Recs

### 26

February 02, 2010 – Comments (11)

A brief description of why I think Beta values are misused.

Frequently when talking about investment returns, an individual will be confronted by a high brow academic with the question: “Yes that’s all very good, but what’s your RISK ADJUSTED return?”  This question is usually followed by smirk or a fist bump to his other MBA buddies.

Investopedia describes Risk Adjusted Return as: A concept that refines an investment's return by measuring how much risk is involved in producing that return, which is generally expressed as a number or rating.

While there are many ways of quantifying the risk involved in an investment, the Beta value is probably the most popular.

Investopedia explains Beta
Beta is calculated using regression analysis, and you can think of beta as the tendency of a security's returns to respond to swings in the market. A beta of 1 indicates that the security's price will move with the market. A beta of less than 1 means that the security will be less volatile than the market. A beta of greater than 1 indicates that the security's price will be more volatile than the market. For example, if a stock's beta is 1.2, it's theoretically 20% more volatile than the market.

Many utilities stocks have a beta of less than 1. Conversely, most high-tech Nasdaq-based stocks have a beta of greater than 1, offering the possibility of a higher rate of return, but also posing more risk.

Investopedia accurately describes both beta and risk adjusted returns and illustrates perfectly my first problem with the use of beta values.  I agree that beta values provide a measure of a stocks tendency to respond to moves in the market, I disagree that this translates directly into risk.

Before we get into a semantics argument about the meaning of risk, let me lay out my definition as it applies to investing.

Risk is the magnitude weighted probability that you will receive below average returns on an investment.

Lets use the market in 2009 as an example.  By March of 2009 high beta stocks were significantly undervalued because as the market fell through 2008, these stocks tended to fall faster.  Then starting in March as the market recovered these stocks far outperformed the market as they had more ground to recover before reaching a “fair” market value.  This time period clearly showed the positive or negative impact that high beta stocks could have on your returns.  But does it indicate that high beta stocks carried increase risk?  The results of myself and others have been downplayed with the argument that “anyone can be a hero who picked high beta stocks in March”.  If you see the beta value for what it represents, you see that this argument is as foolish as saying “anyone can be a hero who accurately identifies highly undervalued companies”.  The trick isn’t to pick high beta stocks, it’s to pick high beta stocks at the right time.  I would actually go one step further to say that the true trick is consistently pick undervalued stocks regardless of the time.

My second problem with beta values is that they correlate the historical price of a stock with a historical market index.  Besides the parallels this has with TA, as a risk measurement is lacks any fundamental component.  In my opinion, the real risk an investor has when buying a stock is that the company will not generate the free cash flow the investor planned on.  I would suggest a better risk measurement would be a correlation between free cash flows and the market index.  Essentially identifying how sensitive the company’s income generation is to broad market movement.

In conclusion, I believe Beta values provide investors exactly what they measure, and idea of how the stock has historically moved with the market.  I think they are inadequate as a measure of risk because they utterly fail to consider the fundamentals, or the position of the market.

I welcome any feedback or comments.

#1) On February 02, 2010 at 5:56 PM, TMFElmosWorld (63.34) wrote:

Great Blog! I agree that beta is widely misunderstood and incorrectly applied as a risk measurement tool. I'm not so sure however that I agree with your suggestion for correlating FCF's with broad market movements to guage risk. IN my opinion, market movements, especialy in the short run, are powered by  investor behavioral and pyshological forces (i.e. the collective emotions of the market) and thus I would expect them to have no effect on the cash generating capacity of businesses, unless of course those businesses applied models that generated cash based upon trading activity (Im thinking discount brokers here).

Report this comment
#2) On February 02, 2010 at 6:12 PM, goalie37 (88.50) wrote:

Great post.  I believe risk/reward to be of more relevance to bonds than to stocks.

Report this comment
#3) On February 02, 2010 at 6:16 PM, goalie37 (88.50) wrote:

To give an explanation of what I was trying to say, bond yields of the same maturity tend to vary depending on the perceived risk of the issue.  "Junk" bonds will have a higher yield than a treasury based on the idea that the interest payments are more in question.

When it comes to stocks, there will be many, many examples of high quality companies returning much greater capital gains (over long periods of time) than there would be of low quality companies returning outsized gains.  Therefore the risk of the bad stocks would not outpace the gains of the good stocks.

