05 Cost and cost effectiveness

There is as yet no cost benefit analysis, nor a business plan. Faced with this anomaly the minister for finance has resorted to hiding costs under different heads. The initial investment towards this is about Rs 20 billion (Rs 1 billion in 2009-10 and Rs 19 billion in 2010-11). This is in addition to an unspecified amount hidden in the census budget. Rs 100 was to be given the gullible, hapless poor is more a bribe rather than an incentive, preying on their poverty and ignorance using public money and is certainly not ‘free’.

Other services that would like to use Aadhaar for verification—for instance the banks—will need to set up the requisite infrastructure. Verification would mean biometric scanners in every ration shop, every bank branch, every school, gas agency, petrol pump… and then the broadband infrastructure and bandwidth to connect to the centralised database. The recruitment, training and hardware costs will add to economic growth, GDP and corporate profit while simultaneously pushing up transaction costs for the individual. This would mean higher costs for inclusion.

Just like the dams across the Narmada, which was started and billions of tax payers money spent on it without the planning commission approval or statutory clearances, UIDAI will also come back with the argument that ‘since a lot of money has been spent’ let it continue. Dismantling this infrastructure, disentangling it from other databases such as the passport etc will cost even more—as the UK has discovered. When the database is hacked, the biometric data collected (fingerprints, iris, face) will need to be replaced, or an other level of data should be added for authentication. Then the plea will be to add a DNA check to it ‘since so much money has already been spent’ else the database will have to be shutdown. This will mean additional expenses. And when that is hacked…

It is ironic that in his earlier avatar as CEO of Infosys, Nilekani would not spend a fraction of this amount with out a business plan, feasibility study or a cost benefit analysis. All of these are missing in this endeavour, resulting in an activity that does not address, much less solve, the problems that it is supposed to solve. It is not even a solution searching for a problem. It is creating a problem where there is none, fully knowing that someone else will have to clean up the mess. It is rather like Wall Street during the financial meltdown of 2008-9. They made the mess the way they made the money, and they made money cleaning up the mess. We will be straddled with this black hole of public expenditure since by then they will be too big to fail and no one will be there to take the blame.

In the use of digital technology, software plays an important part. Here the terrain is highly contested between the proponents of open standards and free and open software on the one hand and the proponents of closed and proprietary standards on the other. Since data is stored and needs to be accessed for a long time, it has to be in a standard that enables access for a long time. It cannot be dependant on the whims and vicissitudes of a company. For instance, if a person’s data is stored from birth to death, it will have to be accessible for about 100 years. Few companies have that kind of longevity. So unless the standards are open, the data may not be accessible if the company goes bankrupt or closes down. Security and privacy concerns are another reason for free and opensource software (FOSS). Aadhaar is not fully FOSS compliant.

The technology is fraught with risks, but more so the operations. As the Biometrics Standards Committee Report says (p22) there is also data to suggest that quality drops precipitously if attention is not given to operational processes…. Empirical data has highlighted several non-technical factors that can impact accuracy. The lacunae are simple operational quality assurance, missing biometric records due to poorly designed processes and biometric software be tuned to local data. They promise an accuracy of above 95%. The SHG federation women have a loan repayment record of close to 100% and, when the PDS outlet is entrusted to them, 100% accuracy in delivery.

Update February 2010

After a very public disagreement with the Ministry of Home Affairs which called the Aadhaar flawed , the UIDAI has 'taken a break' for six weeks from March 2012 so that they can study what went wrong. The officials admitted that in some sense, it was a large pilot project, and there was a need to pause and look at what worked and what did not. A rather costly field pilot study at over Rs 5,000 crores (50,000,000,000 US$1 billion)