The politics and bureaucratics of rural public works: Maharashtra's employment guaranteed scheme

Abstract

The Maharashtra Employment Guarantee Scheme stands out among rural public works programmes in developing countries for its size, longevity, the generosity of its funding arrangements, and the political sophistication of its design. Its mission is highly ambitious: to supply employment flexibly and rapidly by opening and closing public works in response to local, unpredictable weather variations in a poor agrarian economy. We explore the political factors that account for changes in its performance over more than 30 years, and identify the political lessons for the design of similar programmes elsewhere.

Acknowledgments

This paper is the product of a research project on the Employment Guarantee Scheme conducted under the aegis of the Centre for the Future State at the Institute of Development Studies. We are most directly indebted to our co-researchers at the University of Pune and the Institute of Development Studies: Anurekha Chari, Anuradha Joshi, Sujata Patel, Sanjay Savale and Shaji Joseph. Ashwini Kulkarni has helped us with field research. Aruna Bagchee provided the most illuminating commentary in the light of her experience as a member of the Maharashtra cadre of the Indian Administrative Service, implementing the EGS in the 1970s. Participants at workshops held in Pune in April 2003 and in Delhi, by the Centre for the Study of Law and Governance, Jawahrlal Nehru University (JNU), in April 2005, provided much useful feedback. We are especially grateful to Ravi Srivastava. Two anonymous referees provided very helpful comments on an earlier draft.

Notes

1. There is a very large literature on the EGS. Much of it is formal, descriptive, reliant on a few aggregate official data series, and didactic. There has been surprisingly little detailed investigative field research. Echeverri-Gent (1988, 1993), Herring and Edwards (1983) and Lieberman (1984, 1985) are honourable exceptions. They provide detailed insights into the politics and administration of the Scheme in western Maharashtra in its earlier days. The results of the only large-scale reliable independent survey of the EGS (Sathe, 1991) – conducted in two rounds in 1987–88 and 1988–89 in seven of the main EGS districts of western Maharashtra – exist only in typescript form. We have used them extensively. Researchers based in Mumbai (see Acharya and Panwalkar, 1988) often have tended to conduct their EGS research in Thane district, adjacent to Mumbai. Thane is highly atypical of the state: it has a large, vulnerable tribal population; is increasingly industrialised; and has never been a significant site for EGS work (Table 2). Many of the published journal articles purporting to be about the EGS are based on a time series study undertaken by ICRISAT in 40 households in precisely two villages in Maharashtra state (see Gaiha 1996a, b). In preparing this chapter we have depended, in addition to the statistical work reported in Moore and Jadhav (2004), on: an extensive reading of the literature; the research of our co-authors in this book; and on various interviews with politicians, public officials, trades unionists and EGS workers conducted in western Maharashtra between 1999 and 2003.

2. For example, Lipton (1996: 43–4) cites a number of countries in which three to 30 per cent of the workforce were at some point engaged in labour intensive public works.

3. In fact, the creators of the EGS conceived it as a pillar for the construction of a welfare state. The opening sentence of the first resolution of the Maharashtra State Assembly relating to the Scheme reads as follows: ‘In all countries believing in Welfare State it is considered desirable to guarantee employment to all citizens who need it in one for or other’ (Government of Maharashtra, 1969: 1).

4. EGS accounted for an average of 19 per cent of the capital spending of the state government during 1984–85 to 1988–89. It has since declined both absolutely and relatively, and in 1997–98 the figure was 7 per cent. The figures on EGS spending are from the Government of Maharashtra (1998). Those on total development expenditure are from the Bulletin of the Reserve Bank of India, various years.

5. We have been told by senior public servants in Maharashtra that a wise official concerned for his or her career and reputation might even avoid commissioning labour intensive public works because others will automatically assume they are doing this in order to make money for themselves.

