LEAD AUTHOR: Savita Bailur

Photo of people sitting, talking, with sheets of paper

Clients with intermediaries

Doddaraghu: [When people lose their ration card] I give them ration for about two months, but at my own personal risk. I will anyway have their number. In my register, I will also note down the fact that their card has not been produced. At the same time, I will also inform others ‘in case someone else is in possession of such and such a card, please do return it’.… Should the ration card still remain untraceable, we advise our customers to apply for a new ration card and we authorize the letter of request. This letter has to be shown at the Taluk office.

Researcher: Do they take the letter of request for reissue to the Taluk office or do you?

Doddaraghu: They are supposed to take it. But if they say it is difficult for them, then I help them by taking it myself.

Doddaraghu, ration dealer, Garudahalli

Doddaraghu has been a ration dealer for 15 years. He applied for the position after he found life as a farmer in Garudahalli too difficult, particularly with increasing droughts. The exchange above shows the dual roles he plays: he is at once an employee of the state with a formal role and responsibilities for following specific procedures, yet he is also a member of the community in which he has been living since birth, and has personal relationships that he continues to manage socially.

People in Doddaraghu’s community need his help because the PDS system, like all complex “sociotechnical” systems, cannot account for every potential scenario and every individual need—it is designed to work well for most people, most of the time. And it is therefore often the less common edge cases where individuals find themselves stymied by a rigid system, looking for help from someone who can facilitate the processes and interactions that the formal system requires.

Many of our interview respondents had at some point needed help in navigating one or more of the identity systems they encountered. The reasons varied according to individual situations, but the common themes were:

  • To expedite a process, e.g., have their application reviewed more quickly or receive their credential sooner than through “normal” processes
  • To avoid or minimize time spent in-person, for example, if the opportunity cost of spending the day at the government office was too high
  • If they did not have enough information, e.g., if the individual wasn’t sure what the process was, and didn’t have the network, skills (e.g., literacy) or confidence to find that information
  • And finally, if they were physically or mentally unable to follow the process, for example, if they were blind or had mental health issues.

A helping hand

Who provides this help? Intermediaries can be front-line agents (e.g., public servants such as ration dealers or private sector employees such as SIM shop owners), independent workers who may charge more but liaise between all parties, but they can also be friends, family, a stranger in the queue, a stationery shop owner next to a site of an identity-based transaction where people might drop in to buy or borrow a pen.

In our research, we came across many of these intermediaries—both “inside” and “outside” the system. Other than Doddaraghu, Tangamma, and Panchayat officials or SIM card issuers (“inside systems”), others were individuals such as those sitting outside the caste certificate office in Garudahalli, Aadhaar franchisees or PAN card issuers (public-private partners), and even NGOs, such as the labor rights NGO in Kesarpur or those working with waste recyclers in Bengaluru.

The perception of some intermediaries or “middlemen” can often be negative—they can be seen as corrupt,1 exploitative,2 exercising discretion in ways that suit them.3 We saw much of this in our own research, as discussed in essay V3. Part of the exploitation is because many of these intermediaries are for-profit actors—i.e., they provide their services for a fee, or they may not be paid enough—yet both end up serving some of the most marginalized or vulnerable populations and so can take advantage of the discrepancy in their knowledge. And because of the information asymmetry involved (the individual typically doesn’t know as much about the processes in question, hence why they seek help), these intermediaries can easily distort or fabricate the actual processes, fees, and timing required. Yet it’s important to note that it’s not just “the poor” who use intermediaries—wherever it is seen to be more time and cost-efficient, the preferred option is to outsource the entire process over to someone more skilled in the area (even if it means paying extra).

While these intermediaries can be exploitative, they are responsive to demand, and thus can be seen as indicators of where the formal state system has broken down or has friction or pain points that are costly to the individual. In this sense, they provide valuable services that directly address the needs of those for whom the system itself is inaccessible. Intermediaries can provide valuable services to help those for whom the system is inaccessible. Understanding these transactions with intermediaries can highlight points of friction that policymakers can then begin to address.

Intermediaries have always existed, but the discrepancy between new technologies and users who may not have the skills or feel comfortable using them, has made this role even more pivotal. Intermediaries also have varying levels of “capital” such as technical or social capital (who you know).4

Paradoxically, the risk here is also that such assistance may be based on networks and favoritism—for example, Tangamma, the Anganwaadi teacher in Garudahalli, helps Madappa to obtain a loan which strictly only his wife is eligible for. She says we know what will help and what will work specifically for specific persons and who are deserving, responsible and likely to benefit from such loans. The problematic issue is that willingness to help will always be made at the intermediary’s discretion (and therefore may equally exclude others).

