The Union Government hastily passed a Bill to link voter IDs with Aadhaar cards. Such arbitrary approaches without sufficient deliberation have become the norm, it seems. The Government claims that the move will prevent frauds and remove duplicate IDs. But as evidenced by a report by journalist Kumar Sambhav, such a move violates an individual’s Right to Privacy, enables voter profiling and excludes genuine voters — perils that have also been outlined in a statement issued by the Rethink Aadhaar campaign and endorsed by many organisations. Aadhaar was, among other things, purportedly meant to improve efficiency in welfare programmes. However, there are important lessons to learn from the dubious claims made by the Government in Aadhaar linking for welfare delivery which have a strong bearing on the new proposal.
Claims on shaky ground
Cash transfers in many welfare programmes, such as payment of MGNREGA wages, are done using the Aadhaar Payment Bridge System (APBS). For this to work, it is mandatory to link workers’ Aadhaar with their MGNREGA job cards and their bank accounts where the Aadhaar number of the worker becomes their financial address. The Union Government has repeatedly made claims on savings in welfare programmes due to Aadhaar. These have been methodically debunked by Jean Drèze and Reetika Khera, among others. But the Government continues claiming that “the estimated cumulative savings/benefits due to Aadhaar in MGNREGA till March, 2021 is Rs 33,475 crores.” Two Right to Information responses seeking the methodology used to arrive at such savings are relevant. In a recent response, the Government said the “Ministry has been reporting DBT Mission on the estimated DBT savings under the scheme on the assumption that 10% of the wages in the year could be saved.” In an earlier response, it had said: “Savings are in terms of increasing the efficiency and reducing the delay in payments etc.” The savings due to Aadhaar, therefore, appear to be an “assumption” while the other claims are also on shaky grounds.
Wage payment delays in MGNREGA have been persistent. An analysis of more than 18 lakh wage invoices for the first half of 2021-22 by LibTech India demonstrated that 71% of the payments were delayed (called stage 2 delays) beyond the mandated period by the Union Government. Nearly 7 lakh invoices in our sample were done through the APBS; 11.65 lakh were account-based payments where the workers’ name, account number, and the IFSC code of their bank were used to transfer money. Figure 1 compares the time taken by the Union Government (stage 2) in transferring wages for the two payment methods. The axes represent the number of days taken for two kinds of payments. The 45-degree line shows the percentiles of stage 2 for APBS and the dots represent the percentiles of stage 2 for account-based payments. When dots are below the line, the account-based payments are quicker. Barring a few cases, the dots are practically on or below the line. This is perhaps the first large sample empirical evidence demonstrating that the Government’s claim of Aadhaar having “reduced payment delays” is unfounded. Indeed, there is nothing inherent in the APBS that makes transfers faster.
The government’s claims on “increasing the efficiency” is also questionable. Efficiency for whom and how is such efficiency related to accountability? Between 2015 and 2019, there was intense pressure on field-level bureaucrats to increase Aadhaar linking. A recently completed study of nearly 3,000 MGNREGA workers by Anjor Bhaskar and Preeti Singh shows that 57% of job cards of genuine workers were deleted in a quest to show 100% linking of Aadhaar with job cards. Doing such plastic surgery on numbers to show efficiency gains is unethical and sets a harmful precedent.
Another key concern is the opacity surrounding APBS and the consequent dilution of accountability. Cash transfers through both the payment methods can fail. The most common reason for payment failures in account-based payments is when the account number of the worker in the system is incorrect. This can be rectified at the block. However, the most common reason for payment failures through the APBS is enigmatically called “Inactive Aadhaar.” This has nothing to do with an individual’s Aadhaar being inactive but happens when there is a software mapping failure with the centralised National Payments Corporation of India, the clearing house for APBS. Workers and officials alike are clueless on resolving these payment failures.
Moreover, there are several cases of misdirected payments in APBS when the Aadhaar number of one person gets linked to somebody else’s bank account so her money gets credited to somebody else’s account. These are very hard to detect as these will appear as successful transactions on the dashboard. As per UIDAI, its functions include “setting up of facilitation centres and grievance redressal mechanism for redressal for grievances of individuals.” However, no such mechanisms exist.
Beyond technological alibis
So, on at least three counts — timely payment of wages, efficiency gains and grievance resolution — there appears no basis to justify APBS in MGNREGA. These prompt us to move beyond technological alibis for good governance and emphasise the need for a push towards constitutional propriety and accountability for technologies. The mathematician Cathy O’Neil cautions us on how some algorithmic models and technologies for social policies can be at odds with fairness. She writes: “Fairness is squishy and hard to quantify. It is a concept. And computers for all their advances in language and logic still struggle with concepts… Programmers don’t know how to code for it...” Indeed, compromising on fairness, people were coerced into using Aadhaar which had no pilot or independent cost-benefit analysis. No feedback has been collected on the user experience of the recipients or from field-level bureaucrats. Further, it is time to overhaul the nomenclature for recipients of welfare measures. Calling them “beneficiaries” subtly transforms the state from being an institution meant to uphold constitutional rights to sounding more like a charitable institution. Instead, people should be referred to as “rights holders”. This will likely help us better interrogate whether technologies have imbibed democratic principles of transparency, accountability and participation.
When Aadhaar’s use in welfare — for which it was purportedly intended — is itself shrouded in opacity, unreliability, and exclusions, we must be very worried if it is linked to voter IDs as it will further hollow out government accountability. It will fundamentally alter the citizen-State relationship. This must concern everyone as thousands of crores of taxpayers’ money have been spent on it. The Law Minister said that linking Aadhaar with voter id is “voluntary”. But given the prior experience of Aadhaar in other spheres, this will be another example of what Cathy O’Neil refers to as the “authority of the inscrutable.”
Rajendran Narayanan teaches at Azim Premji University, Bengaluru and is affiliated with LibTech India