6. Stone-breaking, to produce metal for road making or other construction work, is sometimes viewed as classic emergency work. In fact, in Maharashtra, stone-breaking is the speciality and sometimes the monopoly of particular caste or tribal groups. Except in extreme circumstances, such as those of the 1972–74 drought, people from other groups will rarely be employed on this work.

7. For evidence, see Echeverri-Gent (1988: 1301).

8. The best analytical description of these arrangements is by Echeverri-Gent (1993: 94–125).

9. These legal provisions help provide effective voice and power to jobseekers and to the politicians and social movements that sometimes organise them. The ways in which this works are however quite complex. The unemployment allowance has never been paid. Government officers have been under clear instructions never to get themselves into a situation where this allowance had to be paid. This provides them with a strong motivation to respond to the demand for work (Aruna Bagchee, private communication).

10. Or five kilometres in hilly areas.

11. The only reliable general evidence we have, from Sathe's 1981 survey of EGS workers in seven ‘prime’ EGS districts, is encouraging: on average, work sites were 2.4 kilometres from home (Sathe, 1991: 363–4).

12. There are in addition EGS Committees at district and block (sub-district) levels.

13. Echeverri-Gent (1993: 114–15) also explains that, because they were forbidden from sitting on elected local councils, the Members of the Maharashtra Legislative Assembly were not keen on transferring authority to that level.

14. After describing the various facilities to which EGS workers are nominally entitled – drinking water, shelter, first aid, child care, goggles for stone breakers, allowances for the use of own implements or for travel to work in excess of eight kilometres, maternity leave, disability compensation, clear explanation of the calculation of work norms, written records of output for each worker, etc – Sathe comments that ‘Essentially, these operational details have found a place in the implementation whenever the local labourers could articulate their demands. Otherwise in most places these rules and provisions often go by default’ (Sathe, 1991: 61).

15. In the early years of the Scheme, local officials were supposed to maintain a permanent register of local jobseekers. Although registration formally was the basis of any claim to EGS employment, more direct routes often were used to make claims, and, from early on, the state government made it clear that lack of registration should not be a basis for denying employment (Sathe, 1991: 40). The registration system appears formally to have been abolished in the 1980s. Yet it still appears in the official account of the EGS. In 1999, one of us was handed a brief undated note in the Planning Department in Mumbai (‘Employment Guarantee Scheme; a Note’) including the claim that ‘The Talati and Gramsavak [village level officers] are the authorities who register the EGS work demands of the villagers, provide receipt of registration and identity cards to the workers’. In the same year, we were shown an EGS registration book by a village Talati in Ahmednagar district in 1999, with entries for 1993 and 1997. And the District Collector talked of registration as an active ingredient of EGS.

16. Two thirds of EGS labourers interviewed in 1987–88 asserted the existence of corruption (Sathe, 1991: 14).

17. Page, like many of the leading figures in the socialist movement in Maharashtra, was a Brahmin (Sujata Patel, private communication). It was partly for that reason that his role in a polity dominated electorally by the Maratha caste was that of an insider rather than an elected politician.

18. During the drought, the political elite in the state committed themselves, in both word and deed, to the goal of avoiding any starvation deaths (Aruna Bagchee, private communication).

19. Marathas accounted for 36 per cent of the members of the Maharashtra Legislative Assembly (MLAs) selected in the elections of 1978, 1980 and 1985. They came dominantly from the Western Maharashtra and Marathwada regions. A further 7 per cent of MLAs were from the Kunbi caste, and predominantly from the Vidarbha region in the eastern part of the state (Thite, 1996).

20. According to the 1981 census, 48 per cent of the rural labour force of Maharashtra were cultivators, 35 per cent were agricultural labourers, and 17 per cent fell into other occupational categories (Sathe, 1991: 22).