Bureaucrats to the rescue?

Doddaraghu is an example of what Michael Lipsky called “street-level bureaucrats”5 — those who have to translate and simplify what seem sometimes as bloated bureaucracy and complex policies issued from a faceless state. Unlike the individual intermediaries who exist outside a system, these street-level bureaucrats are agents of the state or of regulated industries (e.g., banks, mobile operators) and have formal roles with codified procedures and responsibilities.

Yet what we found in our research is that these agents often take on a dual role: in addition to their job as a PDS clerk, Aadhaar enrollment official, or SIM card registrar with formal duties for administering identity systems, they also perform actions and behaviors as individuals with their own code of ethics. And it is this dual role that we see in the quote from Doddaraghu at the beginning of the section, where he acts outside the scope of his formal position because of his personal relationship with certain individuals. While his example of allowing a person to take rations for months without showing a credential may be an extreme case, there are countless other actions that take place every day as bureaucrats act outside the official scope of their position in order to facilitate processes for individuals in need.

What’s important about these interactions is that when a state bureaucrat or regulated service provider acts outside of their role, they often assume personal liability and risk that is not covered by their position. We heard many reasons for why they would “bend the rules,” but one of the most common was because the agent felt they did not have support, training, or current information to follow standard procedure and so had to improvise and be resourceful in order to fulfill the spirit of their job (on the flip side they also didn’t have accountability). For example, Aabid, a PAN card supplier in rural Assam, says he is looking for other jobs because the rules keep changing (specifically in terms of feeder ID documents—initially a School Leaving Certificate and village headman/mukhiya certificate were enough for authentication, but now a School Leaving Certificate, birth certificate and voter ID or driving license is needed). While it is his responsibility to keep up there is little information fed down to him. He believes there may be better communication if he becomes an Aadhaar agent as it seems more organized.

Similarly, because end users often have little concept of which public sector entities are responsible for what6 and as identity-based systems become increasingly complex and interdependent, some intermediaries recount that they are pushed over their boundary of knowledge. For example, when we interviewed the LPG provider in Assam on the identity credentials needed to have access to a gas connection, he said his two biggest challenges were the IVR (interactive voice response) technology introduced to order cylinders as well as the need to have a bank account. In both cases, he said he tried to be the mediator, but particularly in the latter he felt he did not know enough about banking processes to really know what he was doing (Ahmed and his colleagues7 narrate a similar experience of a SIM shop owner who is asked by clients about getting a voter ID).

Though these challenges often apply to agents of private firms, their incentives and reasons for acting outside of the formal system can differ. For example, in some situations banks or mobile operators may earn more revenue if they take shortcuts or otherwise don’t follow all regulations (e.g., sign up new customers without fully authenticating). Equally, they have to shoulder the financial impact of fully complying with any regulations, which can be a disincentive that they must balance with the threat of legal punishment.

Mushtaq, a SIM card provider in rural Assam, told us the lengthy process of checking all details in forms. He needs to ensure that all documents, signatures/thumb impressions and details filled in the form are present and correct as if they are rejected by the distributors (to whom he himself must deliver these) it increases his workload and travel costs. To minimize against this, in cases where customers have difficulties filling in the forms, he does so himself—not out of altruism (his explanation) but because it saves him time and money in the long term.

Costs of being vulnerable

In many ways, intermediaries are no different from end users in feeling powerless to navigate systems. Revathi, a bank manager, told us that she was under pressure to match thumb prints and signatures especially during demonetization when a large number of new customers were illiterate. She also had to go out of her way to explain terms and assuage fears. Jafar, the migrant rights coordinator in Kesarpur said along with the migrants, we also have to also live our lives. Even though I work in an NGO and I am more aware of identity cards than other migrants because of my work, I’ve had to face so much trouble. Just opening one bank account was so problematic. Then getting a SIM card. All these have small challenges. Jafar’s statement made it clear that knowledge does not always transfer to power—or ability to take action.

Our research revealed specific ways in which acting like an intermediary can make someone more, not less, vulnerable. Some formal intermediaries, like those sitting outside the caste certificate office, are dependent on the established processes and personnel of the system not changing, as their livelihood is predicated on inefficiencies and opacity that they can mitigate for their customers. Even small policy changes can have a large impact on their business.

