The making of a supervisor: Determinants and consequences of women’s transition into leadership under MGNREGA

Introduction

The Mahatma Gandhi National Rural Employment Guarantee Act, 2005, contained an internal ladder. Alongside the guarantee of a hundred days of unskilled manual work, the scheme created the position of worksite Mate -a supervisor drawn from among the workers themselves, responsible for maintaining the muster roll, recording attendance, measuring completed work and managing site discipline. The Mate is paid at the notified wage. She is not an officer of the state. She is a beneficiary who supervises other beneficiaries. This is the only position within the scheme that an ordinary participant can realistically attain, and in a programme whose workforce in Tamil Nadu is between eighty and ninety-two per cent female, it is overwhelmingly a position held by women. It is therefore a rare instance of a public employment programme generating supervisory authority for rural women from within its own operations rather than through a reservation or a training scheme layered on top. The empirical literature has almost nothing to say about it. Studies of MGNREGA and women's empowerment are numerous, and the Tamil Nadu literature in particular is well developed (Carswell & De Neve, 2013, 2014; Michael, 2021; Novotný et al., 2013) [1,2,3] . But that literature concentrates on participation, wages, and household outcomes. The transition from labourer to supervisor -who makes it, why, and with what consequence -has not been systematically examined. Finnis (2017) [4] supplies the closest antecedent in her account of how Self-Help Group membership converts into political voice in rural Tamil Nadu, but the Mate role itself remains unstudied. This paper addresses that gap. It asks three questions. First, how widely distributed is the Mate role among women beneficiaries, and does that distribution follow the statute or something else? Second, which characteristics predict whether a woman has held the position? Third, what is the association between having held it and her empowerment across social, political and composite measures?

Data and methods

Sample

The analysis draws on a structured interview survey of 360 women MGNREGA beneficiaries conducted in Krishnagiri and Dindigul districts of Tamil Nadu. A multi-stage design was used: four Community Development Blocks were purposively selected in each district to span agro-ecological and labour-market diversity -Thally, Kelamangalam, Uthangarai and Bargur in Krishnagiri; Athoor, Vedasandur, Dindigul and Reddiyarchathiram in Dindigul -followed by three Gram Panchayats per block and fifteen respondents per Panchayat, selected randomly from the muster rolls and job card register. Eligibility required current job card holding and at least thirty days of personal work under the scheme in the two preceding financial years. The design was realised exactly, with all twenty-four Panchayats returning fifteen respondents. The survey was administered in Tamil during 2024-25 by trained enumerators. It is supplemented here by sixteen key informant interviews with implementing functionaries -Block Development Officers, Assistant Engineers, Panchayat Secretaries, an Assistant Project Officer and serving Mates -conducted over a window extending into early 2026, and by eight focus group discussions, one per block.

Measures

Mate service is recorded as a binary indicator of whether the respondent has ever served as a worksite Mate. Social and political empowerment are measured by two Likert batteries scored 1 to 5: ten items on social empowerment (𝛼 = 0.560) and eight on political empowerment (𝛼 = 0.704). The social scale falls below the conventional reliability threshold, and results resting on it are reported with corresponding caution; the coefficient reflects genuine heterogeneity in the construct rather than measurement failure, since the items span capability, structural position and social connection, which do not covary strongly. A composite Women Empowerment Index was constructed from twelve indicators across five dimensions -economic, financial inclusion, intra-household decision-making, social and political -each normalised to the 0-1 range. Weights were derived by principal component analysis on the standardised indicator matrix, retaining components with eigenvalues exceeding unity and weighting each indicator's absolute loadings by the variance its component explains. The approach follows Kumar et al. (2021) [5] , whose index is grounded in Kabeer's (1999) [6] resources-agency-achievements framework. One caution governs the entire analysis. Mate service is not randomly assigned and this study includes no experimental variation. All associations reported are conditional on the covariates included; none is a causal estimate. Section 5 addresses the direction-of-causation problem explicitly.

How widely is the role distributed?

Of the 344 respondents for whom the item was recorded, 39 -11.3 per cent -had served as a worksite Mate. The aggregate figure conceals the paper's first substantive finding.

Who becomes a Mate?