21. Sathe's (1991) survey is especially useful on these questions of which kinds of people benefited from the EGS in the major EGS/rainshadow/Maratha districts. They were distinctly poor and in that sense deserving (Sathe 1991: 246–7, 371). He does however explain in some detail how the ‘advanced castes’ (i.e., Marathas) tended to rely on EGS employment, while members of Scheduled Castes and Scheduled Tribes were more likely to migrate in search of work (Sathe, 1991: 192–3, 245). There is extensive discussion of problems faced by EGS workers, but the proportion of respondents reporting ‘harassment’ or related problems was the same for women and men. And his respondents generally felt that EGS work was especially useful for women (Sathe, 1991: 282–3). A small survey sponsored by the ILO in three districts in 1978 is especially interesting in relation to the gender issue: the authors suggest that the proportion of females among EGS workers is appreciably higher than is recorded in the official statistics (International Labour Organisation, 1979: 8, 56–7). Dev (1995: 121) summarises the results of nine surveys of EGS employment completed before 1990. In six of these nine cases, the number of EGS workers from households of small and marginal farmers exceeded those from landless households.

22. Half the EGS fund is financed by a specific set of taxes, of which by far the most important is the Professions Tax, which in practice is paid almost entirely by people employed in Mumbai. The other hypothecated revenues are surcharges on other taxes. For details, see Herring and Edwards (1983: 584–6).

23. However, the Scheme inevitably did have adverse effects for rural employers in some instances (Echeverri-Gent, 1993: 112).

24. EGS earnings are pegged to the official minimum agricultural wage. Actual earnings depend on the amount of earth or rock that is shifted. The norms are supposed to be set such that a normal day's work of seven hours will generate the official minimum agricultural wage. For some details of how wage-setting sometimes operated in practice, see Herring and Edwards (1983: 580–613).

25. Employees of the Forestry Department; workers on government farms; village level officials etc. (Brahme and Upadhyaya, 1979: 179–80).

26. This was the arrangement made when the EGS was passed into law in 1977.

27. Many members of the Indian Administrative Service who have served at district level can talk fluently about the EGS. However, their tenures in post are typically short. Our field investigations suggest that, today at least their knowledge of the programme often seems shallow.

28. There have been a series of internal policy discussions within successive state governments about reforming the EGS in some more or less radical fashion.

29. We might have added to this list the fact that the EGS is recurrently denounced for corruption in the Maharashtra press. The implications are however difficult to interpret. There is corruption. In particular, revelations made about Dhule district in the 1990s by Arun Bhatia, a member of the Indian Administrative Service, received wide coverage (Echeverri-Gent, 1993: 110–11). Research by Sanjay Savale in Nasik district in 2003 also indicates some murky stories (Savale, 2004). However, corruption is inevitable in public works, and there is a widespread view that it has been relatively low in the EGS. Arun Bhatia campaigned against corruption with an estimate that it accounted for at least 13.5 per cent of EGS expenditures in Dhule district in 1980–81 (Echeverri-Gent, 1993: 110). By contrast, it is not difficult in India to find politicians and public servants who will explain in detail a set of standard ‘commissions’ on public sector contracts that will collectively amount to at least 40 per cent of total expenditure.

30. Half of the EGS labourers interviewed in 1987–88 by Sathe's team believed that the demand for EGS work from male labourers was declining, principally because of the growing availability of alternative employment (Sathe, 1991: 13). This interpretation fits closely with the prescription that the EGS should be reformed to make it a vehicle for training unskilled workers to do more skilled jobs.

31. The first argument is the more plausible, especially in relation to the original prime focus of EGS – irrigation. Road works feature increasingly in the EGS portfolio (Herring and Edwards, 1983: 579). The need and demand for rural road improvements are considerable, but the expenditure on skilled labour and materials is high relative to the unskilled labour component. If the objective is to generate employment for unskilled labour, road works are not very suitable.

32. The trades unionists and public servants we have interviewed indicate that trades unions become active in relation to EGS only when large workforces are assembled at individual sites.

