Impact of cash transfers in an economy with large fiscal transfers

For its extraordinary, almost global size, India's flagship National Employment Guarantee Scheme (NREGS) remains one of the least evaluated of anti-poverty programs anywhere in the world. For example, how has the massive NREGS cash transfers to rural consumers affected wages and local price levels? More generally, what is NREGS contribution to India's persistent food inflation? Or more specifically, how is the additional disposable income generated by NREGS being spent?

Unfortunately, there is no rigorus enough empirical study of the impact of world's largest cash transfer program on rural wages and resultant inflation. In a related context, Jesse M. Cunha, Giacomo De Giorgi, and Seema Jayachandran compared the relative local price effects of cash and in-kind transfers by studying a large food assistance program in Mexico that randomly assigned villages to receive boxes of food (trucked into the village), equivalently-valued cash transfers, or no transfers and found,

"Both types of transfers increase the demand for normal goods, but only in-kind transfers also increase supply. Hence, in-kind transfers should lead to lower prices than cash transfers, which helps consumers at the expense of local producers...

The price increase caused by cash transfers, based on the point estimates, offsets the direct transfer by 6 percent for recipients who are consumers of these goods. Meanwhile, for in-kind transfers, the price effects represent an indirect benefit to consumers equal to 5 percent of the direct benefit. Thus, choosing in-kind rather than cash transfers in this setting generates extra indirect transfers to the poor equal to 11 percent of the direct transfer. Of course, the welfare implications are reversed if transfers recipients are producers rather than consumers.

We also find that agricultural profits increase in cash villages, where food prices rose, more so than in in-kind villages where prices fell. These effects are due both to the change in the price of goods sold, but also to households responding by producing more (less) when the price of what they produce increases (decreases)."

The study also finds that price effects were particularly pronounced for very geographically isolcated villages, where the most impoverished people live. This is consistent with the fact that these villages are less open to trade and have less market competition, and are therefore more likely to be supply constrained in case of cash transfers. In these cases, the in-kind transfers actually increases supply and lowers prices.

The authors point to two issues that needs to be factored in while calculating the price benefits. One, their study does not look into what kind of effect is generated in the long-term, when the higher prices would signal to increase local production, thereby easing supply constraints. In fact, its long-term benefits are far more than that arising from external in-kind supply.

More importantly, there is the issue of how much does in-kind transfers constrain households' choices or conversely how much does cash transfers increase households' choices. As the authors point out, it is quite possible that an efficient private sector would create more surplus than if the inefficient government were the supplier. So they suggest that the best alternative would a mixture of cash transfers and policies to ease supply-side constraints.

Extending this analysis to NREGS and India will yield interesting possibilities. In India, we currently have a deeply supply constrained market where inflation expectations are on the rise. In such markets, any cash transfer would do little to increase supply. It increases the cash available with consumers, without doing much to boost supply, atleast in the short- to medium-term. Increased prices are inevitable. This increase in prices affect both the specific commodity being subsidized and the general price level.

Consider a cash transfer in place of rice supply through the Public Distribution System (PDS). Assume a family requires 60 kg of rice per month and gets cash equivalent of 35 kg of rice. Let us also assume that the subsidies are calibrated to price variability. Let us assume that there is only one variant of rice available in these remote markets and its price is Rs 15 per kg before the new cash transfer scheme is introduced. In supply-constrained markets (and remote interiors are classic examples of supply-constrained markets), which also experience government fiscal spending, prices generally increase and the following two effects are observed.

1. The beneficiary gets only a portion of his rice through the PDS. He has to purchase the rest from the market at market prices. In this case, he purchases 25 kg from the open market. After the cash transfer scheme is introduced, the price of rice increases to Rs 20 per kg (since the PDS supplies, which is not an insignficant share of total supply in such small markets, is now not available and has to come from the general market supply). He still continues to get cash equivalent to 35 kg. But now he has to shell out an extra Rs 125 per month for the same amount of rice the family was consuming before the program was introduced.

