As a rule, the government likes creating new structures without fully understanding their implications and then disbanding them once criticism inundates the newspaper columns. A lot of time and money is invested in creating these structures and, often, these costs could be higher than the cost they are trying to lower to begin with. The direct cash transfer (DCT) scheme runs a similar risk, since our enthusiasm levels are currently high, and we could go off the track unless certain preconditions are addressed. It is generally felt that DCTs are a more efficient system than, say, physical subsidies. This does hold when conditions are ideal and back-end structures are in place. Otherwise, there could be contradictions that will make the DCT scheme unsuccessful. DCTs come into play for two kinds of transfers. The first is where a new structure is created for transferring cash-for-cash transactions. This holds for, say, salaries, pensions and scholarships and so on. The existing scheme has various departments sending cheques to the recipients, who, in turn, deposit them in their own accounts. The second pertains to cash-for-kind transfers. Here, instead of providing the good to the household, a cash transfer of an equivalent amount takes place and can be used to buy the product. The concept of DCT is based on the much-publicised Aadhaar project where a unique identity (UID) has been provided to people. Since every UID has an account linked to the person, such a transaction would be automatic provided the disbursing authority is linked with the banking systems. Given the volumes involved, this would be a logistical challenge. The advantage for cash-for-cash transactions is efficiency and reduction of leakages provided the identification process is robust. Prima facie, there is nothing amiss here. When it comes to cash-for-kind transactions, the situation is different because we have to give up the existing structures since substitution takes place. There are essentially five "Ss" that have to be tackled before bringing about any change in the transfer system. The first is "structures". We have an elaborate procurement system for food grain that is motivated by, one, procurement for distribution and, two, creation of a buffer. The procurement policy is an open-ended one where farmers can sell a fair average quality to the Food Corporation of India (FCI) at a predetermined price. The idea here is to protect the farmer's income. Have we thought of what will happen to this policy or FCI (an institution set up for this purpose) when we provide cash transfers, and FCI will then have to address only the issue of buffer stocks? Second, "systems" have been created for distribution - the public distribution system (PDS). If we have a "conditional cash transfer" in which money given has to be used to buy grain from fair price shops, then the status quo would be preserved - along with the current inefficiencies. However, if it is not a conditional transfer system, then new issues emerge. There are around 500,000 fair price shops across the country that on an average employ one million workers. By introducing cash transfers and disbanding PDS, there will be an issue of unemployment, since it will be hard for these people to reinvent their stores that are mostly located in rural areas. Today, when there is opposition to foreign direct investment in retail, we are talking of the local kirana shops.
There will be a lot of noise when we think of displacing these one million workers. Do we have a solution here? Third, "selection" is an important consideration for a successful DCT scheme. The problem with PDS, besides the ubiquitous leakages, is adverse selection. A lot of people who are not poor take in these entitlements. This becomes acute as we move to kerosene and liquefied petroleum gas. The new scheme on UID is no different from the existing policy of self-declaration; since no proof of income is asked for it runs the risk of adverse selection. In fact, there are a large number of people holding on to the coloured ration cards and not drawing rations. In the new dispensation of the scheme, this could mean free money for them. Do we have a way of screening households or else will we be back to also helping those who do not require assistance? Fourth, the government is talking aggressively of food "security" with an ambitious target of covering two-thirds of the population. Clearly, there is a major contradiction here. If we are to provide cash transfers, then how do we reach the food grain to the needy, which requires PDS? Fifth, there has been debate on the food "subsidy" burden. The subsidy is the difference between the economic cost and the issue price for wheat and rice. The economic cost varies between Rs 17 and Rs 24 a kg, and the issue price is around Rs 5 to Rs 8 a kg. This is when the food grain is sold at a fixed price. Now, once the people are paid cash, they have to buy food grain on their own from the market. Based on government data, the price of wheat and rice varies from Rs 15 to Rs 35 a kg in different parts of the country. Two practical problems arise here. The cash to be paid in lieu of subsidy will be substantially higher than the present subsidy amount. Second, with inflation being variable, fixing the prices and, hence, subsidy level across states will be difficult, and one can see a lot of politics coming in the way of arguing for higher levels of allocations. To make the DCT scheme effective, we need to fix these five "Ss" first or else we would be running conflicting parallel systems. We also need to evaluate the exact benefits of the cash-for-cash transfers before embarking on the more onerous cash-for-kind transfers. Besides, the cash-for-cash transfers alter the mode of payments without addressing the issue of selection. It is, therefore, advisable that we move one step at a time and not get carried away.
The author is Chief Economist, CARE ratings. These views are personal