magician

I've been looking at a number of financial websites recently - some relatively naïve, some relatively sophiticated, some in between - and I've discovered something extremely alarming (or, at least, disappointing): almost all of them have a warped understanding of beta. To be sure, they usually start out fine when they try to define it or describe it or give examples. But then comes the moment when they'd be better off keeping quiet - but they cannot manage it - and they include something really stupid.

So, who here thinks they know what beta really is, what it does, how it's used, and so on?

Are you game?

whiteflash

My junior college understanding of beta is very simple, and un-complex. I'm positive I don't fully understand it though:

Beta attempts to rank a stocks (or portfolio) rate of return, in regards to its market. My understanding is that 0 = Independent level of correlation between a stock and it's respective market, Positive = A level of positive correlation between a stocks rate of return and it's market, Negative = A level of opposite trends between a stocks rate of return and it's market. If memory serves me right, I remember my professor bitching about how a lot of times this is misconscrewed to somehow represent a stocks volatility, when that is not the case.

e.g. Intel / Tech

Intel Beta = 0. Intel stock trends show that it does not rise or fall with the market, it's ROR fluctuates independently.
Intel Beta = 1+. Intel stock trends show that as tech market increases, Intels ROR follows suit.
Intel Beta = -1-. Intel stock trends show that as tech market increases, it's ROR decreases, or vice versa.

As far as I'm aware Beta is typically used in long term investment measuring, thus is more of a prediction to stock vs market down the line. I.E. If you think in 10 years tech will be down, invest in a cheaper tech with a negative beta. Beta does not attempt to predict short term volatility however, although it's sort of easy how that could be ignorantly misconscrewed. [Read: Ignorantly]

I could be wrong though

Incubus

I thought it had to do with the degree of volatility. A higher value would indicate a more volatile stock.

Of course, I don't go around telling people this.

Malloric

It has to do with both volatility and correlation. A very highly volatile asset with negative correlation would have a large negative correlation. The tricky thing, is that a highly volatile asset that has very little correlation could have the same beta as a much less volatile stock that is highly correlated.

B=Cov(r_a,r_p)/V(r_p); where r_a is the return on the asset, and p is for the portfolio. In most cases, you're talking about the market as a whole instead of a portfolio, but same principle.

What it is is the easy part. That's a fact. How it's used... well, generally it's (miss)used as a measurement of risk. "I don't like risk, so I only buy stocks with a beta of .6 or less." That's sort of like saying, "I don't like vegetables, so I play a lot of tennis." WACC, Hamada, CAPM, I'm sure there are lots of uses for it.

magician

Yes, yes, yes! (Though you should have specified "of returns". Many people believe it has to do with volatility of prices; it doesn't.)

I think you meant "a large, negative beta".

Exactly!

That's the definition, but it's not as useful as it could be. If you multiply and divide by the standard deviation of returns of the asset, and do a bit of algebra, you get a more useful formula:

β = ρ(ra, rm) x (σa / σm)

Beta is the correlation of returns of the asset and the market, times the relative volatility of the asset's returns and the market's returns. This highlights your first sentence: it covers correlation (of returns) and volatility (of returns).

Amen!

Not exactly, but you clearly understand the problem.

It's scary how many people who should understand what they're doing misuse beta. I've read two articles recently that said that low beta stocks are less volatile than the overall market, and several definitions of beta that say the same thing. In both articles, all of the low-beta stocks the mentioned had, in fact, higher volatility of returns than the S&P 500, but the author described them as "low risk" (and in one, said, in so many words, that they wouldn't take you on the rollercoaster ride that the market has; one stock had more than three times the volatility of returns as the S&P 500).

Malloric

Correct on all counts, magician. I've tried to explain that beta just determines the risk floor and not the ceiling. Nobody cares about the risk floor. I mean it's of academic interest, I suppose, but in practical terms? Useless. What good does knowing a stock is at least as volatile as its beta tell you?

There's also the broader point that volatility alone isn't a good way of minimizing risk, at least not if you also care about returns, which is a given if you're in equity markets.
And then you've got the whole fundamental/technical debate. Not touching that.

I've pretty much given up on journalism. The quality is just not there. Part of the problem is mass-media has gotten so good there's almost no production cost any longer. When it cost a few cents to disseminate information, people would hesitate to spend a few pennies disseminate crap. Now that it doesn't, why not? The game is quantity, not quality. Fire the staff reporter and hire the lowest cost freelancer. What incentive does the freelancer have to write anything but drivel? Not to mention the reader doesn't want to deal with it. Low beta, low risk. Short, sweet, to the point. Next article.

magician

Beta tells you something about a security's systematic risk, but little (other than, as you point out, a floor) about its overall risk. Yet people try to use it as a proxy for overall risk.

SlowTeg

Amen to the death of journalism. I'm a little late to this thread, but here's another great piece of "journalism" from bloomberg today :

Headline: Wall Street Bailout Returns 8.2% Profit Beating Treasury Bonds
http://www.bloomberg.com/news/2010-10-20/b...treasuries.html