‘India Needs Banks In Underserved Areas, Not Aadhaar, To Deliver Direct Benefit Transfers’

Cash transfers are projected as ‘hassle-free’ but they are anything but that for the poor: Many find it hard to withdraw cash from distant and crowded banks, or have to deal with middlemen who may demand a bribe, or encounter technology-related issues. With Aadhaar-linkages, their problems have multiplied.

Mumbai: With less than a couple of months for the 2019 general elections, political parties have called for welfare schemes and programmes that will guarantee the development of disadvantaged socio-economic groups in the country.

The Pradhan Mantri Kisan Samman Nidhi (prime minister’s farmer fund) was announced by interim finance minister Piyush Goyal in the 2019-20 (interim) budget presented by the Narendra Modi government. The scheme aims to provide direct annual income support of Rs 6,000 to farmers who own less than two hectares of cultivable land. This amount is to be directly transferred into farmers’ bank accounts.

The Congress party has promised the poor a minimum income guarantee if voted to power in 2019. It claims this will not be a universal scheme, but a “progressive” one (families would get higher income guarantees the further they are from a specified minimum income threshold). A similar idea of universal basic income (UBI) was discussed in the Economic Survey 2016-17.

The implementation and delivery of such schemes depend on the direct benefit transfer (DBT) mechanism, which works through instruments such as Jan Dhan, Aadhaar and mobile payment systems, collectively known as JAM.

Aadhaar and its use are a debated issue with critics pointing to its vulnerabilities and issues. As many as 164 Aadhaar-related frauds had been reported from 2011 onwards, most of them in 2018, an IndiaSpend story said on May 23, 2018. There have been reports of starvation deaths attributed to cancelation of ration cards due to non-seeding of Aadhaar, including this October 23, 2018, IndiaSpend report.

“The belief that direct benefit transfers require Aadhaar-like biometric/digital IDs has been planted in our heads through relentless propaganda,” said Reetika Khera, a development economist, in an interview to IndiaSpend. “What good implementation of DBT requires is an expansion of the reach of our banks to underserved areas.” The problem of duplicate beneficiaries--which Aadhaar is expected to eliminate and thereby enable savings--has been overblown, and no studies have ever been undertaken to estimate the number of fake or duplicate beneficiaries, she said.

Khera is an associate professor of economics at the Indian Institute of Management, Ahmedabad. She holds a PhD in economics from the Delhi School of Economics, University of Delhi, and an MPhil in development studies from the Institute of Development Studies, University of Sussex. She has edited two books, The Battle for Employment Guarantee (2011) and Dissent on Aadhaar: Big Data meets Big Brother (2019), and contributed to several academic journals.

Excerpts from an email interview with IndiaSpend:

The income support for small farmers announced in the 2019 budget is to be transferred to beneficiaries through Aadhaar-enabled direct benefit transfers. Do you foresee any issues in the implementation of this scheme?

Even the government saw problems with using Aadhaar in transferring farmer income. That is why it decided to make Aadhaar optional for the first instalment. This is about as much of an admission of Aadhaar’s failures as we are going to get from the government. These failures range from ‘rejected payments’ (e.g., due to wrongly seeded Aadhaar numbers), ‘redirected payments’ (due to the last-Aadhaar linked-account rule [the system automatically transfers money to the account that was linked last, in cases where beneficiaries have multiple Aadhaar-linked accounts]), ‘frozen accounts’ and so on. There are reports of a massive mess in the banking system created by the requirement of eKYC/Aadhaar and the linking of bank accounts with the National Payments Corporation of India.

As practicalities go, there are obvious reasons to be concerned. We know that other cash transfers--be it social security pensions such as old age pensions or scholarships such as the junior research fellowship--are not regular. Gap months cause hardship to students and the elderly. All these issues are bound to rise with the implementation of the proposed support for small and marginal farmers.

The Congress too has promised a minimum income guarantee. What are the challenges of implementing such schemes through Jan Dhan, Aadhaar and mobile systems?

None of the proposals on the table so far qualify as universal basic income--they are neither universal, nor do they guarantee a 'basic income'. A larger worry with cash transfers is that with inflation, their value erodes over time and no government has shown any inclination to index these transfers. Since 2006, the central government's contribution to old-age pensions had been stuck at Rs 200 per person per month.

