The consensus among traders and foreign exchange analysts in the currency markets is that the rupee is overvalued by about 9-10 per cent. This warrants depreciation of the rupee if exports are to remain competitive.

The perception is based on the index of five-country real effective exchange rates (REER) that the RBI computes and releases with a lag of a month. (There is a 36-country index as well, which comes out with a longer delay.) The index value for June was about 108 and would have increased to about 110 in July.

My colleagues Gaurav Kapur and Rashi Shah argue that if indeed, export competitiveness is the issue on hand, the Reserve Bank of India's [Get Quote] five-country REER might not be the appropriate guide.

In fact, going by an index that they compute and somewhat unimaginatively christen the competitiveness index (CI), the overvaluation is likely to be less. From the average level of 2,000, the five-country REER was up by 11 per cent in July 2005.

The CI, by their estimates, had appreciated by just 3 per cent. There were, however, periods in between when the CI showed a greater degree of overvaluation than the REER.

However, from January 2005, while the REER went up continuously, the CI has persistently shed weight. Thus, India has gained competitive strength. This might just explain why the RBI isn't fretting much over the rupee's persistent strength against the dollar.

How does the CI differ from the REER? It follows the RBI methodology except in one critical respect�the choice of currencies.

The REER considers two things in measuring competitiveness -- the actual exchange rate versus a particular currency and the difference in prices between the trading economies.

The intuition is simple -- an appreciation in the exchange rate makes the exported commodity or service more expensive in a buyer country.

The same holds for rising price differentials -- if, say, Indian prices rise faster than a trading partner's prices, we lose our competitive edge. The REER then combines them by taking a weighted average.

The weighting corresponding to each combination of exchange rates and relative prices reflects the share of the trading partner in the export or trade basket.

The RBI's five-country index considers India's major trading partners in its basket. These are the United States, the United Kingdom, France, Germany, and Japan.

The CI, however, claims that the competitiveness of India's exports depends on how the rupee is faring versus major competitors, not its trading partners. (Crisil, the rating agency, incidentally has a similar index but does not follow the REER methodology closely.)

Thus, India's exports of, say, textiles to the US, do not depend much on what is happening to the Indo-US price-differential or the rupee-dollar exchange rate.

Export growth is affected by how the rupee has fared against a major competitor's currency, say, the Chinese yuan and how local prices have changed compared to Pakistan.

Gaurav and Rashi then go on to identify eight major competitors for India and compute an index based on their exchange rates corrected for the price difference. The weighting used for each exchange rate is the share of that country in global exports.

The intuition that variations in the CI would explain variations in export growth better is borne out by some simple statistical measures.

The simple correlation between export growth and the CI is much stronger at about 0.4 than the correlation with the REER of roughly 0.15.

In fact, if China is dropped from the CI, the correlation improves to 0.56. The relationship incidentally holds ideally with a three-month lag -- that is, if the CI moves up or down, exports respond with a three-month lag.

As expected, the big dent in competitiveness comes from China, which until recently had a pegged currency and lower inflation rates than India.

If China is excluded from the basket, the index actually shows a mild degree of undervaluation. China incidentally has the largest share or weighting (0.36) in the index.

There are a couple of lessons that I have drawn from the CI. First, from an export perspective, things don't seem to be as bad as the official REER would suggest.

Thus, the need to depreciate isn't perhaps that dire. This is important in the current scenario in which oil is trading at $67 a barrel. The more we depreciate, the larger the quantum of inflation that we "import".

Thus, even if supply and demand fundamentals were to result in downward pressure on the currency, I would argue for some degree of intervention by the RBI to prevent too large a decline.

The most important lesson from the CI? The "quality" of India's exports has improved. Here's my rationale for this claim. The currency may be less overvalued at the moment than generally perceived but a value over 100 means that is far from competitive.

Yet exports have grown at close to 20 per cent over the past three years. In fact, the CI rose sharply right when the export boom began.

Clearly, the sharp growth in global incomes led by the US and China has played a role particularly in jacking up the "realisations on commodity exports".

Equally important is the fact that items like pharmaceuticals and engineering goods have seen a rising share. These are associated with higher degrees of value addition and are hence less exchange-rate sensitive. This in effect reduces the volatility of exports and Indian manufacturing on the whole.