Is it time to dump the clich� about India's dire fiscal situation? The consolidated deficit of the federal and state governments put together for 2005-06 is likely to add up to anywhere around 7.6-8 per cent of GDP, compared to 10 per cent just four years ago (2001-02).

The improvement is likely to continue if the Centre gets back on the fiscal track set out in the Fiscal Responsibility Act and states continue to show better fiscal numbers.

The implications are fairly clear. Declining deficit means lower dis-savings of the government and pushes up the overall savings rate. The Central Statistical Organisation estimated the gross domestic saving rate for 2004-05 at a very respectable 29.1 per cent.

The current fiscal year is likely to see similar levels. More visible manifestations are in the government debt market. The central government is likely to end up borrowing Rs 133,000 crore (Rs 1,330 billion) this year against the initial target of Rs 141,000 crore (Rs 1,410 billion).

State governments are actually sitting on large cash surpluses and in a curious twist of fate have actually funded the federal deficit this year by parking their surplus in short-dated treasury bills of the federal government.

The story behind the improvement in the fisc goes thus: first, tax collections are pro-cyclical and tend to pick up over an upswing in the business cycle. The composition of growth is equally important.

Taxes, particularly central taxes, are highly leveraged on the manufacturing sector and the fact that manufacturing has led the current up-tick in the economy has helped. Service tax collection has also improved on the back of the expansion of tax base and strong growth in the sector.

Apart from this cyclical bounce in tax revenues, tax efficiency has also improved remarkably. That is, the tax collection per unit of GDP has picked up as compliance has improved. From a dismal 8.23 per cent of GDP in 2001-02, the ratio of central taxes to GDP moved up to 9.85 per cent in 2004-05.

If tax collections are broadly on target this year, one is likely to see a tax-GDP ratio of close to 10.5 per cent. Measures like the tax information network (TIN), a computerised system for tax payment and monitoring, are an example of how better tax administration can help shore up corporate tax revenues.

Despite their initial misgivings, state governments appear to have benefited enormously from the VAT and it is likely that the six states which did not adopt the scheme last year will jump on to the bandwagon in 2006.

I would, however, warn against ignoring the long-term risks inherent in India's public finances. For one, the demand for funds going forward from different ministries for different social sector and infrastructure-related programmes is enormous.

For 2006-07, the initial demand for assistance from various ministries for social and infrastructure programmes (like the Sarva Shiksha Abhiyan, Bharat Nirman, Employment Guarantee, Railways, the Rural Health Mission, Agriculture and the Highway Development schemes) apparently constitute more than 50 per cent over the budget support levels in 2005-06.

It is unlikely that the finance minister can cobble together these funds simply by weeding out or consolidating some of the existing schemes and transferring resources from them.

The finance ministry does not appear to have the option of pruning revenue expenditures either. The attempt to prune the food subsidy a couple of months ago had to be jettisoned because of opposition from within the coalition.

The scheduled reform in the oil subsidy regime is likely to meet the same fate. Thus, the only way for Mr Chidambaram to ensure fiscal discipline would be to offer less in terms of budgetary support. I sense that the increase in gross budget support to the Centre and state programmes this year would be in the ballpark of 15-20 per cent. The figure doing the rounds of the media is about 19 per cent.

There are some risks to this strategy. First, unless expenditures are prioritised prudently, lower support could mean that sectors like irrigation that are in dire need of funds get less than their optimal share of resources.

Second, given the ideological divide within the coalition, this would also make future fiscal choices contingent on the electoral fortunes of the UPA in the following way: election setbacks in the future would tend to be blamed on the finance minister's tightfistedness. If the finance minister indeed becomes the fall guy for the coalition's political misfortunes, then any adverse turn of events would intensify the pressure to loosen the exchequer's purse strings.

Let me turn to the risks that I see for state finances. State fiscal data is looking better for a couple of reasons. The buoyancy in tax collections has helped but state budgets have also had a helping hand from an extensive bit of fiscal re-engineering over the last three years.

The Twelfth Finance Commission has awarded them a higher share of resources, enabled them to refinance debt at lower rates and provided debt relief. The debt-swap programme of the last three years in which states swapped their high cost loans from the Centre has pared their interest costs.

However, states have been unable to address the issue of charging commercially viable user charges for services like power and irrigation. Consequently, the return on states' physical investments remains abysmally low.

Our estimates show that states' non-tax revenues (a proxy for user charges) as a percentage of GDP have shown a moderate decline from 2000-01 levels.

The effect of fiscal re-engineering and tax buoyancy will last awhile but long-term fiscal correction, particularly reduction in the revenue deficit, will be possible only if non-tax revenues are buoyant. A significant part of fiscal relief from the Twelfth Finance Commission is contingent on states being able to carry out significant fiscal reform.

If states cannot do much to shore up non-tax revenues, they might not be able to access this fiscal relief. This has the potential of reversing the entire process of fiscal reform going forward for the states.