BETA may not be good measure for Indian Markets: ET

ET Intelligence Group tries to find out how powerful is the forecasting ability of beta in the context of the Indian stock market. Our sample comprises stocks of 44 companies, which have performed most consistently over the past decade.

It includes companies like Reliance Industries, Tata Steel, HDFC, Hindustan Unilever and Colgate-Palmolive from the fast-moving consumer goods (FMCG) sector; Ambuja Cements and ACC from the cement sector; auto majors like Mahindra & Mahindra and Tata Motors; engineering giants like ABB, L&T and Siemens; besides leading stocks from pharma, financial services, hospitality and information technology sectors. This has been done to ensure that all sectors in the economy are duly represented in the sample.

To check its effectiveness, we have taken the beta at the start of a year and then observed how the stock fared vis-Ã -vis the Nifty in that year. For instance, the beta of ABB was 0.76 on December 31, 1998.

This indicates that ABB's stock was expected to rise less than the Nifty and fall less than the Nifty. In 1999, ABB's stock price fell by 49.7%, while the Nifty was up by 66.2%. The beta logic did not hold in this case because if the beta is positive, both the stock and Nifty should move in the same direction, instead of the opposite direction. The exercise was repeated for 44 stocks over 10 years from the start of calendar year 1999 to '08 till date.

We ended up with 440 observations, as there are 44 stocks over the course of 10 years. The logic of beta holds for only 172 of such observations, which implies a success rate of 39%. This shows that the odds are against the investor if he takes a call based on beta. In some cases, the success rate can be even worse. For ABB, the basic rule of beta holds good only for one year out of 10. Beta is positive for all 10 years. The beta logic does not hold in years '01 to '08 because though the beta is less than 1, ABB's stock has shown higher volatility. This shows that estimating one-year returns considering the beta at the start of the year can be a self-defeating exercise.

Upon further analysing data, we found that stocks with a beta higher than 1 ' i.e. stocks which rise more with every rise in the market and fall more with every fall in the market ' have given close to three times the returns of stocks with beta less than 1. To establish this, we made two portfolios of stocks: Portfolio 1 includes stocks which had a beta higher than 1 for a major part of the past 10 years. It includes stocks like Reliance Industries, Larsen & Toubro, Tata Steel, ACC, Tata Motors, Wipro et al.

Portfolio 2 includes stocks which had a beta of less than 1 during most years in the past decade. It includes stocks like Asian Paints, Hindustan Unilever, Hero Honda, Colgate-Palmolive and Indian Hotels.

While the high beta portfolio has yielded a return of 19.7% per annum on an average, the low beta portfolio has given only 7.1% return per annum in the past 10 years.