to be prem-ised on access to and protection of sources of livelihoods. In other words, not only does the state need to be brought in, but the state needs to be made and held accountable to its citizens and to its promises. Notes 1 Despite the fact that the nation was assured that census data collected in 2001 would be made available in quick time because of the availability and access to state-of-the-art IT infrastructure, considerable amounts and crucial aspects of the data collected have yet to become accessible. Hence in this paper it has not been possible to un-dertake the level and depth of analysis done ear-lier (using the 1991 census data).2The National Sample Survey Organisation (NSSO) of India collects detailed information on the employment status of the population through large-scale, nationwide sample surveys in which individuals are categorised as employed or avail-able for work but not employed, using different criteria. Rates are calculated as percentages of the total labour force. The NSSO provides four different measures of employment and unemploy-ment, each of which captures different facets of the employment-unemployment situation. One of these is the current daily status (CDS). Based on the reported time disposition of the person on each day of the reference week, person-days in employment (unemployment) are aggregated to generate estimates of person-days in employ-ment/unemployment. The person-day unem-ployment rate is derived as the ratio of person-days in unemployment to the person-days in the labour force. This measure captures the within-week unemployment of those classified as employed on the weekly status. The CDS-measure of unemployment is widely agreed to be the one that most fully captures open unem-ployment in the country [Planning Commission, 2001: 15-16]. 3 Both the Planning Commission reports give a large number of tables containing state-wise data relating to the nature of employment generated, the level of unemployment disaggregated by sex, age and level of education. 4 For a description of what constitutes formal vocational education in India see, India, Plan-ning Commission, Report of the Task Force on Employment Opportunities, New Delhi, July 2001, p 129. 5 Earlier official documents include,Towards Equa-lity, Shram Shakti, and the Time Use Survey. 6 The ‘Livelihood Assessment Survey’ was undertaken in 2004 as part of the preparation for the World Bank’s Tamil Nadu Empowerment and Poverty Reduction Programme. The survey covered 3,864 households and a population of 16,325individuals spread over 11 villages in 10 districts of the state. Three of these 11 villagesare tribal villages. Forty per cent of the surveyed population is from the scheduled caste community, 17 per cent from scheduled tribes’ community and the rest from “others”. The male-female break-up of the sur-veyed population is 50.5 and 49.5 per cent respec-tively. Nearly 15 per cent of the surveyed house-holds are female-headed households (Livelihood Assessment Report, Institute of Development Alternatives, Chennai, December 2004, mimeo). 7 Nussbaum makes a similar observation: “{Thus}, many individual parents who have no objection to educating girls and boys on a basis of equality may be able to afford to educate only one of their children… In many cultural circumstances, exist-ing employment opportunities dictate that the one educated must be a boy because his overall employment opportunities are greater and educa-tion is a necessary passport to these. So the ne-glect of female education may be a matter of sur-vival for parents in many parts of the world” (2003: 340-41).ReferencesDev, Mahendra S and J Mooij (2002): ‘Social Sector Expenditures and Budgeting: An Analysis of Patterns and the Budget-Making Process in India in the 1990s’, Working Paper 43, Centre for Econo-mic and Social Studies, Hyderabad.Folbre, Nancy (1995): ‘Holding Hands at Midnight: The Paradox Report of the Time Use Survey of Caring Labour’,Feminist Economics, Vol 1, Number 1, pp 75-92.Government of India (2000): Central Statistical Organisation, Ministry of Statistics and Pro-gramme Implementation, New Delhi, April. – (2001): Report of the Task Force on Employment Opportunities,Planning Commission, New Delhi, July. – (2002): Special Group on Targeting Ten Million Employment Opportunities Per Year, Planning Commission, New Delhi, May. – (2008): Report on Conditions of Work and Promo-tion of Livelihoods in the Unorganised Sector, National Commission for Enterprises in the Unorganised Sector, Academic Foundation, New Delhi.Institute of Development Alternatives (2004):Liveli-hood Assessment Report, Chennai, December.Koggel, C M (2003): ‘Globalisation and Women’s Paid Work: Expanding Freedom?’,Feminist Economics, Vol 9, Nos 2-3, 163-83.Morgan, K J and K Zippel (2003): ‘Paid to Care: The Origins and Effects of Care Leave Policies in Western Europe’,Social Politics, Spring, 49-85. Nussbaum, Martha C (2003): ‘Women’s Education: A Global Challenge’, Signs: Journal of Women in Culture and Society, Vol 29, No 2, Winter, pp 325-55.Pearson, R (2004): ‘The Social Is Political’, Inter-national Feminist Journal of Politics, Vol 6, Number4, pp 603-22.Razavi, S, R Pearson and C Danloy (2004): Globalisa-tion, Export-Oriented Employment and Social Security: Gendered Connections, Palgrave Macmillan, UK.Swaminathan, Padmini (2001): ‘Labour-Intensive Industries but Units Without Workers: Where Will ILO’s Social Dialogue Begin?’ in A S Oberai, A Sivananthiran and C S Venkata Ratnam (eds), Labour Issues in Export Processing Zones in South Asia: Role of Social Dialogue, ILO, New Delhi, December. – (2002): ‘The Violence of Gender-Biased Develop-ment: Going Beyond Social and Demographic In-dicators’ in Karin Kapadia (ed), The Violence of Development: The Politics of Identity, Gender and Social Inequalities in India, Kali for Women, New Delhi.Swaminathan, Padmini and J Jeyaranjan (2008): ‘Mainstreaming Gender, Engendering Develop-ment: Reflections on a Case Study’, Economic & Political Weekly, Vol XLIII, No 17, April 26.