All the Money on One Stock? [Archive] - Two Plus Two Older Archives

adios

06-03-2004, 02:09 PM

All the Money on One Stock?

I've wondered for a long time about the risk-reward parameters required in order to make betting the entire bankroll on one stock a good bet. I have a few reasons why this problem interests me. One reason is that I know that "diversification" is not a panacea for achieving superior returns (beating the market). Another reason is that if the market is really as highly efficient as many believe, trying to beat the market is a losers game i.e. it can't be accomplished in a systematic way. If the market is very highly efficient then a stock price could never be that much undervalued to my way of thinking. At least undervalued stocks could not be found in a systematic manner. Diversification and market efficiency are topics well worth discussing but that's not what this post is about. I'm going to try and determine when a stock is undervalued enough to make betting the whole bankroll on it a "good bet."

I'm going to use the term Portfolio Theory loosely in this post. Portfolio Theory states that one way to measure risk is the utilization of a quantity known as beta. The value of an individual stock's beta is given by:

Correlation between stock price movement and the stock market portfolio (covariance) divided by the variance of the stock market portfolio.

I'm using SPY as being representative of the stock market portfolio. The Capital Asset Pricing Model (CAPM) gives the return percentage for the holding period on a compounded basis for a stock selection necessary to make the selection worthwhile. The CAPM is a function of beta, the risk free rate, and the equity risk premium.

Required Return = Risk Free Rate + Beta * Risk Premium

Portfolio Theory states that an investor is only paid a risk premium for risk that can't be diversified away i.e. only for market risk. The standard procedure for estimating beta is to regress a company's stock price returns against the market return:

Stock return = intercept of the regression + slope of the regression * Market return.

The intercept of the regression is commonly referred to as alpha while the slope of the regression is referred to as beta. The final useful statistic that the regression gives is a quantity called R squared which measures the goodness of the fit of the regression. Portfolio Theory states that R squared provides an estimate of the proportion of the risk in investing in a company's stock that is attributable to market risk versus the proportion that is attributable to firm specific risk. The firm specific risk is (1 - R squared). Therefore if we want to put the whole bankroll on only one firm we are assuming all the firm specific risk but only being compensated with a risk premium for the percentage of risk attributable to R squared, the market component of risk. The risk undertaken in assuming firm specific risk amounts to making a 0 EV bet. According to Portfolio Theory a more efficient portfolio would be constructed by using the market portfolio to assume the same amount of risk but be compensated with a risk premium for all the risk undertaken, not just a portion of the risk taken. For instance if an individual stock has a beta of 1 and an R squared of 50%, one should prefer the market portfolio since all risk is compensated.

If an investor is not being compensated for the risk assumed, the stock price should be undervalued by the appropriate amount to make up for the additional, uncompensated risk being undertaken. The question is what is the required rate of return for investing in only one firm?

Again the CAPM indicates the required return for investing, CAPM is a function of beta and assuming the risk indicated by the beta, then the market portfolio can be utilized to yield a return of

Return_Indicated_by_Stock_Beta_Acheivable_Using_SP Y

(remember SPY is the proxy for the market portfolio) where all the firm specific risk is diversified away.

If Return_On_Stock_Due_to_Being_Undervalued is the actual return that the investor will receive due to the stock being undervalued then the return expected is:

Return_On_Stock_Due_to_Being_Undervalued * R Squared

Therefore

Return_Indicated_by_Stock_Beta_Acheivable_Using_SP Y
= Return_On_Stock_Due_to_Being_Undervalued * R Squared

Solving for Return_On_Stock_Due_to_Being_Undervalued:

Return_On_Stock_Due_to_Being_Undervalued
= Return_Indicated_by_Stock_Beta_Acheivable_Using_SP Y / R Squared

As an example:

If a stock's beta indicates a return of 10% is required for the holding period desired and the stocks R Squared is 20% then the stock has to be undervalued enough to have a 50% return for the holding period:

.10 / .20 = .5

Similarly if a stock's beta indicates a required return of 10% and has an R squared value of 80% then the required return for the holding period is 12.5 %:

.10 / .80 = .125.

Conclusions

Betting the whole bankroll on one stock is no doubt a bad idea. We already knew that but showing the kind of returns that are required seems to, at the very least, add some perspective to how bad this is. I wasn't sure myself and I'm guilty of overweighing my portfolio in one stock and I have a better perspective of the risk involved. One can achieve the necessary returns by using the market portfolio, in this case SPY. Usually to assume a lot of risk one would have to use leverage in utilizing SPY. In other words if you think a small cap stock that you feel has a ridiculously low price, betting the whole bankroll on it is equivalent to using a great deal of leverage in buying SPY.

