It had to happen. The market meltdown, I mean.  Not because the Chinese market lost 8.5 per cent in one day. Not because the finance minister increased taxes on dividend distribution, or broke promises to give IT exporters a tax holiday, or increased the excise duty on cement supposedly to bring down inflation!

Any of these could have been an excuse for a sorely needed market correction. But these were not the reasons. Then what were? Was it the fact that the average Nifty stock was selling at 21.5 times earnings? Well, no and yes.

Why no? Let's look at a mythical consumer goods company - Hindustan Foods and Soaps Ltd. It produces a wide variety of mass consumption goods for the middle classes of India. Though its margins fluctuate with surges in commodity prices, it spends enough on marketing to build and maintain attractive brands.

HFSL's net profit margins tend to revert to eight to nine per cent of turnover, and its topline growth is consistent with the Indian economy, that is, about 15 per cent per annum (nine per cent GDP growth and six per cent inflation). Am I willing to pay 21 times earnings for such a company? Absolutely.

So why then did I say, "Yes, the market needed correction"? For the reason that the 50 companies on the Nifty include many that are not exactly like HFSL. For example, it includes cement companies (three), which are cyclical stocks with highly fluctuating margins and profitability.

Steel and aluminum stocks (three) fall into the same category. Banks, of which four are included in the Nifty, usually find it difficult to show high returns to equity, since increased business usually forces them to increase capital.

Further, ten of the Nifty stocks represent the public sector, where profitability can always be shaken by political decisions.

So, while I am prepared to pay 21.5 times earnings for HFSL (and for IT stocks, despite the MAT provisions), it would be too expensive for the motley bunch on the Nifty.

Further, the Nifty also includes high flyers like Suzlon. At its peak of Rs 1,500, Suzlon was selling at a little over 50 times earnings for financial year 2006. To justify this kind of P/E, a company needs to grow at above 40 per cent per annum for a sustained period. When the company skips a beat (as most do, once every so often), the street loses faith, and such stocks tank.

In other words, the recent correction was not about triggers outside the market; instead, it can entirely be explained by the relationship between company financials and their pricing.

At some time rationality has to set in. When it does, and over-priced stocks begin to be sold, the ripples spread - those who have borrowed funds to buy are the first to quit; first-time investors get spooked by even small losses; and if there is redemption in mutual funds, fund managers are forced to sell.

However, the downshift is not uniform. Even in a correction as sharp as the one between 26 February and 5 March, some stocks swim against the tide.  Britannia, for example, moved up eight per cent from Rs 1,200 to Rs 1,300 while the markets corrected by over 10 per cent. NIIT too held up well. At a broader level, between December 23, 2006 and March 5, 2007, the market lost 7.6 per cent, while my portfolio was in the black, even if only by 0.2 per cent!

Looking ahead, I am convinced my money is safe in HFSL. If the market drops further, I will buy some more. HFSL will find the right level irrespective of whether the market takes a while to steady, or recovers sharply.

But, there is no real HFSL. There are, however, a host of companies out there that are well positioned to cash in on growth - Indian and international. Which companies, what prices? In a companion piece, we throw up some ideas for you.

The author is an investment advisor to a select group of clients. He can be reached at msatyanand@yahoo.com