The role of informal caste networks in public service delivery

  • Blog Post Date 06 May, 2013
  • Articles
  • Print Page
Author Image

Sheetal Sekhri

University of Virginia

Large public programmes designed at the national level are often undermined by corruption at the local level. This column proposes local level monitoring and enforcement through informal networks as a potential low-cost solution. It shows how caste networks can facilitate monitoring and enforcement in the Targeted Public Distribution System, and infers implications of these findings for the food security bill.

In India, as well in other developing countries ,there is evidence that the effectiveness of some vast public programmes designed at the national level has been consistently undermined by serious corruption at the local level. Consequently, the intended beneficiaries fail to see benefits from programs that draw significant amounts of government resources.A pressing policy question – particularly in an era that increasingly emphasises decentralised delivery of public services – is how to mitigate this corruption and ensure that services reach their intended targets. Improved monitoring and enforcement from higher levels of government may prove effective, but may be too costly in many circumstances, even with improved technology. Re-designing programmes to limit the incentives for corruption among those delivering the services is an ideal solution, but may be politically difficult and may still leave important amounts of discretion with local agents, whether private or public. To address these issues, informal networks may be one potential solution.

National-level monitoring and enforcement can effectively discipline local authorities in certain settings (Olken 2007). But, this is not always forthcoming or cost-effective, and implementation often requires some accountability at the local level. Simple technological solutions, coupled with strictly enforced changes in incentive structures, can overcome obstacles of cost (Duflo et al 2012). However, in some cases, implementing the necessary technology or enforcing the proper incentive structures may be too expensive. In these cases, finding low-cost, local-level monitoring and enforcement mechanisms becomes more pressing. Bjorkman and Svensson (2009) demonstrate that community-level disciplinary actions, encouraged by NGOs, effectively improved service delivery by public health providers in the context of Uganda.We examine how informal monitoring and enforcement mechanisms operate and what is the resulting benefit in terms of improvement in public service delivery.

The context

In a recent study, we focus on exploring the impact of informal caste networks in the efficient delivery of India’s Public Distribution System (PDS) (Nagavarapu and Sekhri 2012). The PDS is India’s largest response to the persistence of malnutrition. The system distributes food grains and other goods through over 450,000 Fair Price Shops, where households are entitled to purchase goods at below market rates from a locally appointed shopkeeper. The subsidy for grain is primarily targeted toward below poverty line (BPL) households, where BPL status is determined by local elected officials. The scale of the programme is massive - the food subsidy accounts for approximately 1.3% of India’s GDP.At the same time, leakage in the system is persistently high. Recent estimates for 2007-2008 place the fraction of grain diverted by shopkeepers and others to the black market at an astounding 43.9% across India, which is actually an improvement over previous years (Khera 2011).

The programme faces a fundamental problem of monitoring and enforcement. With limited chances of being caught and punished, shopkeepers and their government suppliers face clear incentives to black market goods at the market price rather than sell them at the mandated below-market price. Improving formal monitoring and enforcement would be helpful, but it can be costly in terms of time and money. The push towards a national biometric identification system (Unique Identification (UID)1will hopefully provide the government the ability to detect some forms of corruption, but it may not address all forms of corruption and shopkeepers may find ways to work around technological solutions. Therefore, this is a setting where informal networks may play an important role.

Exploring the role of caste networks

Specifically, we explore the role that caste networks can play by comparing the traditionally disadvantaged members of the Hindu caste system, Scheduled Castes (SCs), to other castes. We examine household survey data from Uttar Pradesh in 1997, the year when the targeted version of the PDS (targeted toward BPL households) called TPDS, was introduced. Under TPDS, subsidies for grain were targeted primarily to BPL households, but subsidies for sugar and kerosene were available to all households. A distinct pattern emerges in the data - SC households have higher take-up of goods when they face SC shopkeepers, but this is only the case for grain and not for sugar and kerosene. We establish that this pattern is not explained by basic socio-economic differences across the castes, and not explained by differences across villages that depend on the caste of the shopkeeper2.

We show that the role of SC networks in monitoring and enforcement can explain the patterns we observe.Why should the patterns be different for sugar and kerosene versus grain, and why should this be true for SC households only? Sugar and kerosene differed from grain in 1997 because everyone in a village was entitled to sugar and kerosene, but not to grain. This creates a situation where information about the activities of the shopkeeper is likely to be widely available when it comes to sugar and kerosene, but harder to come by when it comes to grain. In a sense, everyone is on the look-out for sugar and kerosene, but only BPL households – a fraction of total households – are on the look-out for grain. Moreover, in 1997 BPL cards in Uttar Pradesh were concentrated among SC households, so SCs had a substantially larger interest in the shopkeeper’s treatment of grain than other households. With the additional, plausible assumption that having a shopkeeper of one’s own caste gives someone the ability to find out more information about him, these facts can explain why SC households benefit from SC shopkeepers, but only for grain – a good where the value of information is high since it is harder to come by.