The educational threshold

Mates differ from other respondents on every socio-demographic dimension measured, and the pattern identifies the selection criterion operating in practice. The contrast is stark at both ends of the distribution. Among non-Mates, 33.8 per cent have no formal education whatever; among Mates the figure is 2.6 per cent -a single respondent. At the other end, 43.6 per cent of Mates have completed higher secondary schooling against 8.9 per cent of non-Mates. Mates are also markedly younger, averaging 36.5 years against 46.5 (p < 0.001), and are nearly twice as likely to belong to a Self-Help Group (64.1 against 33.8 per cent; χ 2 = 11.93, p < 0.001).

Multivariate determinants

Logistic regression identifies which of these characteristics predict Mate service independently of the others. Self-Help Group membership, strongly associated with Mate service in bivariate comparison, ceases to be significant in the multivariate model (p = 0.725). The group effect operates through the educational composition of its membership rather than independently -Self-Help Groups recruit the better-schooled, and it is schooling that the Mate role requires.

The requirement is intelligible on its face. The Mate maintains muster rolls, records attendance through the National Mobile Monitoring System application, and measures completed work. The Panchayat Secretary at Mathur described the women Mates in his block as excelling at attendance recording through the mobile application and at measuring and allocating work quantities accurately. These are literate and increasingly digital tasks. The Block Development Officer at Uthangarai noted the corresponding limit, observing that while women Mates perform well in fieldwork, shortcomings appear in maintaining documentation.

The implication is uncomfortable and should be stated plainly. The scheme's one internal ladder is accessible principally to the most educated third of its participants. The third of respondents with no formal education are effectively excluded from it. Whatever empowerment the role confers is distributed towards those who began with the greatest advantage.

What difference does it make?

The empowerment differences between Mates and other respondents are large by any conventional standard. Panel B shows the scale means with confidence intervals that do not overlap.

The differences survive multivariate control. In regressions of each outcome on Mate service together with days received, years of participation, age, education, Self-Help Group membership, social category and district, the Mate coefficient remains large and highly significant: +0.314 on political empowerment (p < 0.001), +0.265 on social empowerment (p < 0.001), and +0.123 on the composite index (p < 0.001). Mate service is not a proxy for education; it carries an association of its own.

Selection or effect?

The direction of causation cannot be established from these data, and the paper does not claim to establish it. Two mechanisms are plausible and both are likely operating. The role may confer empowerment. A woman who supervises a worksite, deals routinely with officials, maintains records and allocates work acquires standing, capability and confidence through doing so.

Alternatively, the more empowered may be selected into the role, the administration identifying women who are already confident, educated and articulate. The finding that Mates are markedly better educated supports the selection mechanism; the finding that they do not differ in days worked or years of participation indicates that what distinguishes them is not their engagement with the scheme as labourers. What can be stated without resolving the question is that Mate service is the single strongest marker of empowerment in this dataset; that its distribution is determined by block-level administrative practice rather than by anything in the statute; and that the practice is demonstrably scalable, since one block administration has appointed 294 women to the role while three blocks in the neighbouring district have appointed none of the respondents surveyed.

The mechanism of appointment

The focus group in Thally block -where a third of respondents had served -described how the appointments there came about, and the account is the most directly actionable finding of this paper: "When we made a request to the Panchayat administration, women were appointed as SHG leaders and site supervisors (Mates)." -Focus group participant, Thally block, Krishnagiri The mechanism is a collective request to the Panchayat administration, granted. No statutory amendment, no additional entitlement and no budgetary provision was required. Thally is also the only block in which the focus group reported securing a formal Gram Sabha resolution on worksite facilities. The two achievements are of a piece: they describe a block in which women have learned to make collective demands of the local administration, and an administration that has learned to accede to them.

Aspiration is not the constraint

A natural objection is that Dindigul women may simply be less interested in supervisory office. The data contradict it. On the political empowerment item asking whether women like themselves should hold leadership roles in the village, Dindigul respondents score 3.66 against Krishnagiri's 3.49 (p < 0.05), and on likelihood of electoral participation 3.60 against 3.25 (p < 0.001). Dindigul women endorse women's leadership more strongly than Krishnagiri women and hold the office at roughly one-eleventh the rate. The gap is therefore not one of willingness. It is the difference between a district administration that has opened the pathway and one that has not.

Discussion and policy implications

Three conclusions follow. First, the Mate role is the strongest marker of empowerment identified in this dataset and the most unequally distributed. Effect sizes between 0.91 and 1.78 standard deviations, on a distribution ranging from a third of respondents in two blocks to none in three others, under an identical statute. Since neither days worked nor years of participation distinguishes Mates from other beneficiaries, and since the ad-ministrative testimony records no shortage of willing candidates, the distribution is a product of implementation practice. Second, the pathway is scalable at negligible cost. The Thally account identifies the mechanism precisely -a collective request, grantedand one block administration in this study has appointed 294 women.