We heard multiple accounts of people relying on their own personal finances when coordinating help or transactions for those in need. Tangamma, the Anganwaadi teacher for example, uses WhatsApp to communicate with her supervisor at the district office if there are delays with Aadhaar processing— but frequent power outage or running out of her (personal) mobile data means that she falls behind on calls. While we were interviewing her, she had to take a break to serve some clients, only to find that she couldn’t send a WhatsApp because she was out of data. We offered our mobile instead, which she used to contact her supervisor.

Another challenge is that sometimes the dual roles collapse, and in the eyes of the individual being helped the intermediary becomes the face of the state.8 For example, while we were interviewing Mahadev, the Aadhaar center franchisee in Bengaluru, he was interrupted by a phone call from a customer asking why his application was rejected. Mahadev mentioned that they are not supposed to give personal numbers, but when an applicant is desperate, he sometimes does, but he often cannot explain that the rejection is at UIDAI, not by him and he feels caught in the middle.

Such expenses also lead to opportunistic behavior such as making extra money on the side. At a Delhi liquid propane gas distribution center, for example, the franchisee said they were forced to sell on the black market because: the distribution boys come from a very poor background. They don’t get enough commission from the government for the distributorship. Sometimes the delivery boys have to travel 15/16 km daily just to deliver one gas cylinder. In Delhi, the boys get Rs 19.50 ($0.30) per cylinder per delivery as commission. In that Rs. 19.50, boys have to manage their vehicles, fuel the vehicle, get a helper, satisfy all the traffic keepers [i.e., sometimes bribe] and then deliver the cylinder. Sometimes the customers might stay on the fourth floor. The delivery boy needs to climb all the floors and then he should deliver the empty cylinder back to the warehouse. All for under Rs. 20.


It is almost impossible in the Indian context to imagine identity-based transactions without people both outside and inside systems playing the role of intermediary. The role of the intermediary is pivotal but it is not a simple role—intermediaries hold different levels of capital, skills, networks, and have different incentive structures and vulnerabilities. If we borrow from financial inclusion literature, Maurer et al (2013) distinguish mobile money agents as either neutral channels or more helpful bridges.9 In identity-based transactions, human intermediaries can be obstructive, neutral or helpful (either for profit or not). We found no evidence that those inside the system were “channels” or those outside “bridges”—as we saw, there were many equally helpful and exploitative. In an ideal world, identity-based transactions would be disintermediated both for efficiency and to avoid bias, but as we found in our research, we are far from this and the bias can be just as positive (helping someone out, being flexible in the moment) as negative (acting out of prejudice). To build policy for how low income users access identity credentials, we first have to address the role of intermediaries, and acknowledge that while they may be exploitative or obstructive (as seen in essay V3), in cases where they are helpful, they also need support in many ways, particularly if they fall between multiple systems.

  1. Silvia Masiero, “Digital Governance and the Reconstruction of the Indian Anti-Poverty System,” Oxford Development Studies 0, no. 0 (November 16, 2016): 1–16, doi:10.1080/13600818.2016.1258050.

  2. Reetika Khera, “India’s Public Distribution System: Utilisation and Impact,” The Journal of Development Studies 47, no. 7 (July 1, 2011): 1038–60, doi:10.1080/00220388.2010.506917.

  3. Aloysius Irudaym S. J., Jayshree P. Mangubhai, and Joel G. Lee, eds., Dalit Women Speak Out: Caste Class and Gender Violence in India (Chicago: University of Chicago Press, 2011).

  4. François van Schalkwyk et al., “Open Data Intermediaries: Their Crucial Role,” Web Foundation: Open Data in Developing Countries Phase 2, August 15, 2015.

  5. Michael Lipsky, Street-Level Bureaucracy: Dilemmas of the Individual in Public Services, 30th Anniversary Expanded Edition (New York: Russel Sage Foundation, 2010).

  6. Stuart Corbridge, Seeing the State: Governance and Governmentality in India (Cambridge, UK: Cambridge University Press, 2005).

  7. Syed Ishtiaque Ahmed et al., “Privacy, Security, and Surveillance in the Global South: A Study of Biometric Mobile SIM Registration in Bangladesh,” in Proceedings of the 2017 CHI Conference on Human Factors in Computing Systems, CHI ’17 (New York, NY, USA: ACM, 2017), 906–918, doi:10.1145/3025453.3025961.

  8. Masiero, “Digital Governance and the Reconstruction of the Indian Anti-Poverty System.”

  9. Bill Maurer, Taylor C. Nelms, and Stephen C. Rea, “‘Bridges to Cash’: Channelling Agency in Mobile Money,” Journal of the Royal Anthropological Institute 19, no. 1 (March 1, 2013): 52–74, doi:10.1111/1467-9655.12003.