You are here: Home » Opinion » Columns
Business Standard

Can you trust long-term Sensex data?

Debashis Basu 

Debashis Basu

The stock market is down in the dumps. For the past six years, barring a handful of stocks in the consumer goods, pharmaceutical and financial sectors, most companies have not even delivered positive returns. Those who believed that stocks were a way to create long-term wealth are having a hard time keeping the faith. They advise that you should look at the long term. After all, didn't the Sensex rise by 16 per cent, compounded, between 1979 and 2013? The data provided by the Bombay Stock Exchange say that it has.

This number, 16 per cent, is liberally used by mutual funds, stockbrokers, financial planners and advisers to justify equity as a great long-term investment. I do view equity as the best option to beat inflation, but here is why the 16 per cent figure is wrong.

The Bombay Stock Exchange launched the Sensex in 1986. But it has calculated the Sensex data since April 1979 to provide a longer series, with a base index value of 100. It is on the basis of this that many analysts calculate historical returns for the Sensex. If the Sensex is quoted at 19,000, according to these data, the index has gone up 16 per cent, compounded, between 1979 and now. Only if the data are correct, that is.

For one, the Bombay Stock Exchange does not give year-wise Sensex data from 1979. The data for the Sensex that are freely available on the website of the Bombay Stock Exchange are from the year 1991 onwards. To get the data before 1991, you need to pay. According to what the Bombay Stock Exchange has generously shared with Moneylife, the Sensex rose from 124.15 as on April 3, 1979, to 991.26 as on January 2, 1991. The Sensex value as on April 3, 1979, was 124.15 and not 100, owing to the fact that the value of the index divisor has been arrived at by taking the average market capitalisation of all the 30 stocks for 1978-79. But are these numbers accurate?

From around 100 in 1979, the index moved up to around 500 in 1986 (according to data available with the Securities and Exchange Board of India), when the Sensex was actually launched. This would equate to an annual compounded growth rate of nearly 25 per cent in seven years! Is this credible? That would amount to a massive stock market bubble in the 30 top Indian stocks, almost equivalent to Nasdaq's dot-com bubble. Does anybody remember 1979-86 as the period of a great Indian stock market bubble? If you assume that the index value of 500 in 1986 is correct, the data for the reconstituted Sensex of 1979 are suspect. Or the base data are correct and the index value of 500 is possibly wrong.

However, not only did the Sensex go up four times in seven years, it jumped to 1,193.61 on April 1, 1991, from 500 in 1986 - a rise of 139 per cent in five years. How credible is that? Remember that this was one of the worst periods for the Indian economy: the Bofors scandal, the disillusionment with Rajiv Gandhi's government, V P Singh's venomous raids on top Indian businessmen, his own short-lived coalition government, his regressive pro-reservation policies, and L K Advani's Rath Yatra. It was an extremely difficult period for Indian businessmen, especially against the backdrop of galloping short-term foreign debt that culminated in near-bankruptcy for India in 1990. We had only two weeks of money for imports that year. To stave off the crisis, India had to pledge gold abroad. Now, should we believe that during this period, which was one of the most difficult periods in recent history, the Sensex was up 139 per cent, or 19 per cent, compounded?


Taking the entire period into account, the Bombay Stock Exchange would have us believe that the Sensex was up 19 per cent, compounded, annually from 124.15 as on April 3, 1979, to 1,194 in April 1991, a period of 12 years. If you believe the Bombay Stock Exchange, you will soon come across even more bizarre figures. According to the data, the Sensex moved up from around 124 in 1979 to a peak of 4,467 in April 1992, which translates into an annual compounded growth rate of 31 per cent.

One of the reasons why I strongly suspect that there are errors in these data is that the Bombay Stock Exchange does have a record of such blunders. The Bombay Stock Exchange reported that price-to-earnings (PE) figures for the Sensex were as high as 55-56 during the 1992 peak. On June 10, 1995, however, The Economic Times exposed this as wrong. In response, on June 14, 1995, the Bombay Stock Exchange dropped its PE to 42. This means an almost overnight drop in PE by as much as 25 per cent! Also, before September 1, 2003, the Sensex was calculated using the full market capitalisation of the constituents and dividing it by a number called the index divisor. And from that day onwards, the calculation method for the Sensex was changed to the "free-float market capitalisation" methodology. Under this, the level of the index at any point of time reflects the free-float market value of 30 component stocks relative to a base period. That leaves us with another thought. Is it fair to compare Sensex data before September 2003 with the data of the present index?

So next time someone tells you that equities are a great investment option because the Sensex has gone up 16 per cent, compounded, over the long term, take it with spoonfuls of salt. Once again, my argument is not against equities. I do believe that stocks must be bought to beat inflation, and for the long term. I am only against sloppy research that is widely shared.



The writer is the editor of www.moneylife.in
editor@moneylife.in

First Published: Sun, August 04 2013. 22:44 IST
RECOMMENDED FOR YOU