We use the 1997 household survey to show indirect evidence that our hypothesis that caste networks facilitate monitoring and enforcement in TPDS is true.In an IGC fundedfollow - up survey, we directly asked 300 survey respondents (both SCs and non-SCs) about the role of caste networks in the provision of information about the PDS and the enforcement of discipline on the shopkeeper.We find strong evidence that caste networks facilitate monitoring and enforcement. In our survey, respondents reported that people of own caste provide information about availability of goods to own caste peers. It is easier to get information about availability of goods if the shop keeper is from own caste, and shop keepers are chided for any wrongdoing, if proven, by excluding them from caste based assistance like loans.

While this is encouraging in terms of understanding why informal networks may matter, there is a remaining question involving how much they matter. Our findings suggest that the value of this monitoring is large, approximately 20% of the total grain subsidy value3.

What does this mean for PDS policy going forward?

Our findings have a few important implications for policy design. First, there is a chance that increasing the value of the food subsidy – as envisioned in the National Food Security Bill – could actually lead to fewer households taking up goods. This is because, if monitoring is not increased sufficiently, a larger subsidy gives shopkeepers a larger incentive to black market goods. Second, expanding the number of people entitled to the PDS goods – another element of the National Food Security Bill - could create a larger swathe of people monitoring the shopkeeper and punishing illegitimate activities, and have positive effects on take-up.

However, these two opposing effects on welfare from increased generosity of the programme can tilt the balance perversely. On one hand the demand will go up, but on the other hand black market incentives will also go up. Our findings indicate that welfare would fall due to the dominance of the negative effects.

Third, informal networks can potentially channel information and provide enforcement. Community based monitoring can be a cost-effective solution to improve the take-up of the program.Fourth, the current PDS system is designed in such a way that there are strong incentives to black market goods. Considering re-designs of the system – with food vouchers, for instance – may ultimately be a better solution for households, though informal networks could still be useful in ensuring that households get the vouchers they are entitled to. However, more research is required to examine the effects of vouchers before making any quantum leaps.

This column is based on Nagavarapu, S. and S. Sekhri (2012), “Informal Monitoring Mechanisms in Public Service Delivery: Evidence from the Public Distribution System in India”. Authors wish to thank Frank Wolack for his extremely useful comments and suggestions.


  1. 12 digit individual identification number issued by the Unique Identification Authority of India (UIDAI) on behalf of the Government of India which serves as a proof of identity and residence through India
  2. Please refer to Nagavarapu, S. and S. Sekhri (2012), “Informal Monitoring Mechanisms in Public Service Delivery: Evidence from the Public Distribution System in India” for details.
  3. The final part of our paper provides a detailed economic model of the interaction between the households and the shopkeepers, uses the 1997 data to estimate the model, and then quantifies the value of the monitoring provided by the caste network to households.

Watch a video of Sheetal Sekhri presenting the underlying research at the 3rd IGC ISI India Development Policy Conference (July 2012; New Delhi)

Further Reading

  • Afridi, Farzana (2008), "Can Community Monitoring Improve the Accountability of Public Officials", Economic and Political Weekly, 18 October
  • Banerjee, Abhijit, Esther Duflo, and Rachel Glennerster (2008), "Putting a Band-Aid on a Corpse: Incentives for Nurses in the Indian Public Health Care System", Journal of the European Economic Association, 6(2-3):487-500, April-May
  • Bjorkman, Martina and Jacob Svensson (2009), "Power to the People: Evidence from a Randomized Field Experiment on Community-Based Monitoring in Uganda", The Quarterly Journal of Economics, 124(2), 735-769
  • Duflo, E., R. Hanna, and S. P. Ryan (2012), “Incentives Work: Getting Teachers to Come to School, "American Economic Review, 102(4), 1241-1278
  • Khera, R. (2011), “Trends in Diversion of Grain from the Public Distribution System," Economic and Political Weekly, 46(21), 106-114
  • Niehaus, P., and S. Sukhtankar (2013), “Corruption Dynamics: The Golden Goose Effect, " American Economic Journal: Economic Policy, Forthcoming.
  • Olken, B. A. (2007), “Monitoring Corruption: Evidence from a Field Experiment in Indonesia, "Journal of Political Economy, 115(2), 200-249
No comments yet
Join the conversation
Captcha Captcha Reload

Comments will be held for moderation. Your contact information will not be made public.

Related content

Sign up to our newsletter