In contrast, with in-kind transfer, let us assume that the price falls (or it could remain the same) by say Rs 2 per kg after its introduction. This in turn leaves the farmer with a savings of Rs 50 per month. The difference between the two programs, with these assumptions, is therefore Rs 175.

2. There is also the likely spill-over effect on the general price level due to the increase in the price of rice. Further, the additional disposable incomes generated by way of NREGS and the resultant higher rural wages, will increase the demand for other items, mainly meat and other protein foods. Econ 101 would tells us that, when supply remains the same (which is likely to be the case with most products, atleast in the medium term), additional incomes (or higher aggregate demand) will have the effect of increasing the general price level.

All this in turn increases the burden on the family by say, Rs 100. Taken together, both these effects have the effect of reducing the real income of rural household by Rs 225 per month. The combined effect of cash transfers in an NREGS context is captured in the graph below. Note that it is possible that even the actual aggregate consumption could fall, rather than increase, especially if inflationary pressures get out of hand.



This highlights the importance of easing supply-side constraints in ensuring the effectiveness of any cash transfer scheme. In fact, taken together with NREGS, cash transfers could, without policies to increase supply, exacerbate inflationary pressures. Whatever the analysis, India's biggest obstacle to growth and successful implementation of process reforms that can increase growth, is a deeply supply-constrained economy. Its inflation problems too are just a symptom of supply bottlenecks.

Cash transfers and negative income tax credits

Cash transfers are the flavour of the season in India. The two commonly discussed cash transfer strategies are direct cash transfers to replace subsidies (as in case of food grains, kerosene, fertilizers, cooking gas etc) and conditional cash transfers to incentivize social outcomes (immunizing, sending children to school, maintaining nutritional standards etc).

In both cases, it is proposed that the cash can be transferred using the Aadhaar identity number to the beneficiary's Aadhaar-linked savings bank account. Though this will not completely resolve the problem of beneficiary selection, the issues of pilferage (by ghost and duplicate beneficiaries) and administration of the transfers will be satisfactorily addressed.

Apart from the two aforementioned types of cash transfers, there is a view, still marginal, that advocates direct payment of cash to all those below the poverty line. Such a universal minimum income guarantee, it is argued, should replace all subsidies and, if delivered through biometrically validated Aadhaar-linked accounts, will simplify program administration, minimize leakages, and limit incentive distortions.

I am not interested in this post to get into the relative merits of the three approaches to cash transfer. Needless to say, there are some very formidable challenges with a universal minimum income guarantee cash transfer. However, if this is adopted, the most effective strategy to implement it would be to package it as a negative income tax (NIT) scheme.

An NIT is reimbursed to the income tax assessee whose income falls below the basic minimum income level for which no one pays income tax. Just as tax payers pay tax at a percentage of their positive taxable income, NIT assessees can be reimbursed, into their Aadhaar-linked accounts, at a percentage of their income deficit (or shortfall from the basic minimum level at which tax kicks-in, the negative taxable income).

Milton Friedman had first proposed the NIT in the late sixties to replace all other welfare programs for the poor. A variant of this, the earned income tax credit (EITC), was introduced in the US in the seventies for the working poor.

The NIT has several benefits in the Indian context. Apart from the numerous benefits arising from dispensing with all other subsidies, the NIT, by making every individual file his income tax returns, would be big step in legitimizing all income streams. Though there will be the risk of people under-reporting incomes, it will open up a large part of the massive parallel economy in India.

The formidable challenge, as mentioned, will be to get people to report their incomes with reasonable degree of accuracy. Rigorous analysis of the massive database so created would itself provide ample information that can further uncover the parallel economy.

In any case, this would be a definite improvement over a direct minimum income transfer where cash would be transferred without any pre-conditions to all the identified beneficiaries. But with NIT, the government would be benchmarking all transfers to the reported incomes of the individual.