Besides, it is not obvious that the best use of available fiscal space is to introduce new cash-transfers--either of the sort the Congress is promising or the farm-income support announced by the current government. Ideally, other sectors--I have primary healthcare in mind--should be prioritised.

Cash transfers are projected as 'hassle-free' but they are anything but that for the poor: They find it hard to withdraw cash from distant and crowded banks; they may have to have to deal with middlemen such as banking correspondents, who may demand a bribe, or encounter technology-related issues such as the lack of connectivity or biometric recognition failures. With Aadhaar-linkages their problems have multiplied. Now the demographic data in the Aadhaar database needs to match their demographic data with the bank, along with ensuring that their accounts do not get cross-wired with someone else's Aadhaar number, etc.

A 2016 World Bank report said that using digital identity, India could potentially save $11 billion (Rs 77,000 core) every year through efficiency gains and plugging of leaks. That’s the equivalent to nearly 30% of India’s annual subsidy bill. You had contested this figure in February 2018 but it is being widely quoted by ministers and the media. How do you respond to that?

This is the most egregious example of sponsored research that we have encountered. In its 2016 World Development Report, the World Bank claimed that Aadhaar could result in savings of $11 billion if linked to all welfare programmes. This was when India’s total welfare spending was $11 billion. The World Bank passed off total spending as potential savings.

When this was pointed out to them, instead of admitting their error, they made further convoluted claims which Jean Drèze and I wrote about in The Economic Times in early 2018. They claimed that the $11 billion estimate was based on a 10% of total welfare spending, and that they had underestimated the total welfare spending by a factor of 10. They inflated welfare spending by classifying things such as horticulture, dairy, wildlife, etc. as “welfare”.

There have been reports of fake Aadhaar cards. But surely Aadhaar has, in some measure, been able to reduce corruption--one of its primary objectives--in welfare schemes such as the public distribution system and NREGA. This estimate says Aadhaar has removed 66 million ghost beneficiaries from various welfare schemes such as liquefied petroleum gas (LPG) subsidy and public distribution system (of food).

I’m glad that you bring up the information given in Parliament. Many of the deletions, especially in LPG connections, pre-date the use of Aadhaar.

More importantly, if you read the answer carefully, it clearly states that the cancellations were on account of duplication, ineligible, fake/non-existent, inactive, “including some due to migration, death, etc”. Of these, Aadhaar can only help identify “duplicate” users. This means that the other deletions had nothing to do with Aadhaar. The breakup has never been made public.

In a nutshell, the government has been passing off all manner of deletions as Aadhaar-related. In an answer to Parliament dated July 19, 2018, regarding NREGA job card deletions, they admit clearly that the deletions had nothing to do with Aadhaar.

Question: “(b) whether all these cancellations have been made as per Aadhaar based verification and if so, the details thereof; and (c) the details of savings expected due to such cancellations?”

Answer: (b): No Madam (c): Does not arise.

Source: Lok Sabha

Are there other technological solutions that could enable direct benefit transfers while eliminating duplicate or fake beneficiaries? What would constitute a fool-proof system?

One major problem with the Aadhaar project is that it is ‘anecdote-based’ policy-making masquerading as ‘evidence-based’ policy-making. Look at the problem of duplicate beneficiaries. We have always maintained that we should first document the size of this problem. No government has shown any interest in doing this.

Interestingly, exactly the same debate played out in the US when biometrics were sought to be introduced in welfare: The problem of duplication was overplayed without any evidence, and the lack of evidence was presented as proof of its existence. Shoshana Amielle Magnet’s book When Biometrics Fail suggests that the biometrics industry lobby was behind this charade.

The belief that direct benefit transfers require Aadhaar-like biometric/digital IDs has been planted in our heads through relentless propaganda. Let me explain. For DBTs, you require bank accounts with modern banking systems, such as Centralised Online Real-time Exchange (CORE) banking, National Electronic Funds Transfer (NEFT), etc. Even before Aadhaar came along, many of us had access to a modern banking system, and for those who found it difficult to open bank accounts, successive governments had been taking steps such as allowing NREGA job cards as a valid KYC document to open accounts. Yet, we were made to believe that Aadhaar was necessary to reach the unbanked.

