The central government has announced the conversion of 29 poverty schemes to Direct Cash Transfers. Should this be implemented through banks or the postal system? This article assesses the pros and cons, and recommends experimenting with different systems.

The central government has announced that 29 government poverty schemes in 51 districts would be converted to Direct Cash Transfers (into). These schemes include pensions for senior citizens, scholarships, maternity benefits to pregnant women, and MNREGA wage payments. The direct cash transfers are designed to go into the bank accounts of the beneficiaries linked to Aadhar. The success of the initiative depends on improving the last mile connectivity and it is crucial therefore that the banks become accessible enough to rural beneficiaries to be a part of the scheme. There is some anxiety among many that it may be premature to contemplate such a plan when the banking network and other infrastructure in rural areas are conspicuous by their absence. Do the banks have a natural incentive to deliver cash to the poor and needy when they are likely to withdraw it as soon as it arrives leaving little for the banks to invest and earn returns? Isn’t it better to do it through the existing network of post-offices? In this article, I intend to consider these questions.

First, it is important to note that presently the payments are taking place through the post office or the brick and mortar bank branch and the outcome is far from being satisfactory - monthly pensions do not get delivered regularly, pregnant women end up getting what are supposed to be pre-natal benefits after the delivery, MNREGA wage payments are regularly delayed, and often some of the benefits are siphoned off. The relevant question therefore is not whether the proposed system is imperfect but whether it is likely to deliver an improved performance. Direct Cash Transfers are meant to correct the problems mentioned above in reverse order.

Let us first understand why these problems arise under the present system so that we can assess the proposed solutions. The source of delays in payments can be at the top or bottom of the bureaucratic hierarchy but more likely the bottom. The funds for the central government schemes come down from the centre via electronic transfers to the state agency and then down to the district or block level. However, the village post offices are not equipped with electronic fund transfer system and the movement of funds toward the village beneficiaries slows down, as now a cheque has to be issued to the village post office. The village postman disburses the funds to the intended beneficiaries and this involves further delays. Or the beneficiary has to travel all the way to the bank branch mostly situated at the block place. Moreover, there is no effective mechanism by which the beneficiaries are informed of either the arrival of funds or the amount. These problems in the last mile connectivity can be perhaps solved through the technological solutions implicit in the proposed channel of delivery. Now let us take the questions one by one.

(1) Since a network of post offices already exists, why should we incur the expense of setting up a rural banking network from scratch?

(This is not the same question as put in first paragraph above) Firstly, with the Banking Correspondent (BC) Model of RBI, the Brick and Mortar branches are not needed for offering banking services. Instead, the same services can be supplied by a single person (bank correspondent) equipped with a battery operated Micro ATM which runs on a mobile SIM card. This takes away the necessity of electricity and broadband and works with any SIM card that works in that location.

Secondly, implementing the cash transfers through the postal system would require an amendment of the Postal Act that limits the disbursement by a Postmaster to Rs. 20,000 in a single day. Since 2008, the Ministry of Rural Development has been attempting to amend the Act unsuccessfully. Moreover, it would entail equipping the postal network with the infrastructure needed to carry out electronic fund transfer that the banks are already equipped with. Moreover, the postal bureaucracy itself is reluctant to accept new technology. Lastly, the rural post offices are often nothing more than the homes of the postmaster and therefore not equipped to deal with sophisticated technological changes.

(2) The poor beneficiaries are likely to withdraw most of what they receive, leaving little in their accounts for the banks to earn any returns. What incentives do the banks have in supporting the project?

First of all, the conclusion that poor beneficiaries of schemes such as MNREGA would withdraw the entire amount is unwarranted. Say, two members of a household work on a MNREGA site for a week. That would bring them an amount not less than Rs.1200. It is not likely that they would want to take out Rs. 1200 at once; their consumption expenditure for a week would be much less than that. If partial withdrawals by beneficiaries were allowed that would leave something in the banks, and thus keep the banks’ incentives in tact.

Second, the incentives would depend on how the system is designed. If the UID platform is used for identification then the basic investment required by the banks for identification would not be necessary. If the banks do not have to incur these fixed costs, they would be more inclined to be conduits for direct cash transfers.

Third, the banks can be paid on ‘per transaction’ basis creating incentives for them to maximise the volume of transactions.

Fourth, if the banks take a long-term perspective, it should make a lot of sense for them to cover the rural countryside and register a wide swath of the rural poor as their account holders. Once they enter the rural area, they can also attract some of the rich farmers whose deposits could generate substantial profits for the banks.

Note that the banks use electronic fund transfers through their network. The information technology part can ensure transparency and leave an audit trail with date stamp. This can help show the delays immediately. For too long, the poor have had to live with a different system than the rich. For example, the poor have used money orders through the postal system while the rich have used the banking network. The system that the rich use works better as the rich have connections that help them hold the officials accountable, while the poor lack such clout and suffer on account of the dysfunctional system. This sort of segregation has been detrimental to the interests of the poor. Indeed, there are externalities for the poor when the rich too use the same system. The net result of the proposed schemes could be that the poor would get to be a part of a better performing system than the post office.

(3) How can we ensure that the Bank Correspondents do their job effectively and in a timely manner?

In the last mile connectivity there are two problems: the beneficiaries do not get informed in time of the arrival of the payment (or even the amount), and the agent (e.g. postman) responsible for disbursing the funds may demand ‘bakshish’ or take a cut from their legitimate dues. The first of these problems can be tackled by establishing a system that would automatically send out a SMS to the beneficiaries informing them of the arrival of the payment. This may not be possible, as many beneficiaries may not have mobile phones. However, if even one of the beneficiaries has a phone, the word could spread. This is not a perfect solution but with electronic fund transfer system all the way through the last mile, there are possibilities that do not exist otherwise. Banks do offer advantages that the post offices do not. The hand held ATM machine does speak to the beneficiaries telling them the amount due to them and in this respect unlike a postman a BC cannot take advantage of the ignorance of the beneficiaries. As far as the demand for ‘bakshish’ is concerned, the BC does not necessarily have an advantage over a postman.

One could argue that appointing a local village youth rather than an outsider as a BC for the village may be more convenient. For instance, the villagers would not have to trek to a taluka town to receive their payments. A village person may be held accountable more easily than an outsider. However, this claim can be contested on the ground that the advent of panchayat raj has not solved the problem of lack of accountability - there is a great deal of graft and inefficiency by local officials. Would giving power to disburse cash on a regular basis confer political power on the BC that he or she could exploit? We do not know. It is an experiment worth trying.

Yet, there are different models of BC’s possible. For example, in Andhra Pradesh, private firms such as Zero Mass and FINO were contracted to undertake the BC’s tasks. In other words, they offer a service similar to that of a courier service, except that they courier funds. However, in AP, these firms found the activity unprofitable for some reasons. In a UID based system, the government will bear some of these costs and improve the profit margin. An interesting question is who should monitor these firms – the banks or the government? If the ‘per transaction’ payment system to the banks creates natural incentives for the banks, it would make sense for the banks to do the monitoring. The bank management would then find it in their interest to curb unseemly behaviour by BC’s who would be their agents.

As is evident, all these untried methods of delivering direct cash transfers are not guaranteed to succeed but they have a lot greater promise to improve the performance of the poverty schemes than the current system. It goes without saying that it would be better to experiment with different systems to sort out what works and what does not, rather than applying the same system to the whole country. And lastly, from a development perspective, it would be a great step forward to financially include so many who had been excluded before, which would be added bonus to using banks rather than the post office for cash transfers.