Yes, it's that time of year again, when every self-respecting columnist feels obliged to tender fiscal policy advice to the finance minister and his overworked colleagues in the ministry.

Well, who am I to shirk obligations? I shall limit my comments to the tax policy side of the Budget.

First, at the risk of repetition, let's briefly review some contextual background. Thirty years of global tax reform have been dominated by a handful of big themes.

Run-up to the Budget: Complete Coverage

First, corporate profits should be taxed at a flat, moderate rate (in the range of 25 to 35 per cent), with the tax base being as free of exemptions and incentives as possible.

Similarly, personal income tax rates should be modest, with revenue buoyancy being sought by broadening the tax base through good compliance and minimal exemptions. Some progressivity in rates remains the norm despite the growing debate on flat rate taxes.

Third, customs tariffs should be low and, preferably, at a uniform rate across commodities to promote sound trade policy.

Fourth, a single rate VAT up to the retail stage is the preferred form of domestic taxation of goods and services. Finally, a modern, IT-based tax administration is seen as a central requirement for a good tax system.

Ever since V P Singh's Long Term Fiscal Policy (1985) and the Chelliah Tax Reform Committee report (1991-92) these big themes have also guided efforts at tax reform in India.

There has been substantial, if slow, progress, occasionally marred by some backsliding. What should be done in the coming Budget?

Before giving some suggestions, organised by the major central taxes, one more contextual remark is in order.

Against a background of high fiscal deficits, soaring government debt, ambitious expenditure prorammes and the legal requirement (embodied in the FRBM Act) to reduce fiscal and revenue deficits, I assume that such suggestions should ideally be revenue-enhancing or, at worst, revenue-neutral.

More bluntly, we shouldn't assume that tax reform only means reductions in tax rates!

Central excise

The gradual move towards a single rate (16 per cent) CENVAT for all goods in a production chain, buttressed by additional special excises for a small number of luxury (income elastic) consumer goods, had been a laborious endeavour on the part of successive finance ministers through the 1990s, culminating in the Yashwant Sinha Budgets of 1999 and 2000.

Since then, and particularly with the Jaswant Singh Budget of 2003, there has been a loss of focus, with that Budget outlining a triple rate excise structure (8, 16 and 24) and failing to distinguish between the CENVAT rate and (additional) special excises.

As experience has shown, the legitimisation of the 8 per cent rate (as anything other than a temporary way station for goods being progressively moved up to the 16 per cent CENVAT) invites continuous lobbying for commodities to be moved down to this rate.

Moreover, the propensity to grant outright excise exemptions to various products has increased, undermining the basic logic of the VAT chain.

In the last two Budgets alone, computers, bicycles, tableware, toys, tractors, dairy machinery, hand tools and mosaic tiles have been totally exempt from excise for less than compelling reasons.

The coming Budget needs to reemphasise the centrality and (near universality) of the 16 per cent CENVAT rate by critically reviewing the present long list of products with full exemptions or concessional rates.

Nor does it have to shy away from levying (additional) special excises at 8 per cent or higher on a limited range of consumer luxury products. Not, at least, if raising revenue in a non-distortive manner is a serious objective.

There have been repeated attempts to bring the hugely important, but fragmented, textile industry into the CENVAT system.

The last attempt by Jaswant Singh in 2003 appeared to be reversed in Chidambaram's 2004 Budget.

The issue merits revisiting. The guiding consideration should be what kind of tax system will most help the Indian textile industry to compete successfully in the post-MFA world.

More generally, central excise revenues have shown disturbingly low growth in recent years, at rates well below the growth of output value in manufacturing.

The politically sensitive exemptions for small-scale industry cannot be the main reason (there is a separate and obvious economic argument for whittling those down).

More likely, there is urgent need to modernise the administration of central excise and find ways of checking the abuse of the CENVAT credit procedures.

Customs duties

During its six-year tenure, the NDA government reduced "peak" customs tariffs from 40 per cent to 20 per cent. However, the number of commodities (especially agricultural) with tariff rates higher than the "peak" has grown over this period and needs review.

The trade policy case for further reductions of customs tariffs "to ASEAN levels" remains strong. But for revenue reasons the decline has to be gradual.

Actually, there is still substantial dispersion of tariff rates between zero and 20 per cent. It should be possible (and desirable) to reduce the peak rate to 15 per cent, while raising all (or most) unbound rates below 10 per cent to that level, without any loss of customs revenue.

Revenue could also be augmented by pruning the very large number of exemptions (often varied by end-use), which today complicates hugely the customs tariff structure.

According to press reports, the Lahiri committee on taxation of petroleum products has recommended continuation of a significant gap between the tariffs on oil products and the (lower) one on crude.

If so, this is a pity as it perpetuates the grant of heavy effective protection to the low value-added petroleum refining industry. There was a time in the mid-1990s when these rates were broadly unified.

The economic case for reunification in the coming Budget is strong.

Direct taxes

The personal income tax structure has become opaque after the last Budget's attempt to relieve taxable income up to Rs 100,000 from tax without altering the existing exemption limit of Rs 50,000 or the slab structure.

In effect, the 10 per cent rate for the slab Rs 50,000-60,000 has become redundant. And the curve of marginal tax rates has acquired curious kinks.

A cleaner, more transparent structure would be to raise the exemption limit to Rs 75,000, apply the 10 per cent rate to the slab Rs 75,000--100,000, a 20 per cent rate to the bracket of Rs 190,000-200,000, 30 per cent to the range Rs 200,000-10 lakh, and introduce a new rate of 35 per cent for taxable incomes above Rs 10 lakh (Rs 1 million).

This is not really a new slab, given that those earning more than Rs 850,000 are already paying a marginal rate of 33 per cent because of the surcharge.

This proposal would have the important added benefit of equalising the top personal income tax rate with the general corporate rate and thus encourage present non-corporate business forms (like partnerships) to corporatise.

The present general rate of 35 per cent for companies is reasonable and should be retained.

The real challenge is to reduce the large number of exemptions for various sectors and kinds of income, which reduce the effective average rate of company taxation to below 25 per cent.

Tinkering with MAT (minimum alternate tax) may help but it could easily be counter-productive.

Our current taxation of capital and capital gains defies economic logic and breaches basic notions of equity. But the structure has been jerked around six times in the last eight years! For that reason alone, I would let it be.

Finally, beware schemes for voluntary disclosure and "using black money productively". They reward the dishonest, undermine compliance and rarely augment public revenues in a genuine and durable way.

None of the above is the stuff of "dream budgets". But perhaps these suggestions reduce the risks of the real nightmare taxes � low growth and inflation.