What good implementation of DBT requires is an expansion of the reach of our banks to under-served areas. Simple steps can achieve that: Opening extra counters at existing bank branches can provide huge relief to those who use them for drawing their NREGA wages, pensions, transfers for housing schemes, etc.

You talk about the potential misuse of Aadhaar in your new book Dissent on Aadhaar: Big Data Meets Big Brother. Can you explain these issues?

One, even your very basic demographic details, such as your date of birth and address, can lead to identity fraud. Phishing attacks also require very little information. So, the more places your Aadhaar data is stored, the greater the chances of it becoming public, and leading to personal harm in the form of identity fraud and phishing attacks. This is just one aspect of privacy, commonly called ‘informational privacy’ in the literature.

Two, the availability of your basic Aadhaar information in more and more databases combined with the rise of the data broking industry makes it very easy to combine data from various sources to create a 360-degree profile of my life: My rail and air travel, phone use, online shopping, etc. which exposes me to payday loans--a short-term unsecured loan to be repaid on the borrower's next salary day--surge pricing, cynical use of my profile for targeted advertising, etc. Bruce Schneier [a well-known American cryptographer and computer security and privacy specialist] refers to this as corporate surveillance.

Third, by making us ‘seed’ or link our Aadhaar number to all databases, it creates a bridge across hitherto distinct data silos. That’s how Aadhaar facilitates profiling, tracking and surveillance. There is always the danger that the government will demand our data from various sources to profile and surveil us. This amounts to a violation of our right to privacy and our right to freedom.

Finally, given the low levels of tech-digital-legal literacy--when the Telecom Regulatory Authority of India chief tweeted his Aadhaar number, it allowed people to get some of his bank information, email passwords, etc.--the scope for personal harm is much greater.

Do you oppose the very idea of Aadhaar, or just how it has been rolled out?

I think IDs play a useful role in welfare administration, but what we need are programme-specific IDs because the pool of beneficiaries is often quite distinct--someone who needs housing may not require a social security pension. Such decentralised IDs (programme-wise and perhaps even state-wise) are better than a centralised, single national ID such as Aadhaar. By design, decentralised IDs are more secure and provide protection against profiling and tracking.

Another problem with Aadhaar is that for the government and the corporations pushing for it, it only makes sense when it is made mandatory. The problems with a mandatory Aadhaar are, by now, well understood.

Would multiple, programme-wise IDs not cause overlap of effort and create more difficulties for the poor?

An integrated single ID opens the door to profiling and surveillance in a way that programme-specific IDs won't. Besides, my sense is that other IDs aren't going to go away--the proponents of Aadhaar like to compare it with the United States’ SSN [social security number, a nine-digit number issued to US citizens, permanent residents, and temporary (working) residents that is used as a national identification number for various purposes including taxation], where as far as I'm aware passports, driver’s licenses and birth certificates continue to exist. In pre-Aadhaar days, a person who needed housing or an old-age pension or a mid-day meal was not required to acquire a separate ID for those programmes. That person was assigned a unique number by the administrative department of that particular scheme (which they used internally for administrative purposes). For example, being enrolled in a government school automatically entitles a child to a school meal without having to submit or link Aadhaar. Aadhaar creates additional hurdles. On the other hand, possession of Aadhaar does not automatically entitle anyone to anything. They continue to have to make applications for each form of social support.

You have compiled a list of 75 starvation deaths since 2015, of which 42 were linked to Aadhaar-related problems. However, the government has denied any such deaths.

In 2002-03, there was a similar wave of reports of starvation deaths and the response of the governments at the time was exactly the same: Denial. Sometimes it may be hard to pin a death down as starvation- or hunger-related, but once you have seen the condition in which these people were surviving, it only requires common sense to see that there is a serious problem. Further, in those cases where the persons were cut-off from niggardly yet life-line forms of social support, the role of Aadhaar in causing exclusion is clear. Why does the government continue to deny these problems? One can only guess. One possibility is that if they acknowledge that there was a lapse, they will have to take remedial action, which they want to avoid.

