Tuesday, March 29, 2016

71. UID project: Does Evidence matter?, Ideas for India, 5 February, 2014.


UID project: Does Evidence matter?
Ideas for India
5 February, 2014.


UID project: Does evidence matter? 
Posted On: 05 Feb 2014 
Section: Perspectives
Tags: 
UID , IT , PDS , MNREGA
Post Comment



Reetika Khera Indian Institute of Technology, Delhi
reetika.khera@gmail.com
The Indian government has been pushing to make UID compulsory, and is now moving towards charging for use of authentication services. This article asserts that the three key selling points of UID – corruption control, inclusion and portability – have been achieved in several states in major welfare programmes, by creatively using simpler technology that is free of the hassles attached to UID.


The sales pitch for the ‘Unique Identification’ (UID)1 project had three important planks – corruption control, inclusion, and portability. All three claims were based on a poor understanding of the relevant programmes. Recent experience in several states suggests that the same goals can often be achieved with simpler and more appropriate technologies without the hassles attached to the UID project. On the other hand, UID application pilots initiated by the Unique Identification Authority of India (UIDAI) and others have faced serious challenges. Yet, the government has been pushing towards making UID compulsory, and worse, is now moving towards charging for use of (free thus far) authentication services.


Corruption in welfare programmes
A key claim made by the advertising blitz to ‘sell’ the UID project was that corruption would be eliminated, especially in the Public Distribution System (PDS) and the Mahatma Gandhi National Rural Employment Guarantee Act (MNREGA). As far as corruption in the PDS is concerned, Tamil Nadu has been running a clean system without UID. Intelligent applications of simpler technology (for example, computerisation, SMS alerts etc.), along with other reforms, have contributed to the turnaround of the PDS in Chhattisgarh and Odisha. In these states estimated leakages of PDS grain have come down from around 50% in both states, to 10% and 20% respectively (between 2004-2005 and 2010-2011).


In MNREGA, the scope for corruption was greatly reduced when wages began to be paid through bank and post office accounts in 2008-20092. Yet, corruption continues in the following three ways: coercion, collusion with labourers, and collusion with post office officials. Coercion occurs when labourers’ wages are forcibly ‘shared’ with corrupt officials after being withdrawn from the bank or post office. Collusion can be between implementing officials and labourers: days of work are inflated to defraud the system. In such cases, UID authentication cannot help. Collusion also occurs between MNREGA functionaries and post office officials (“identity fraud”), who operate the labourer’s account without his or her knowledge. It is only in this case that the UID-authentication can control corruption (Even here, identity fraud can be controlled further by switching to banks).


Welfare programmes and social inclusion
Two, UID was supposedly a project of ‘social inclusion’ for the millions who were excluded from social welfare programmes because they do not have an identity. It is true that millions of Indians are excluded from the government’s social welfare programmes, but the main cause for exclusion is not the lack of identity. The main reason is the measly budget for such programmes and exclusion errors. Rather than the budget being determined by the number of people in need, the budget determines the size of the target population. To make things worse, there are serious exclusion errors whereby poor people are classified as non-poor for the purpose of the PDS. UID alone can neither remedy the budget constraint, nor can it reduce exclusion errors.


The National Food Security Act (NFSA) increases the coverage under the PDS (from about half of the population to two-thirds). Independent of the NFSA, there has been a trend towards greater coverage (or, ‘inclusivity) as state governments, one after another, have rejected the central caps on coverage. Starting with Tamil Nadu, Andhra Pradesh and Himachal Pradesh, which run universal or quasi-universal PDS, Chhattisgarh, Odisha, Rajasthan and even Jharkhand have expanded their PDS by committing state funds.


Portability
The third claim of UIDAI was “portability” of benefits. In the current system, most entitlements are tied to the recipient’s place of residence. When people migrate, they are forced to forego their entitlements. For instance, one can only draw subsidised PDS rations from the PDS outlet to which one is tied. Indeed, the ability to run a PDS which is ‘portable’ would be a great improvement.


The Food Department in Chhattisgarh took up the challenge of designing a portable PDS, and in April 2012, they launched “CORE PDS”. Ration card holders have been given a “smart card”. Smart cards (like ATM cards) can be used to draw rations at any CORE PDS outlet which has a “point of sale” (POS) machine. The POS machine is, in fact, a mini-computer with connectivity. Once a smart card is inserted, the machine authenticates it, and sales can proceed. Offline sales are also possible when there are issues related to the server or mobile connectivity. In such cases, the ration card number needs to be entered for sales to proceed.