33. Irrigation work accounted for 78 per cent of EGS expenditures in 1974–75, 20 per cent in 1987–88 (Sathe, 1991: 76), and only 15 per cent in 1990–91 to 2000–01 (Table 5). Members of Scheduled Tribes, who have become increasingly significant in the EGS workforce, tend to prefer stone-breaking and road construction work. Government has responded to their needs (Echeverri-Gent, 1988: 1308).

34. Since the Mandal Commission of 1990, the Kunbis have been recognised as separate caste entitled to the job and educational preferences attached to the status of Other Backward Caste (OBC). Whereas upwardly mobile Kunbis previously had often tried to become accepted as Marathas, the trend is now more likely to be in the opposite direction, with some Marathas seeking an OBC identity as Kunbis. In addition, there is probably also a link with the fact that the Maratha political-cum-economic elite have shifted their own economic interests away from food crop agriculture and rural areas toward urban and export enterprise. They no longer have or perceive much interest in the rural poor (Jadhav, 2004).

35. The Mumbai-based Shiv Sena party did well in the 1988 elections to district councils (Zilla Parishads), and, in a coalition with the BJP, controlled the state government between 1995 and 1999. That coalition government diverted many of the staff of the Planning Department, which managed the EGS centrally, into new employment programmes for urban areas (Aruna Bagchee, private communication).

36. The sugar cooperatives, that performed similar political roles to the EGS, were also economically dynamic in earlier decades. Their recent decline (Baburao Baviskar, private communication), like that of the EGS, reflects the crumbling of the hegemony of a distinct and organised Maratha-led political bloc.

37. Aruna Bagchee, private communication.

38. Typical of public policy in India generally, there are actually a number of schemes through which government agencies can provide direct cash subsidies for well-drilling and water supply projects aimed at increasing horticultural production. Our fieldwork indicates that, here as in other domains, government officers often splice together subsidy packages that are nominally from distinct programmes.

39. None of the figures on EGS employment are very reliable. The official series are actually based on estimates from recorded attendance at worksites on the last two days of each month.

40. It would be equally plausible to suggest that this new use of EGS funds would increase political support for the Scheme.

41. However, the figures might be misleading. In a private communication, Aruna Bagchee explained that government officers sometimes kept EGS projects formally incomplete, even if close to actual completion, so that they could easily meet small-scale local demands for employment without going through the process of opening new worksites.

42. The fact that the Government of Maharashtra pursued this policy of suppressing demand for EGS work is now beyond doubt. Some of the administrators involved are now willing to be frank about this. In 2003, one of us encountered a former EGS worker who had since made a career in Nasik district working as a muster clerk in a government construction agency. He explained how, for quite some years, officers at the lower levels of the public service had put less effort into promoting the EGS work opportunities than previously. He cited in particular the discontinuance of the practice of sending drummers around the villages announcing the opening of new worksites, and a shift from measuring work accomplished (and submitting payment requests) from a weekly to a fortnightly cycle. Ravallion and colleagues undertook quantitative simulations of what would have happened to EGS expenditures had there been no clamp down by government at the end of the 1980s, and conclude that the clamp down substantially reduced demand (Ravallion et al., 1993).

43. It follows that EGS employment is concentrated in the dry season. For some statistics on this, see Dev (1996: 243).

44. Echeverri-Gent commented in 1988 that despite ‘extensive corruption and the relative autonomy of the technical departments, implementation of the EGS remains surprisingly responsive to the demands of the rural poor’ (Echeverri-Gent, 1988: 1303–4).

45. See, for example, Ravallion and Dutt (1995), Lipton (1996), and Ravallion (1999).

46. A recent study of the Employment Assurance Scheme in five districts in the states of Bihar, Jharkhand and west Bengal finds a much higher knowledge of and uptake of the Scheme in the west Bengal district where the rural population have been mobilised by the Communist Party of India (Marxist) (Véron et al., 2003).