A target for women Mates per block, monitored and published, would require no change to law, entitlement or budget. This is among the least expensive gender-responsive interventions available in rural employment policy. Third, and against the preceding, widening the pathway requires addressing what presently restricts it. Education is the dominant predictor and a single Mate in the entire sample comes from the third of respondents with no formal education. Simply instructing administrations to appoint more women would reproduce the existing selection, drawing further from the same educated stratum. Assisted or simplified digital attendance procedures, and targeted numeracy support for record-keeping and measurement, are the conditions under which the ladder becomes available to those who most need it. These findings acquire immediate relevance from the legislative transition. MGNREGA was replaced with effect from 1 July 2026 by the Viksit Bharat-Guarantee for Rozgar and Ajeevika Mission (Gramin) Act, 2025 [7] , which retains the Mate role and requires that not less than one-third of administrative cost be applied at Gram Panchayat level. The successor framework preserves the pathway. It does not address its unequal distribution, and nothing in it would prevent the pattern documented here -a third of women supervising in one block and none in another -from being reproduced unchanged.

Limitations

The design is cross-sectional and Mate service is not randomly assigned; the associations reported are not causal estimates, and Section 5.1 sets out the competing interpretations the data cannot adjudicate between. The social empowerment scale returns a reliability coefficient below the conventional threshold. Thirty-nine Mates is a modest subgroup, and the confidence intervals in Figure 4 for the smaller categories are correspondingly wide. The sample is confined to two districts of one State selected for contrast rather than representativeness. Finally, the study observes only women who have served, not those who sought the role and were passed over -a distinction that would sharpen the selection question considerably and that a purposive follow-up design could capture.

References

1. Carswell G, De Neve G. Women at the crossroads: A village study of MGNREGA implementation in Tamil Nadu. Economic & Political Weekly. 2013;48(52):82-93.
2. Carswell G, De Neve G. MGNREGA in Tamil Nadu: A story of success and transformation? Journal of Agrarian Change. 2014;14(4):564-585.
3. Michael SCA. MGNREGS and women empowerment in Tamil Nadu: Prospects and challenges amidst the existing socio-cultural reality. IAR Journal of Humanities and Social Science. 2021;2(1):1-13.
4. Finnis E. Collective action, envisioning the future and women's self-help groups: A case study from South India. Indian Journal of Gender Studies. 2017;24(1):1-23.
5. Kumar N, Raghunathan K, Arrieta A, Jilani A, Pandey S. The power of the collective empowers women: Evidence from selfhelp groups in India. World Development. 2021;146:105579.
6. Kabeer N. Resources, agency, achievements: Reflections on the measurement of women's empowerment. Development and Change. 1999;30(3):435-464.
7. Government of India. The Viksit Bharat-Guarantee for Rozgar and Ajeevika Mission (Gramin) Act, 2025. Ministry of Law and Justice. 2025.
8. Khera R, Nayak N. Women workers and perceptions of the National Rural Employment Guarantee Act. Economic & Political Weekly. 2009;44(43):49-57.
9. Narayanan S, Das U. Women participation and rationing in the Employment Guarantee Scheme. Economic & Political Weekly. 2014;49(46):46-53.
10. Novotný J, Kubelková J, Joseph V. A multi-dimensional analysis of the impacts of the Mahatma Gandhi National Rural Employment Guarantee Scheme: A tale from Tamil Nadu. Singapore Journal of Tropical Geography. 2013;34(3):322-341.
11. Pankaj A, Tankha R. Empowerment effects of the NREGS on women workers: A study in four states. Economic & Political Weekly. 2010;45(30):45-55.
12. Zaidi M, Chigateri S. Making Mahatma Gandhi National Rural Employment Guarantee Act (MGNREGA) more care-responsive: Programmatic notes for Women's Economic Empowerment policy and programming. Institute of Development Studies and Institute of Social Studies Trust; 2017.
13. How to cite this article: Kumar D, Nakkeeran K. The making of a supervisor: Determinants and consequences of women's transition into leadership under MGNREGA. International Journal of Agriculture Extension and Social Development. 2026;9(10):247252. doi: https://www.doi.org/10.33545/26180723.2026.v9.i10d. 4390