Cash transfers, whether conditional or unconditional, are getting a lot of attention in India these days. The success of cash transfers program in the some parts of the world, especially Latin America, and not so much of success of India’s own schemes and welfare programmes, have led many to think about cash transfers more seriously.

Unfortunately, there is not much empirical evidence on the success of  cash transfer programs in the Indian context. One simple reason is that there aren’t many in the first place. There are a few cash transfer programs at the central and the state level but the lack of concrete data and rigorous evaluation of these programs has meant that we can’t really say much about the effect of these cash transfer schemes, in general. We know a lot about the problems in say, distributing food through public distribution system but that does not necessarily mean that the cash transfer programs will work.

A recent paper in EPW, ‘Small but Effective: India’s Targeted Unconditional Cash Transfers’  (by Puja Dutta, Stephen Howes, Rinku Murgai, December 25, 2010, Vol. XLV, No. 52) compares a cash transfer program (pensions to elderly, widows, disabled) with a non-cash transfer program (PDS) on basis of parameters like coverage, targeting and compliance, and draws some interesting conclusions.

Government pensions are provided to the poor mainly a) elderly, b) widows, and c) severely disabled. These pensions make up at the most 4% of total spending by the central government on various safety net programmes. On the other hand, subsidized food programmes constitute close to half of the spending on safety net programmes.

Coverage:

The analysis in the paper shows that the coverage of the pensions is very low. Even among the poorest quintile, the coverage reaches only 10%. Lack of awareness, lack of knowledge about the application process, complexity of the process might be some of the reasons for such low coverage. On the other hand, coverage of the PDS is much higher, reaching to more than 30% among the poorest quintile.

Targeting:

The pension scheme is meant for elderly/ widows who are poor. The authors’ calculations suggest that the scheme is progressive in the sense that proportion of pension recipients decline as we move up the wealth quintiles. The targeting performance of the PDS is also quite similar.

Compliance/ Leakage:

The authors identify five possible types of non-compliances in case of the pensions and try to get an estimate of leakage due to each of these reasons (with a focus on the state of Karnataka).

The types of non-compliances are 1) duplicate records, leading to overpayment or under payment to fraudulent pensioners, 2) ‘missing’ pensioners i.e. genuine pensioners, who are ‘missing’ (either moved or died), 3) the recipients might not receive pension in full, 4) they might have to pay bribes at the time of joining or during the year to receive the pension, and 5) the recipients may not be eligible.

In Karnataka, the pension recipients’ list is computerised. Each recipient’s record contains name, father/ husband’s name, post office, as well as an identifier, which is meant to be but may not be unique for each household. An algorithm was run to assess the closeness of different records. Only 0.2% of records were virtually identical. 1% of records had very high similarity score (0.9 and above), and 6% a high score (0.8 or above). The authors assume that 1% of the records as duplicates (hence 1% leakage).

To get an idea about the extent of missing pensioners, the authors, through a survey team, tried to find all the individuals enrolled in the scheme. In 9% of the cases, the enrolled individual could not be found (hence 9% leakage).

The authors also collected data on amount of pension actually received by the pensioners. They found that on average, pensioners receive 96% of their pension (hence 4% leakage).

Those who applied for pension, but unsuccessfully, paid an average Rs. 200. Assuming that the successful recipients paid the same bribe on average, it adds 2% to leakage (see the paper for calculations). It was also found that 8% of pensioners had paid a median amount of Rs. 100-200 in the last 12 months to various officials to receive the pension. This adds less than 1% to total leakage.

This gives us about 17% leakage.

On the other hand, it has been reported that the consumption of grains from the PDS shops was only 46% of the grains supplied to them in 2004-05. Leakages vary from state to state. In Karnataka, the corresponding ratio is 64%.

To summarise, the PDS coverage is much higher. Both do well in terms of targeting. But the pensions fare way better when it comes to the extent of leakages.

Now the relevant question is: What would happen to leakages if the coverage of the pensions goes up? The answer depends on reasons for low coverage in present. The authors put forth two possibilities: a) low levels of discretion in the pension scheme i.e. once enrolled (may be after paying bribes), payments follow more or less automatically.  There is little scope for diversion of funds. On the other hand, for the PDS, ‘every extraction of benefits from the system requires effort and is a potential rent seeking opportunity’; b) both the payments involved and the number of recipients are small, and hence not worth targeting for corrupt individuals.

Both are plausible and have very different implication for leakages if the pension schemes are scaled up. Hence the authors suggest progressive scaling up of these schemes, combined with effective monitoring, which might be a tad easier for the pensions, where the data indicates that the leakages are likely to be concentrated in specific areas, and relatively easier to spot in administrative checks and surveys, especially after computerization.

Ambrish Dongre is a Senior Researcher with the Accountability Initiative.