Direct transfer mechanisms were devised to effectively deliver income transfers and transfer payments (in the Indian context, think old age pensions and scholarships for the underprivileged). Some transfers are conditional on a socially desirable action - e.g. ensuring children attend school and are vaccinated, as in the famous Brazilian Bolsa Familia programme.
How does this work for subsidies? Typically, subsidies are embedded in the price. That is, the price for the beneficiary is lower than for all others e.g. wheat for the poor is sold at, say, Rs 4 per kg when the market price is Rs 20 per kg. The DBT solution is to dissemble the subsidy (Rs 16) and deliver it through the hi-tech infrastructure. At one stroke, this results in a single market price for all, while the intended beneficiary receives the subsidy, thus leaving her effectively facing the same price, namely Rs 4 per kg. The single price regime (SPR) destroys rent-seeking opportunities and any possible distortions in resource allocations.
To focus on the DBT mechanism, assume that Aadhaar coverage is universal, bank accounts are functional for all and the hi-tech infrastructure linking all these is in place.
Let's begin with LPG, on which annual subsidy is Rs 22,000 crore at today's oil prices. First off, note that the 'unsubsidised' market price is below cost, with upstream refining companies picking up a separate subsidy bill (no SPR). DBT will indeed deliver the subsidy. But this begs the logically prior question: should LPG be subsidised? This is not a subsidy for the poor; it is a grant to the non-poor (read rich). Poor villagers with LPG connections: seriously? It is Rs 2,500 per household today; and would have been Rs 8,000 but for the crash in oil prices. If so, isn't it simpler to just increase the "subsidised" price through pre-announced increments? The re-direction of the marginal rupee saved to programmes actually intended for the poor yields higher social welfare. Moreover, should oil prices start rising, the subsidy will automatically rise. Better a gradual increase now than any shock tomorrow.
Consider fertilisers, with an annual subsidy of Rs 73,000 crore (with Rs 70,000 crore in arrears). The subsidy is non-conditional in terms of eligibility. All farmers (kulaks and smallholders) can buy fertilisers in whatsoever quantities at the subsidised price. DBT will work and deliver the subsidy. However, the SPR may not work: there will always be an arbitrage opportunity - i.e. an eligible beneficiary can always purchase more than she needs and strike a deal with a third party for side-payments. And, the DBT does not address the problem of fertiliser overuse because the price faced by the farmer does not change. So, the huge subsidy on urea (over 70 per cent) continues, prices relative to other fertilisers do not change and the N-P-K balance remains unaddressed. Once again, why not simply move to a pre-announced reduction of the subsidy? Decide on the level (in percentage terms) of subsidy for N, P and K and chart a price correction path. Since there is no intention to dismantle the subsidy altogether, DBT c
an continue. The key change required for DBT to be effective is the imposition of an eligibility condition.
Next, kerosene, with an annual subsidy of Rs 14,000 crore at current oil prices. The finance minister has rightly announced the phase-out of this subsidy through the Public Distribution System (PDS) except where electrification is yet to take place. Once again the solution is DBT: but how to identify areas that are not electrified? (What are the criteria?) And till that hurdle is crossed, the subsidy regime continues. Isn't it simpler to just have pre-announced reductions in the subsidy? This could be made to coincide with the government's ongoing programme to electrify all remaining villages in 1,000 days.
Lastly, food, with an annual bill in excess of Rs 100,000 crore. DBT can work, but, at present, is not operational. Requests to state governments to move to DBT fall on deaf ears; economists call such requests incentive-incompatible. Why shift from the present regime? Identifying eligible beneficiaries is costly and willy-nilly compels state governments to take action that is politically unpopular. With the noblest of political will, it will take at least three years (that's how long it took on LPG) to put the DBT mechanism in place.
Now relax the heroic assumptions on Aadhaar coverage, inter-linkages and glitch-free infrastructure. That in itself sets us back three years (probably more). In effect, DBT as a subsidy-reform instrument is considerably weakened. Lastly, note that Aadhaar certifies identity not eligibility. Without credible lists of eligible beneficiaries, where is the reliability of DBT in reaching targeted beneficiaries?
This is not a critique of DBT. It can and does work well for some purposes. And, it can and will stop leakages provided eligibility is clearly spelt out. However, in the hands of politicians, it has become a means of evading critical public policy choices and postponing actual reforms. It has enabled a dodge on issues like: who should get subsidies, how to determine and independently verify eligibility, how to deal with unintended consequences of subsidies (fertilisers), what is the sustainable level at which subsides can be given, and how to stop rent-seeking. DBT is the go-to instrument to signal political commitment to reform; when it is clearly understood that this is just kicking the subsidy-reform can down the road.
Politicians have used DBT to certify their pro-reform credentials. In reality, this is a diversionary tactic to postpone the difficult discussions and hard decisions that lie at the core of subsidy reform. The UPA did it. Now, the NDA government is doing it. Plus ca change plus c'est la meme chose. (A French epigram that means, "the more things change, the more they stay the same.")
The writer is a former Union commerce secretary