Report this comment
#4) On February 02, 2010 at 8:45 PM, ChrisGraley (29.31) wrote:

Love the post, and would like to add a little bit to the discussion.

Beta is just a tool and I agree it's an over-used tool by the MABS. (monkies in business suits)

Beta is not risk, it's volatility. It just shows that a stock has big swings up and down and that isn't always at the same time that the market goes up and down.

If you want to see if the stock moves up and down with the market then the next thing to look at is correlation. A stock that is correlated with the market will move up when the market goes up and move down when it goes down. A stock that is negatively correlated with the market will go up when the market goes down and visa versa. Another thing to look at is if the correlation is predicting or lagging of the market. A stock that predicts the market is very valuable. It helps you to re-evaluate your holdings before the market moves.

Now once you measure a stock against the market, it's also a good idea to measure it against it's industry. Also, it's a good idea to look at predictive and lagging correlation against the industry as well.

So now you have a toolbox with a bunch of tools, but there are a couple more tools that you can use. These are independent variables and these are the best tools in the box in my opinion and the hardest ones to find. In almost every industry there are specific things that make the stocks in that industry move up or down independent of the market. There will always be stocks in that industry that respond more than their peers. These are great for hedges against certain situations. Lets say I'm long a bunch of domestic infrastructure companies and I'm worried because a certain Middle Eastern country is saber rattling. I know that the government might postpone infrastructure spending if this becomes a conflict. The first thing I do is look for a correlation in the market for when this happened before. Sure enough, the industry tanked as soon as a conflict started.  Next, I look for a negatively correlated company to hedge my risk and there are a bunch of defense companies to choose from. Last I narrow my search to conflicts with just Middle Eastern companies and it shows me that ABC oil company (entirely made-up for demonstration purposes) reacts far more than defense stocks and stocks in the oil industry. It gives me a better chance to hedge my bets than if I chose a defense stock.

The last tool is probably the best, but I'll admit that I don't find it very often. When I do, it's money in the bank! It's a predictive indicator of stock that you might own. Like I said certain stocks move before the rest of the industry, but there are plenty of people that keep an eye on those canaries and you won't be able to react quickly enough. What you want to look for are indicators that predict the industry early enough to predict the canaries. I'll give you an example. Let's say that I think the market is going to crash, but I don't know when. When I look at previous crashes, I see that the "Wipe your rear with \$20 bills" toilet paper company usually predicts the carnage pretty well. So I buy some puts and move the rest of my investments to low beta stocks (utility companies that seem to lag the market) Once I see that my puts are going through the roof, I move my utility stocks to the stocks that tend to negatively correlate to the market.

It's not that simple. It's never that simple! But it's a lot simpler when you realize the tools that you have. If you can add these tools to the ability to research a stock correctly and come up with a decent valuation, you will win a lot more than you can lose.

I hope this helps.

Report this comment
#5) On February 02, 2010 at 11:06 PM, Tastylunch (29.28) wrote:

I agree that beta values provide a measure of a stocks tendency to respond to moves in the market, I disagree that this translates directly into risk.

I agree completely. I've seen more than a few low beta stocks spontaneoulsy collapse before.

Not really much more to add, your interpretation is correct. Beta is a cheap,lazy and very poor proxy for risk.

Report this comment
#6) On February 03, 2010 at 11:09 AM, SkepticalOx (99.22) wrote:

Why are they still teaching this, along with MPT, EMH, etc. in B-school finance courses? It still makes up a majority of finance education in most universities, even though each of these things have been proven flawed, time and time again.

Report this comment
#7) On February 03, 2010 at 4:07 PM, TMFJake (28.26) wrote:

I would agree with Chris that Beta can be useful as one tool among several to properly hedge your portfolio.  Understanding a company's correlation to the market, to its industry, etc., and the magnitude of it's historical price movement relative to this market correlation, can be useful in extracting whatever value that is unique to the company.  I can beta weight my hedged positions and bank the profit that comes when the market comes to its senses and realizes that it has undervalued my company bet.