When overweighing one's portfolio, one is probably better off buying a large cap, well known company, with lots of revenues, decent margins that has been beaten up unjustly (determining when this happens is another matter altogether). For instance 5 or 6 years ago Proctor and Gamble got hammered on some bad news really bad. P&G wasn't going to go into chapter 11 and their business was going to be good for the foreseeable future. It might have been well worth overweighing ones portfolio to some extent with this stock. I would guess that P&G has at least a fairly high R squared value. Just a thought.

Also I think it points to the issues involved with diversification. For instance buying a portfolio of all MREITS isn't anywhere close being diversified. Many of the firm specific issues with individual MREITS apply to all of them. MREITS typically have low betas implying low risk but they also have low R squared values so buying a portfolio of these types does not diversify away firm specific risk. It's hard to beat SPY for diversification.

Also in evaluating risk an investor needs to look not only at beta but also at R squared. I plead guilty to not doing this enough. The goal is to be compensated appropriately for all risk undertaken. In my mind this is my most money managers do not beat the averages over the long run, they don't get compensated for all the risk that they undertake in their portfolios.

AceHigh

06-03-2004, 10:04 PM

Actually it works if you know the company really, really well. It's probably you best chance to go from the outhose to the penthouse. See Bill Gates, Mark Cuban, etc.

Unfortunately you usually have to be a pretty high ranking officer in the corporation for you to have the kind of knowledge you need.

winchem21

06-03-2004, 11:01 PM

I normally keep portfolio in no more than 3-5 stocks with the goal of high return and trying not to mimick market.

Right now happen to be nearly all in with one biotech play and am willing to wait for it to hit. Pipeline has several diversified products that could each hit big, the earliest one has minimal technical risk and represents 3-4 bagger potential value. Big bets can be made IF the edge is large.

The IF is the key word....that is what due diligence is all about

wc21

Ray Zee

06-04-2004, 05:52 PM

an easy way for a person to make kind of an offhand guess as to what risk they would take with all of their investment money, is to ask at what price would they need to flip a coin for it all.

GeorgeF

06-05-2004, 09:44 PM

If you use CAPM to determine the optimum 100% concentrated portfolio you will end up with the risk free rate i.e. the optimum portfolio is treasury bills. This sounds like an ECO 101 question.

You can use other methods to create optimum portfolios. Needless to say in addition to Mr Zee's excellent advice there is always Harry Callahan: "I know what you're thinking. I can make a mint off a 100% concentrated portfolio. Well, to tell you the truth, I don't know myself. But being that you could also loose 100% of your money, you've got to ask yourself one question: Do I feel lucky? Well, do ya gambler?"

Rooster71

06-14-2004, 09:21 PM

I can't imagine someone even considering placing 100% of their portfolio in one stock. With so many corporations being riddled by corruption, this is an insane idea. The fact that a company is public (owned by its stockholders and having its share traded publicly) means nothing to the crooked SOBs at the tops of some firms who make $1 mil - $100+ mil/year and their company has only been marginally profitable, if even profitable at all.

adios

06-15-2004, 01:27 AM

[ QUOTE ]
If you use CAPM to determine the optimum 100% concentrated portfolio you will end up with the risk free rate i.e. the optimum portfolio is treasury bills.

[/ QUOTE ]

Don't follow your logic here. It seems obvious that most would prefer a well diversified portfolio that returned at a rate of treasury bills + a risk premium.

My main point was to quantify the risk-reward tradeoffs for assuming uncompensated risk i.e. assuming risk where a risk premium does not exist.

I didn't say anything about an investors utility function which has a great deal to do with how much risk an investor should assume.

adios

06-15-2004, 01:28 AM

.....

adios

06-15-2004, 01:31 AM

If you believe that the stock(s) are undervalued enough then the risk is worth taking. My main point was to quantify the necessary risk-reward relationship when doing so. It should be noted though that you can theoretically acheive tha same returns by using leverage and trading SPY.

adios

06-15-2004, 01:36 AM

Sometimes yes, sometimes no. Doing such a thing with GE, I'd say no.

adios

06-15-2004, 01:38 AM

[ QUOTE ]
I can't imagine someone even considering placing 100% of their portfolio in one stock.