(Mallapur is a senior policy analyst with IndiaSpend.)

We welcome feedback. Please write to respond@indiaspend.org. We reserve the right to edit responses for language and grammar.

Mumbai: With less than a couple of months for the 2019 general elections, political parties have called for welfare schemes and programmes that will guarantee the development of disadvantaged socio-economic groups in the country.

The Pradhan Mantri Kisan Samman Nidhi (prime minister’s farmer fund) was announced by interim finance minister Piyush Goyal in the 2019-20 (interim) budget presented by the Narendra Modi government. The scheme aims to provide direct annual income support of Rs 6,000 to farmers who own less than two hectares of cultivable land. This amount is to be directly transferred into farmers’ bank accounts.

The Congress party has promised the poor a minimum income guarantee if voted to power in 2019. It claims this will not be a universal scheme, but a “progressive” one (families would get higher income guarantees the further they are from a specified minimum income threshold). A similar idea of universal basic income (UBI) was discussed in the Economic Survey 2016-17.

The implementation and delivery of such schemes depend on the direct benefit transfer (DBT) mechanism, which works through instruments such as Jan Dhan, Aadhaar and mobile payment systems, collectively known as JAM.

Aadhaar and its use are a debated issue with critics pointing to its vulnerabilities and issues. As many as 164 Aadhaar-related frauds had been reported from 2011 onwards, most of them in 2018, an IndiaSpend story said on May 23, 2018. There have been reports of starvation deaths attributed to cancelation of ration cards due to non-seeding of Aadhaar, including this October 23, 2018, IndiaSpend report.

“The belief that direct benefit transfers require Aadhaar-like biometric/digital IDs has been planted in our heads through relentless propaganda,” said Reetika Khera, a development economist, in an interview to IndiaSpend. “What good implementation of DBT requires is an expansion of the reach of our banks to underserved areas.” The problem of duplicate beneficiaries--which Aadhaar is expected to eliminate and thereby enable savings--has been overblown, and no studies have ever been undertaken to estimate the number of fake or duplicate beneficiaries, she said.

Khera is an associate professor of economics at the Indian Institute of Management, Ahmedabad. She holds a PhD in economics from the Delhi School of Economics, University of Delhi, and an MPhil in development studies from the Institute of Development Studies, University of Sussex. She has edited two books, The Battle for Employment Guarantee (2011) and Dissent on Aadhaar: Big Data meets Big Brother (2019), and contributed to several academic journals.

Excerpts from an email interview with IndiaSpend:

The income support for small farmers announced in the 2019 budget is to be transferred to beneficiaries through Aadhaar-enabled direct benefit transfers. Do you foresee any issues in the implementation of this scheme?

Even the government saw problems with using Aadhaar in transferring farmer income. That is why it decided to make Aadhaar optional for the first instalment. This is about as much of an admission of Aadhaar’s failures as we are going to get from the government. These failures range from ‘rejected payments’ (e.g., due to wrongly seeded Aadhaar numbers), ‘redirected payments’ (due to the last-Aadhaar linked-account rule [the system automatically transfers money to the account that was linked last, in cases where beneficiaries have multiple Aadhaar-linked accounts]), ‘frozen accounts’ and so on. There are reports of a massive mess in the banking system created by the requirement of eKYC/Aadhaar and the linking of bank accounts with the National Payments Corporation of India.

As practicalities go, there are obvious reasons to be concerned. We know that other cash transfers--be it social security pensions such as old age pensions or scholarships such as the junior research fellowship--are not regular. Gap months cause hardship to students and the elderly. All these issues are bound to rise with the implementation of the proposed support for small and marginal farmers.

The Congress too has promised a minimum income guarantee. What are the challenges of implementing such schemes through Jan Dhan, Aadhaar and mobile systems?

None of the proposals on the table so far qualify as universal basic income--they are neither universal, nor do they guarantee a 'basic income'. A larger worry with cash transfers is that with inflation, their value erodes over time and no government has shown any inclination to index these transfers. Since 2006, the central government's contribution to old-age pensions had been stuck at Rs 200 per person per month.

