So far as the budgetary process is concerned, the rate of growth of revenue/current expenditure has been alarming. Continuous rise in revenue expenditure, financed by borrowings is dangerous since the debt incurred does not create an asset, which will yield returns to pay back the debt.

Instead, borrowings used for revenue expenditure lead to higher interest burden the next year, which causes next year's revenue expenditure to be higher still and requires further borrowings. This spiral is not only threatening to but has already gone out of hand.

If the government expenditure to GDP ratio had been rising on account of increase in capital expenditure, it would not have been a cause for as much worry. So long as it is ensured that capital assets created yield a reasonable return over a period of time the debt incurred in creating these assets could be retired over a period of time.

Even if the asset itself does not yield sufficient returns, it would likely boost growth and raise tax revenues, which could be used to repay the debt. Therefore, not only is a debt incurred for asset creation likely to be self-financing, the asset itself will boost future growth of the economy.

Given the situation that the budgetary position finds itself in, fire fighting measures are called for. It is well known that the three major factors responsible for rising expenditure are interest payments, subsidies and defence.

Interest payment accounts for 30 per cent of total expenditure (Budget 2000-01), 36 per cent of revenue expenditures and 51 per cent of revenue receipts.

Subsidies account for about seven per cent of total expenditure, eight per cent of revenue expenditure and 11.5 per cent of revenue receipts.Defence accounts for 17 per cent of total expenditure. We have computed defence only as a proportion of total expenditure since some part of defence is also covered by capital expenditure.

Thus, these three items account for 54 per cent of total expenditure; interest payments plus subsidies account for 44 per cent of revenue expenditure and about 53 per cent revenue receipts.

These numbers serve to indicate the lack of flexibility that the government faces with respect to financing development expenditures. Of the three items the government has very little discretion over two of these, namely interest payments and defence.

Interest payments are an obligation on past debt and have to be met if government credibility is to be ensured. Only by reducing debt creation now can the government hope to reduce future interest payments.

The level of defence expenditure, on the other hand, is often dictated by external considerations. In the wake of the Kargil episode it was only to be expected that spending on this score would rise. With two of the three major expenditure heads offering little or no flexibility the finance minister was left with only one choice: Pruning subsidies to reduce revenue expenditure.

In the context of subsidies, it is important to remember that there are two kinds of subsidies: Explicit subsidies and implicit subsidies. Explicit subsidies are those that are mentioned clearly in the budget (see for example, Annexure 3-1 in the Expenditure Budget 2000-2001, Volume 1).

In addition, there are implicit subsidies that have to be estimated from the revenue and expenditure sides of the budget. Implicit subsides arise from the gap between the cost of providing a public good or service and the revenue earned by the sale of this good or service.

Various studies have shown that such subsidies are as important as the explicit subsidies. The finance minister, in this year's budget, has focused on explicit subsidies only.

Two kinds of subsidies have been reduced in Budget 2000-01: Food subsidies and fertiliser subsidies. Food subsidy has gone down from Rs 9,200 crore in 1999-00 (revised estimates) to Rs 8,100 crore (2000-01), a fall of about 12 per cent.

Fertiliser subsidy on indigenous urea has gone down from Rs 8,670 crore to Rs 8,058 crore, a fall of seven per cent. On the other hand, fertiliser subsidy on imported urea has gone up from Rs 80 crore to Rs 500 crore, an increase of 525 per cent.

When subsidies are given as well as when they are cut it is important to be clear who the beneficiaries of subsidies are. It is also important to be clear regarding the reasons for which subsidies were given in the first place.

Subsidies may be given for those activities where social benefits are available over and above private benefits. This is a justification of subsidies from the efficiency point of view. Thus giving food subsidy may be justified on this score: The recipient is able to buy enough food to keep hunger at bay. In addition, from society's point of view, a person with adequate amount to eat is also more productive. Apart from this there is also a strong distributive reason for subsidising food for those who cannot afford it.

The usual behavioural assumption in economics is that of selfishness. Given this there is a temptation for all individuals to avail of subsidised benefits (food) if they are under no compulsion to pay the full price for it. This results in the subsidy bill being extremely large, creating problems for the budgetary process.

Thus targeting of food subsidy has been attempted since 1998-99, when different subsidised prices were charged for those above and below poverty lines. There are two problems here. One, separating individuals into two categories on the basis of the poverty line. This separation will not be perfect since the government does not have enough information to do so.

Hence there will be wrong categorisation of individuals with its consequent injustice and lowering of individual and overall welfare.

The other problem is rather more serious and involves a political economy problem. The question may seem strange and the answer to it obvious.

However, the answer may not be as obvious as it seems. The questions is: Who are the true beneficiaries of food subsidies?

The true beneficiaries of food subsidy are actually large farmers and not the consumers of subsidised food! The quantum of food subsidy is determined by the economic cost of providing, say, wheat. The economic cost consists of (1) procurement price paid to farmers by the government (2) costs of storage, distribution and wastage of the Food Corporation of India (FCI).

The average proportion of these two components, as far as wheat is concerned, over the period 1991-92 to 1998-99, is 67% for procurement price and 33% for costs incurred by FCI.

The procurement price for wheat has been growing over the years at a rate of 14% per annum, while the rate of inflation over this period has been much lower at 8% per annum. Further, procurement price surprisingly bear a positive correlation with changes in production of wheat.

I would have expected that in years of decline in wheat production, procurement prices would have risen, while it would have been lower in years of increase in wheat production. In fact, the reverse has happened in India. Thus the procurement price has become an instrument for bailing out large farmers.

The net result of continuously rising procurement prices is visible in the food grain stock that FCI is forced to carry, which for the last few years has been more than double the minimum buffer stock norms.

Further, I suspect, the reduction in food subsidy this year will not lead to lower procurement prices. Thus consumers will suffer, but the true beneficiaries of food subsidy i.e. the big farmers, will continue to receive favourable treatment from the government.

The author is RBI Ambedkar Professor of Political Economy, Department of Economics, University of Mumbai.