Since it was launched in two shops of Raipur, the system now covers 400 PDS outlets of which 40 are rural outlets. About one-fifth of transactions at these shops have used the portability option – ration card holders chose to go to a PDS outlet other than the one to which they were originally allocated. In most cases, portability was exercised within the ward (The density of PDS outlets in Raipur seemed very high.) Convenience was the prime reason for switching: the outlet to which people were originally tied was not the nearest, tenants who moved residence, when supplies ran out, etc. To cut a long story short, portability has been achieved without UID.


This raises the question of choosing between UID and smart cards for portability. Smart cards are a tested and affordable technology. In Chhattisgarh, it cost Rs. 50 ($0.8 approx.) per cardholder to implement smart cards (on the other hand, UID enrolment of each person costs Rs. 50). Apart from cost considerations, UID has raised other issues: surveillance and tracking, data security, personal and civil liberty concerns, privacy etc. (Ramanathan 2013 and Drèze 2011).


UID pilots’ crash
While state governments have been using simple technology creatively, UIDAI’s pilots have faced serious challenges - often crashing. A pilot in Mysore in the state of Karnataka, linking UID with Liquefied Petroleum Gas (LPG) cylinder purchase was aborted as it did not take off at all. In Jharkhand, an attempt was made to integrate MNREGA payments with UID. Notwithstanding tall claims in media reports, on the ground, the pilot turned out to be miniscule: for example, it was implemented in just three Gram Panchayats of Ranchi district, one of three pilot districts; hardly any payments had been made through the new system, and almost always under great scrutiny for the benefit of a visiting dignitary (Bhatti 2012). The pilot was quietly wound up.


The only ‘successful’ UID pilot is the integration of PDS with UID in East Godavari (Andhra Pradesh). It required nearly two years of preparatory work, a large and dedicated team with dynamic leadership (provided by the sub-Collector). In September 2012 the pilot was launched in 5% of PDS outlets of the district. It has not seen any scaling up since. It did not offer portability.


If the ration card and UID number match, the electronic-POS (e-POS) prompts the card holder to place one of their “best fingers” for real time authentication. Five trials are allowed, failing which, a one-time password (OTP) is issued. OTP is an “override facility” when biometric authentication fails (either due to biometrics, connectivity etc). Data from the first year suggest that OTP had to be used in 13% of all transactions. This ‘failure’ rate is not small, but has not been discussed at all.


Yet the UID juggernaut rolls on: in November 2012, the government announced Aadhaar-enabled cash transfers for existing schemes. Seven months later, a review at the Prime Minister’s Office (PMO) acknowledges that preparatory work was complete for only 9% of all beneficiaries. Undeterred, in October 2013, the UIDAI put out its ‘pricing policy’ approach paper for authentication services which are free only until December 2013. Private and public entities will be charged for iris, fingerprint and demographic data authentication, electronic Know Your Customer (e-KYC) etc.


Concluding remarks
Against its success in generating more than 50 crore unique numbers, there are many questions, even failures, from the UID pilots. Meanwhile, state governments using other technologies innovatively have demonstrated that corruption can be controlled; inclusion and portability can be achieved without the hassles that are a part of the UID package. Evidence, it seems, does not matter.


Notes:

Unique Identification (UID) or Adhaar is a 12 digit individual identification number issued by the UIDAI on behalf of the Government of India which aims to serve as a proof of identity and residence through India.
In fact, MNREGA has contributed much to financial inclusion even before UID existed.


Further Reading

Bhatti, Bharat (2012), “Aadhaar-enabled payments for NREGA workers”, Economic and Political Weekly, 8 December, Vol 47, No 49.
Drèze, Jean (2010), ‘Unique Facility or Recipe for Trouble’, The Hindu, 25 November.
Khera, Reetika (2011), “The UID Project and Welfare Schemes”, Economic and Political Weekly, 26 February, Vol 46, No 9.
Khera, Reetika (2013), ‘Lessons from the East Godavari Pilot’, The Hindu, 11 April.
Ramanathan, Usha (2010), “A Unique Identity Bill”, Economic and Political Weekly, 24 July, Vol 45, No 30.



- See more at: http://www.ideasforindia.in/article.aspx?article_id=250#sthash.hc8I7HK3.dpuf