Besides, it is not obvious that the best use of available fiscal space is to introduce new cash-transfers--either of the sort the Congress is promising or the farm-income support announced by the current government. Ideally, other sectors--I have primary healthcare in mind--should be prioritised.

Cash transfers are projected as 'hassle-free' but they are anything but that for the poor: They find it hard to withdraw cash from distant and crowded banks; they may have to have to deal with middlemen such as banking correspondents, who may demand a bribe, or encounter technology-related issues such as the lack of connectivity or biometric recognition failures. With Aadhaar-linkages their problems have multiplied. Now the demographic data in the Aadhaar database needs to match their demographic data with the bank, along with ensuring that their accounts do not get cross-wired with someone else's Aadhaar number, etc.

A 2016 World Bank report said that using digital identity, India could potentially save $11 billion (Rs 77,000 core) every year through efficiency gains and plugging of leaks. That’s the equivalent to nearly 30% of India’s annual subsidy bill. You had contested this figure in February 2018 but it is being widely quoted by ministers and the media. How do you respond to that?

This is the most egregious example of sponsored research that we have encountered. In its 2016 World Development Report, the World Bank claimed that Aadhaar could result in savings of $11 billion if linked to all welfare programmes. This was when India’s total welfare spending was $11 billion. The World Bank passed off total spending as potential savings.

When this was pointed out to them, instead of admitting their error, they made further convoluted claims which Jean Drèze and I wrote about in The Economic Times in early 2018. They claimed that the $11 billion estimate was based on a 10% of total welfare spending, and that they had underestimated the total welfare spending by a factor of 10. They inflated welfare spending by classifying things such as horticulture, dairy, wildlife, etc. as “welfare”.

There have been reports of fake Aadhaar cards. But surely Aadhaar has, in some measure, been able to reduce corruption--one of its primary objectives--in welfare schemes such as the public distribution system and NREGA. This estimate says Aadhaar has removed 66 million ghost beneficiaries from various welfare schemes such as liquefied petroleum gas (LPG) subsidy and public distribution system (of food).

I’m glad that you bring up the information given in Parliament. Many of the deletions, especially in LPG connections, pre-date the use of Aadhaar.

More importantly, if you read the answer carefully, it clearly states that the cancellations were on account of duplication, ineligible, fake/non-existent, inactive, “including some due to migration, death, etc”. Of these, Aadhaar can only help identify “duplicate” users. This means that the other deletions had nothing to do with Aadhaar. The breakup has never been made public.

In a nutshell, the government has been passing off all manner of deletions as Aadhaar-related. In an answer to Parliament dated July 19, 2018, regarding NREGA job card deletions, they admit clearly that the deletions had nothing to do with Aadhaar.

Question: “(b) whether all these cancellations have been made as per Aadhaar based verification and if so, the details thereof; and (c) the details of savings expected due to such cancellations?”

Answer: (b): No Madam (c): Does not arise.

Source: Lok Sabha

Are there other technological solutions that could enable direct benefit transfers while eliminating duplicate or fake beneficiaries? What would constitute a fool-proof system?

One major problem with the Aadhaar project is that it is ‘anecdote-based’ policy-making masquerading as ‘evidence-based’ policy-making. Look at the problem of duplicate beneficiaries. We have always maintained that we should first document the size of this problem. No government has shown any interest in doing this.

Interestingly, exactly the same debate played out in the US when biometrics were sought to be introduced in welfare: The problem of duplication was overplayed without any evidence, and the lack of evidence was presented as proof of its existence. Shoshana Amielle Magnet’s book When Biometrics Fail suggests that the biometrics industry lobby was behind this charade.

The belief that direct benefit transfers require Aadhaar-like biometric/digital IDs has been planted in our heads through relentless propaganda. Let me explain. For DBTs, you require bank accounts with modern banking systems, such as Centralised Online Real-time Exchange (CORE) banking, National Electronic Funds Transfer (NEFT), etc. Even before Aadhaar came along, many of us had access to a modern banking system, and for those who found it difficult to open bank accounts, successive governments had been taking steps such as allowing NREGA job cards as a valid KYC document to open accounts. Yet, we were made to believe that Aadhaar was necessary to reach the unbanked.

