COVID-19 and Aadhaar: Why the Union Government’s Relief Package is an Exclusionary Endeavour

In response to the COVID-19 pandemic, Finance Minister Nirmala Sitharaman on 26 March 2020 announced a relief package for the informal sector. This package was a combination of cash and food assistance, which included:

1. Doubling public distribution system (PDS) ration entitlements for three months for beneficiaries covered by the National Food Security Act, with the extra ration to be given free of cost.
2. For pensioners (old age, widows, and persons with disabilities), Rs. 1000 in two instalments is to be disbursed over the coming three months. 
3. For female Jan Dhan Yojana bank account holders, Rs 500 per month is to be provided for three months.
4. Free LPG cylinders are also to be given to beneficiaries of the Ujjwala Yojana scheme for three months. 

The officially stated value of this relief package, that is Rs. 1.7 lakh crores, is deceptive. For one, its value is overstated—it includes money already committed under the PM-KISAN Yojana. This package ends up amounting to roughly Rs 1 lakh crore, or 0.5% of the gross domestic product (GDP), and is clearly inadequate. It ignores the workers employed under the Mahatma Gandhi National Rural Employment Guarantee Act (MGNREGA) because, in most states, MGNREGA works have been suspended due to the risk of community transition. Similarly, no relief was announced for migrants or for the one-third who do not have PDS ration cards.^[1] This brief note focuses on the hurdles that Aadhaar places in accessing the announced relief.

Food Transfers

In the PDS, Aadhaar is an important source of exclusion. It excludes people in three main ways: cancellation of cards (or names on ration cards) if people do not have an Aadhaar number, failure to link or if they fail to link it with their ration card, and failure of the Aadhaar-based biometric authentication (ABBA) at the time of purchasing grain (Khera 2017). 
Administrative data on PDS transactions is available from two sources: State government portals (accessible through https://pdsportal.nic.in) provide monthly transaction data and Annavitaran (https://annavitran.nic.in), which is a national portal that covers all states and union territories with consolidated data on the number of ration cards and beneficiaries, transactions, and information about authentication modes. 

We define the recorded “quantity transaction rates” (QTR),—the quantity of grain disbursed to PDS beneficiaries’ households (as recorded by the point-of-sale [POS] machine)—as a proportion of the total “in-principle” entitlements for that state (Khalid 2018). In-principle entitlement is calculated as the sum of per capita entitlement (5 kg/month) multiplied by the number of people in priority households and the entitlements of Antyodaya households (35 kg/month) times the number of Antyodaya Anna Yojana ration cards. Information on the number of PDS (priority and Antyodaya) beneficiaries is taken from the Ministry of Food’s “Monthly Foodgrain Bulletin.” The share of the in-principle entitlement which is not disbursed is often (but not always) related to technology failures at the last mile.^[2] Note that the recorded QTR is not the same as the “purchase-entitlement ratio” (PER) which measures quantity fraud. The recorded QTR may be higher than the PER due to continued quantity fraud (for example, a dealer taking a cut of a few kilograms of grain) even with ABBA (Drèze et al 2017).

We present this analysis for a handful of states using both the state PDS portals and the Annavitaran portal. The choice of states is based on the availability of data and to show the variation in transaction rates across states, based on the method of last mile authentication, which refers to the mode of recording the sale of PDS grain to the ration card holder. The central government’s favoured method, ABBA, is used in Jharkhand and Rajasthan. Tamil Nadu uses non-biometric smart cards (like an ATM card) and Chhattisgarh provides each PDS outlet with a tablet, which is used to upload a photograph of the PDS beneficiary with the grain that they have received (Khera 2018).

The monthly QTR for these states is presented in Figure 1 (state portal) and in Figure 2 (Annavitaran portal). Data from both sources suggests that Tamil Nadu and Chhattisgarh, where ABBA is not used, consistently record QTR of close to 100% for most months. On the other hand, for Rajasthan and Jharkhand, both data sources suggest that the QTR, averaged over 24 months (Jan 2018–Dec 2019), is lower, at 86% and 91% respectively. 

Figure 1: Transacted Quantities from PDS as a Share (%) of Total In-principle NFSA Entitlements (State Portals) 

Figure 2: Transacted Quantities from PDS as a Share (%) of Total In-principle NFSA Entitlements (Annavitaran Portal)

For long we have suggested that instead of using a technologically-demanding and unreliable option such as ABBA for last-mile authentication, the central government should explore more reliable options such as non-biometric smart cards. Figures 1 and 2 confirm that states using alternatives to ABBA have consistently higher QTRs than those who rely on ABBA (sometimes in combination with an override mechanism). Earlier evidence from small field studies is therefore confirmed by official data as well (Drèze et al 2017; Muralidharan et al 2020).

Cash Transfers

A cash-only relief strategy at this time is inadvisable for the following reasons: The possibility of panic buying, hoarding, and supply disruption may lead to an increase in prices and could also erode the value of cash. Further, the density of bank branches in rural areas is thin. Mass cash transfers will create crowding, which in turn will increase the risk of community transmission of the virus (Khera 2020). 
There is also a largely unacknowledged issue (by the government) in the payment system. A mess in the banking system has been brewing since the integration of bank accounts with Aadhaar began some years ago. MGNREGA workers and pensioners were essentially coerced into linking their Aadhaar with bank accounts as wages and pension payments were made conditional on Aadhaar-linkages. This led to chaos, harassment, and confusion on the ground resulting in disruptions and exclusion. ^[3]

Here, we present some data on payment failures, from the Direct Benefit Transfer (DBT) Mission obtained through Right to Information requests filed by Srinivas Kodali. We focus on social security pensions and maternity entitlements. 

The DBT mission data provides us with the proportion of total payments that are “Aadhaar-based payments” and those that are “non-Aadhaar based payments.” No formal documentation is available, but Aadhaar-based payments refers to payments made through the Aadhaar Payment Bridge System (APBS) while non-Aadhaar payments refer to payments made through the National Electronic Fund Transfer (NEFT) system.

Aadhaar seeding of bank accounts has been de facto mandatory even before the Supreme Court formally allowed it in 2018. Further, as Dhorajiwala et al (2019) point out, before an account transitions to the APBS using the NPCI mapper, bank accounts have to be linked (or “seeded”) with Aadhaar. Thus, even among non-Aadhaar based payments, Aadhaar has created multiple issues such as wrongly-seeded, inactive, and deactivated Aadhaar, among others. 

Between 2017–18 and 2018–19, the share of APBS increased for most of the listed schemes (See Table 1). In 2018–19, between 20% and 35% of social security pensions and 65% of maternity entitlement payments under the Pradhan Mantri Matru Vandana Yojana (PMMVY) were made through APBS. With the transition to APBS, it is possible that the situation may have worsened. One fallout of faulty transitions is “diverted payments.”^[4] This is what had happened in the case of Airtel, when people found that their LPG subsidy was being diverted to their mobile wallet accounts—often opened without their consent or knowledge (Venkatnarayanan and Lakshmanan 2017). 

Table 1: DBT Payment Failures for Select Schemes^[5]