According to the “Capital Asset Pricing Model” (promoted by several Nobel Laureates in Economics, including the late Merton Miller, a ratio called"beta" can be calculated for every stock. This ratio shows how a stock price has reacted IN THE RECENT PAST, to changes in a broader index (like the DJA or S&P 500).From what I understand, this ratio, being calculated from past data, cannot have any predictive power about the FUTURE of an individual share price. However, in looking at my insignificant portfolio, I am struck by how individual stockprices track the broader indexes…indeed, the corellation is astounding (someimes); for example, I own a small, local cororation…and over the past 3 weeks, its share price has tracked the NASDAQ index perfectly! Why should this be so?
And, can the BETA of any individual stock actually predict future pricing (at least over the next 2-3months)? Has anyone checked this out?
Of course, this wouldlead one to conclude that stock prices are predictable, which is disproved by experience, but, anyway…
No. The beta only would have predictive power if one could predict how the broader market was going to do.
The beta is a measure of risk relative to the market as a whole, usually reflected by the S&P 500 I believe. If the market goes up and the beta is greater than 1.0, then the stock (or mutual fund) will rise more than the S&P. However, it also works the other direction. If the S&P falls, the investment will fall even more.
If you desire less risk than the general market, you’d invest in instruments with betas less than 1.0.
Beta is simply a measure of the riskiness of a stock. If the market (which technically includes every possible investment in the world) moves up 1%, how much should you expect the stock to move? If the Beta is 2, then the stock is expected to move at the risk-free rate (or alpha - the rate at which the gov’t borrows money) + 2 X the market premium.
But, as you said, there is no reason that the future price movements of the stock should reflect any of this. In fact, Betas chnage over time. And they wouldn’t be great for making short term gains on stocks.
But many people forget about the purpose of Beta. Beta separates out market-related risk from idiosyncratic or company-specific risk. With efficient markets, company-specific risk can be diversified away. So the pricing of stocks depends only on market-related risk (in theory). You are not compensated for idiosyncratic risk. So while your hunch that Company X will get the patent that can make or break the future of the company, the price that you pay for the stock will not be discounted due to this risk.
I hope all that was clear.
To add a specific example to the very capable explanations above…
A stock for a company that was generally steady, despite downturns in the market (economy as a whole)… for instance… a hypothetical manufacturer of a staple food product (wheat farmer) would have a beta close to zero.
A stock for a company that manufactured a luxury chocolate, on the other hand, would be very sensitive to changes the the economy as a whole, and would have a high beta.
And (yes, there is such a thing) companies that did well when the rest of the economy is the pits (staying with my food example, maybe the company that makes Ramen noodles) would have a negative beta.
Hope that helps clarify the technical explanation above.
A Mathematician Plays the Stock Market, by John Allen Paulos talks about these technical indicators and the role they play in predicting stocks. (He wound up buying WorldCom, a decision that looked good at the moment.)
I haven’t read the book, but it’s getting very good reviews.
The main problem here is in the theory versus application. Beta measures sensitivity of a security to the complete market (including itself by the way). BUT, as people try to use it to PREDICT movement they have to approximate beta. They do this most often by taking the last 60 months of data and simply regressing it against the market’s movement (with the risk-free rate as the intercept). Past performance does not guarantee future results. Beta is theoreticaly the real-time sensitivity, not the last 60 month’s, but if you wanna predict with it then you gotta know it in advance.
Also, beta is often used to determine cost of equity (that’s the CAPM). The CAPM is a big deal, because it says that the only things that determine cost of equity are the following:
- rate of return of the market
- risk-free rate of return
- beta
and - Nothing else, no credit stats, no market cap, no pro forma no nothin’.