What good implementation of DBT requires is an expansion of the reach of our banks to under-served areas. Simple steps can achieve that: Opening extra counters at existing bank branches can provide huge relief to those who use them for drawing their NREGA wages, pensions, transfers for housing schemes, etc.

You talk about the potential misuse of Aadhaar in your new book Dissent on Aadhaar: Big Data Meets Big Brother. Can you explain these issues?

One, even your very basic demographic details, such as your date of birth and address, can lead to identity fraud. Phishing attacks also require very little information. So, the more places your Aadhaar data is stored, the greater the chances of it becoming public, and leading to personal harm in the form of identity fraud and phishing attacks. This is just one aspect of privacy, commonly called ‘informational privacy’ in the literature.

Two, the availability of your basic Aadhaar information in more and more databases combined with the rise of the data broking industry makes it very easy to combine data from various sources to create a 360-degree profile of my life: My rail and air travel, phone use, online shopping, etc. which exposes me to payday loans--a short-term unsecured loan to be repaid on the borrower's next salary day--surge pricing, cynical use of my profile for targeted advertising, etc. Bruce Schneier [a well-known American cryptographer and computer security and privacy specialist] refers to this as corporate surveillance.

Third, by making us ‘seed’ or link our Aadhaar number to all databases, it creates a bridge across hitherto distinct data silos. That’s how Aadhaar facilitates profiling, tracking and surveillance. There is always the danger that the government will demand our data from various sources to profile and surveil us. This amounts to a violation of our right to privacy and our right to freedom.

Finally, given the low levels of tech-digital-legal literacy--when the Telecom Regulatory Authority of India chief tweeted his Aadhaar number, it allowed people to get some of his bank information, email passwords, etc.--the scope for personal harm is much greater.

Do you oppose the very idea of Aadhaar, or just how it has been rolled out?

I think IDs play a useful role in welfare administration, but what we need are programme-specific IDs because the pool of beneficiaries is often quite distinct--someone who needs housing may not require a social security pension. Such decentralised IDs (programme-wise and perhaps even state-wise) are better than a centralised, single national ID such as Aadhaar. By design, decentralised IDs are more secure and provide protection against profiling and tracking.

Another problem with Aadhaar is that for the government and the corporations pushing for it, it only makes sense when it is made mandatory. The problems with a mandatory Aadhaar are, by now, well understood.

Would multiple, programme-wise IDs not cause overlap of effort and create more difficulties for the poor?

An integrated single ID opens the door to profiling and surveillance in a way that programme-specific IDs won't. Besides, my sense is that other IDs aren't going to go away--the proponents of Aadhaar like to compare it with the United States’ SSN [social security number, a nine-digit number issued to US citizens, permanent residents, and temporary (working) residents that is used as a national identification number for various purposes including taxation], where as far as I'm aware passports, driver’s licenses and birth certificates continue to exist. In pre-Aadhaar days, a person who needed housing or an old-age pension or a mid-day meal was not required to acquire a separate ID for those programmes. That person was assigned a unique number by the administrative department of that particular scheme (which they used internally for administrative purposes). For example, being enrolled in a government school automatically entitles a child to a school meal without having to submit or link Aadhaar. Aadhaar creates additional hurdles. On the other hand, possession of Aadhaar does not automatically entitle anyone to anything. They continue to have to make applications for each form of social support.

You have compiled a list of 75 starvation deaths since 2015, of which 42 were linked to Aadhaar-related problems. However, the government has denied any such deaths.

In 2002-03, there was a similar wave of reports of starvation deaths and the response of the governments at the time was exactly the same: Denial. Sometimes it may be hard to pin a death down as starvation- or hunger-related, but once you have seen the condition in which these people were surviving, it only requires common sense to see that there is a serious problem. Further, in those cases where the persons were cut-off from niggardly yet life-line forms of social support, the role of Aadhaar in causing exclusion is clear. Why does the government continue to deny these problems? One can only guess. One possibility is that if they acknowledge that there was a lapse, they will have to take remedial action, which they want to avoid.

(Mallapur is a senior policy analyst with IndiaSpend.)

We welcome feedback. Please write to respond@indiaspend.org. We reserve the right to edit responses for language and grammar.