The price volatility problem with cash transfers

I have blogged earlier that price volatility, especially with food grains, is the most difficult challenge with cash transfers. One of the strategies to overcome this is to provide a large enough subsidy, inclusive of a volatility buffer, so that even if price increases, the buffer will cover for the increase. The nutrient-based subsidy (NBS) for non-urea fertilizers under implementation since April 2010 uses this strategy to cover for price volatility.

It is therefore instructive to examine the experience with NBS in fertilizers. Under this model, fertilizers are sold to farmers at its decontrolled MRP (minus the subsidy) and the subsidy amount paid separately to the fertilizer companies. The nutrient-wise fertilizer subsidy is fixed once a year, based on its market and import prices, and with a reasonable buffer to cushion the producers against price fluctuations over the year.

What has been the progress report with this subsidy regime in operation for more than a year now? A recent report in Businessline highlights the sharp volatility in the prices of fertilizers, as reflected in the manner in which the subsidy fixed for 2011-12 (operational from April 1, 2011) was sharply revised between November 2010 (when it was originally fixed) and March 2011 (when the subsidy amount was frozen). It now appears that even this may have to be revised.

Recent imports of DAP have been for $612 per tonne (or Rs 27,540 per tonne), which with duties, freight and other charges will retail at about Rs 30,735 per tonne. Against this, the MRP of DAP is currently at Rs 10,750 a tonne and the NBS payable from April 1, 2011 is Rs 18,474 (fixed assuming a landed import price of $580 a tonne), thereby grossing the company Rs 29,224 for every tonne of DAP. This deficit of more than Rs 1500 a tonne or Rs 75 a bag will either have to be covered through an increase in subsidy or passed on to the farmer.

Given important state elections coming up, it is almost inevitable that despite the original decision to fix the subsidy once a year, the subsidy rates will be revised again. Similarly, even on MoP, where global suppliers are apparently pushing for a landed price of $420 a tonne (which is more than the current reference rate of $390 a tonne), a subsidy revision looks certain.

Further, there is also the problem of market expectations that get formed by the subsidy price announcements. Since India is one of the largest fertilizer importers, its procurement decisions and prices signals get embedded into the global market prices. Though the subsidy rates announced includes a volatility buffer, the market prices naturally get anchored at the upper end. As the Businessline report says,

"For 2010-11, the Centre had fixed the NBS rates on nitrogen (N), phosphorus (P), potash (K) and sulphur (S) with reference to corresponding import prices of $310 a tonne on urea, $ 500 on DAP, $370 on muriate of potash (MoP) and 190 a tonne on sulphur. For 2011-12, the Centre – in a bid to talk down world prices – initially, on November 19, pegged these reference prices lower at $280, $450, $350 and $125 a tonne. But with global prices showing no signs of easing, the Centre, on March 9, announced new NBS rates for 2011-12 based on higher landed prices of $350 for urea, $580 for DAP, $390 for MoP and $180 for sulphur."

The Businessline report also quotes industry sources who claim that the government's decision to tinker with the DAP reference price has contributed to the steep increase in its import price in recent months. Also, the NBS regime has had no effect on lowering the subsidy burden. In fact, subsidy burden is also projected to go up by Rs 30,000 Cr to Rs 82, 245 Cr, up from the budgeted provision of Rs 52,837 Cr for 2010-11.

The experience of fertilizer highlights the difficulty of managing price volatility and also ensuring access to fertilizers (or food grains) at affordable prices. The much greater price volatility and political sensitivity associated with food grains means that any buffer will have to be much larger. Far from lowering it, the food subsidy outgo will therefore increase.

Admittedly, with fertilizers, given the reasonably integrated national market, it may be possible to calibrate subsidy to a price index. However, this approach cannot work with food grains whose prices vary so widely across the country at any point in time, that it is impossible to capture it in one index with any reasonable degree of reliability. So what is the way forward? More food for thought!