Of course, I agree with you Anticitrade, that knowing that a portfolio has a high beta over any short time frame tells you nothing.  Similarly knowing that a stock has a high or low beta historcially tells you nothing about whether the stock is a compelling opportunity right now.

A better measure of risk-adjusted returns is looking at the return-to-volatility ratio, using the Sharpe ratio for instance.  I'll evaluate a portfolio that returned 10% with 5% vol different that 20% returns with 20% vol. Now, if the 10/10 portfolio had a Beta of 0 and the 20/20 portfolio had a Beta of 1.3, then I'd be very interested in these beta characteristics based on my outlook for the market in the coming year.

Of course, my crystal ball doesn't do a very good job of telling me the outlook for the coming year...so I put snow flakes and a snowman in it instead. :)

Report this comment
#8) On February 03, 2010 at 5:31 PM, anticitrade (99.64) wrote:

Thanks everyone for the comments.  I should clerify that I think that Beta values are a good indicator of exactly what they measure, and can be an effective tool when used in this way.

elmosworld,

You can definately see my fundamental bias coming out in my proposed alternative to a beta value for measuring risk.  My personal opinion is that a companies ability to generate free cash flows is the base reality in the stock market, and everything else is just grades of deviation from that reality.  I am not saying I am right, but that is the perspective that I currently hold.

goalie37,

Thanks for the comment.  I don't have enough experience on the bond market to add anything to your comment.

ChrisGarley,

Your comment is insightful enough that it should be its own blog post.  I like and agree with pretty much everything you said.  I particularly like the analogy of "canaries" and I wish I had more time to devote to identifying and tracking indicators that I believe would act in this way.  I actually had a business idea around this concept.  Basically provide shopping malls with low cost sensors to monitor foot traffic.  Tie all these systems from the various malls you service into a centralized system to monitoring shopping sentiment.  Then correlate this data with movement in the stock market.  I think if done right, you could own a pretty important canary.

Tastylunch,

I hope you don't mind if I quote you in the future whenever someone starts using beta values as a weapon.

Beta is a cheap,lazy and very poor proxy for risk.

This statement perfectly describes my feelings as well.

SkepticOx,

I think (most) finance proessors must believe in these theories (particulary the efficient market hypothesis) in order to reconcile their personal decision to not actively manage money.

TMFJake,

I agree with pretty much everythign you said (especially about our crystal balls not working), but I think I may have a harder stance on beta values than you.  I see volatility only being a risk factor when an investor must react according to deadlines.  Obviously this is less of an issue for true buy and hold investors, and most traders try to benefit from short term volatility anyway.

Report this comment
#9) On February 04, 2010 at 9:22 PM, sentinelbrit (63.22) wrote:

I've known quite a few great investors in my time and I don't know any who think about a stock's beta when they're investing. They're thinking about the business fundamentals, earnings potential and valuation.

The whole MPT thing is still clung to by academics because otherwise they'd have nothing to lecture about!

Report this comment
#10) On February 10, 2010 at 11:00 PM, FleaBagger (27.82) wrote:

A beta value is extremely useful when betting on the direction of the market. Instead of buying or shorting the market itself, you can get much more bang for your buck buying or shorting VHB or UHB stocks once you feel confident of the trend. Of course, this is very "risky."

Report this comment
#11) On February 12, 2010 at 2:23 PM, lh100 (23.83) wrote:

Great post as always.

However, everything you said is based on a stock being under-overvalued compared to its FCF.

I think the recent crash was a good example of how a high beta stock would not be considered "risky" if you expected it to be undervalued compared to it's FCF generation... but can you correctly identify that number better in a volatile market than in a "normal" market ?

I am nowhere good as you are when it comes to this, but i think it is harder. So chosing high beta  stocks, even if they dropped a lot more than they "should" compared to it's fundamentals, is "riskier" than chosing low beta stocks, because the margin of error should be bigger. During pretty much the whole down period of 2007-2009, you could find a crapload of undervalued high beta stocks all along, because it was pretty hard to find out how badly credit would hit everybody. Yet you would end up with a pretty bad return, but if you found the bottom and did the same strategy, you would end up a hero(and so riskier strategy can bring bigger wins/losses).

Your risk definition is pretty close to correlation, depends on how you define your "average returns"(i am guessing you are comparing to a benchmark for the same period).

Report this comment