One of the major factors behind the transformation of India's current account deficit, from a deficit of 3.1 per cent of GDP in 1990-91 to a surplus in 2001-02, was the increasing prominence of invisibles receipts making them almost as significant as earnings from merchandise exports.

Invisibles (net) inflows have increased manifold on account of India's burgeoning services, specially IT and ITES exports, and heavy remittances from Indians working abroad.

Indeed, from a deficit of $242 million in 1990-91, invisibles earnings (net) have increased to $40.9 billion in 2005-06. The past five years (that is, 2001-02 to 2005-06) have seen such earnings grow at an annual average of 28 per cent. As a proportion of GDP, invisibles (net) exceeded merchandise trade balance (net) in 2001-02, leading to a surplus in current account after more than three decades.

However, from 2004-05, the trend has again reversed with trade deficit overshooting invisibles surplus, turning the current account into a deficit that crossed $10.6 billion and amounted to 1.3 per cent of GDP in 2005-06. Indeed, after 1999-2000, this was the first year when the CAD became larger than 1 per cent of GDP.

India and Thailand are the only leading economies of developing Asia running CADs, while China, Indonesia, Japan, Malaysia, and Philippines, are all generating surpluses. A CAD, per se, is a good sign as it reflects a net inflow of foreign savings supplementing domestic savings and can thereby be used for financing investment requirements of a fast-growing economy.

It is, however, important that the CAD be financed by stable long-term capital inflows (that is, FDI). In India, debt-creating flows (that is, external assistance, commercial borrowings and short-term trade credit) and short-term portfolio flows have become dominant in the capital account since 2004-05.

Indeed, the share of debt-creating and portfolio inflows in total capital receipts has increased from around 48 per cent in 2001-02, to almost 75 per cent in 2005-06. It is, therefore, important to not only monitor the CAD, but also the quality of capital flows financing it.

In this context, it becomes important to examine how large the CAD can be in 2006-07? We attempt to answer the question by projecting four different scenarios. Our projections are based on an assessment of disaggregated current receipts and payments, both merchandise and invisibles, for the past five years (2001-02 to 2005-06).

On the basis of this past performance, we assume year-on-year export and import growth of 23.5 per cent and 28.7 per cent, respectively (in dollar terms).

One of our key assumptions is regarding crude oil prices. While projecting an import growth of 28.7 per cent, we have assumed India's crude oil import basket price at $70 per barrel for 2006-07, and a volume growth of 7 per cent for POL imports.

For invisibles, our key assumptions pertain to software exports and private transfers. We assume growth rates of 28 per cent for software earnings and 5 per cent for private transfer receipts.

This is based on the NASSCOM projections indicating annual growth between 27 and 30 per cent for IT and ITES exports for 2006-07, despite such exports growing by 33 per cent in 2005-06. We also expect India to remain one of the highest remittance-receiving countries in the world.

This is based on growth of private transfers rising from $2.1 billion in 1990-91, to $24.1 billion in 2005-06. We project private transfers (net) to touch $25.2 billion in 2006-07.

We have assumed GDP at current market prices to grow by 12.5 per cent during 2006-07. Our assumptions are based on a real GDP growth rate of 7.5 per cent and an annual inflation of 5 per cent. This is a somewhat conservative estimate for GDP growth.

Scenario 1: Scenario 1 yields a trade deficit of $62.1 billion, which, offset by an invisibles surplus of $51.1 billion, produces a CAD of $11.0 billion, or 1.2 per cent of GDP.

Scenario 2: We retain our earlier projections of export and import growth (that is, 23.5 per cent and 28.7 per cent, respectively), while assuming software earnings to grow at a higher rate of 33 per cent. We get a higher invisibles surplus of $52.6 billion, producing a CAD of $9.6 billion, which, as a proportion of GDP, is marginally lower at 1.1 per cent, vis-�-vis scenario 1.

Scenario 3: We change our assumption for POL imports in this scenario, while retaining export growth at 23.5 per cent and the invisibles projection of software earnings growth at 28 per cent and private transfer receipts at 5 per cent. While in the earlier two scenarios we had assumed the import price for the Indian crude basket at $70 per barrel, here, we assume price at $74 per barrel, given the unrest in the Gulf.

Maintaining the volume growth at 7 per cent (as in the earlier scenarios), this implies a 43 per cent growth in oil import bill (compared with 35.2 per cent earlier) and 31 per cent growth in overall imports. Indeed, during April-June 2006, oil imports have grown by 39 per cent, and are likely to grow further during the year.

Assuming the above, we have a trade deficit of $65.6 billion, which, offset by invisibles worth $51.1 billion, results in a CAD of $14.5 billion. This is the highest CAD among all our scenarios at 1.6 per cent of GDP.

Scenario 4: We retain exports growth at 23.5 per cent, imports at 31 per cent (as in Scenario 3) and assume the invisibles projections of Scenario 2 (that is, with software earnings growth of 33 per cent). The CAD reduces by about $1.5 billion to $13 billion, and to 1.5 per cent of GDP.

Non-customs trade balance is a critical assumption for all scenarios. This item reflects the difference between merchandise trade figures reported on customs and BoP basis. In recent years, the difference has become strikingly large, with the gap being as much as $10.7 billion in 2004-05, and almost $12 billion in 2005-06.

We have assumed a smaller difference of $6 billion for the current year as we expect the ongoing efforts for reconciling the two data sets to yield positive results.

Our worst-case scenario yields a CAD/GDP that is 0.3 percentage points higher than 1.3 per cent recorded in 2005-06. A marginally higher CAD that supports a stronger manufacturing sector growth, which seems to be the case going by the 10.9 per cent growth in manufacturing during April-May 2006, will be quite acceptable, even welcome.

In any case, we can live with an even higher CAD, as a proportion of GDP, provided, of course, it is financed predominantly by stable long term capital flows like FDI.