[/ QUOTE ]

You'd be surprised at how much this happens and by people I'd consider knowledgable. I'm not saying your points aren't valid in fact I think they're extremely valid.

laserboy

06-15-2004, 08:02 PM

How exactly do you quantify risk? Is it possible to quantify it in an empirical fashion? My contention is that it is not. My personal view is that risk can be minimized by conducting proper research and fully understanding the companies that you are investing in. I personally would much rather take ownership in a great business than place my money in an amorphous blob of companies, the vast majority of which I know nothing about.

From my perspective, investing in companies is an entirely different discipline than investing in index funds. To draw comparisons between the two is like comparing apples to oranges. Investing in a good small cap has absolutely nothing to do with leveraging up in an index fund and vice versa. I take ownership in good companies because they make tons of money and because I would like to share in their profits. Taking ownership in index funds is more of a bet on the prosperity of the American economy as a whole (a bet that I would not be willing to take right now, BTW).

Am I in favor of putting everything into one stock? Of course not. That is the equivalent of playing with your whole bankroll in front of you... One bad beat and you are done. Like Warren Buffett, though, in general I do believe in putting your eggs in one basket and watching that basket closely. It is, after all, much easier to come up with a couple good investing ideas a year than twenty good ideas. In fact, investing in fewer companies is one of the few significant advantages that the individual investor has over the institutional investor. Most money managers have portfolios spread out over dozens and dozens, or even hundreds of stocks. The more stocks they are invested in, the closer their returns approach the market return. It is no wonder that, after fees, they are underperforming the market.

By concentrating your bets on only the very best of your investing ideas, you can acheived much better results. Think of it as the equivlalent of being tight and aggressive. The key of course is identifying and distinguishing the GREAT investing ideas from the merely good ones and of course the bad ones.

That's my 2 cents...

adios

06-16-2004, 03:07 AM

[ QUOTE ]
How exactly do you quantify risk?

[/ QUOTE ]

beta is one way, a very popular and accepted way.

[ QUOTE ]
My personal view is that risk can be minimized by conducting proper research and fully understanding the companies that you are investing in.

[/ QUOTE ]

Portfolio theory would argue against this. IMO what you're saying is that you can identify companies that the market has not priced correctly i.e. you can identify companies that are undervalued. That doesn't say anthing about the risk you're undertaking. The idea is to maximize your expected return for the risk that you are undertaking.

[ QUOTE ]
I personally would much rather take ownership in a great business than place my money in an amorphous blob of companies, the vast majority of which I know nothing about.

[/ QUOTE ]

Ok but you aren't compensated by a risk premium for undertaking individual company risk that can be diversified away. It's easily provable mathematically that this can be accomplished. In actuality Ray Zee's comments in this thread were very accurate. My post was about quantifying the risk-reward parameters when someone does do this. If you could have higher returns by undertaking the same amount of risk from an alternative investment you'd be crazy not to select the alternative investment.

[ QUOTE ]
From my perspective, investing in companies is an entirely different discipline than investing in index funds. To draw comparisons between the two is like comparing apples to oranges. Investing in a good small cap has absolutely nothing to do with leveraging up in an index fund and vice versa.

[/ QUOTE ]

Sure it does. Stocks in the aggregate have had a historical risk premium of about 5.5% above bonds. Let's say that you have a holding period of ten years. Therefore the risk free rate is the rate paid by the appropiate government bond for the duration of your holding period which in this case is a 10 year Treasury. The 10 year is currently yielding approximately 4.8%. Therefore your expected returns for SPY is 10.3% for the holding period. If you leverage 2-1 it's 20.6%. Alternatively if you put it all on an individual company or a select few companies where your expected return for the same 10 year holding period is 20.6% then it is the same. The question is which choice involved less risk. My post was about how to answer that question according to portfolio theory.

[ QUOTE ]
Am I in favor of putting everything into one stock? Of course not. That is the equivalent of playing with your whole bankroll in front of you... One bad beat and you are done. Like Warren Buffett, though, in general I do believe in putting your eggs in one basket and watching that basket closely. It is, after all, much easier to come up with a couple good investing ideas a year than twenty good ideas.

[/ QUOTE ]

And your investing ideas lead you to believe that you will acheive at least a certain amount of return, the question is what risk are you taking to acheive those returns and when you know that you can evaluate if there are more favorable alternatives that will acheive higher returns for the same amount of risk taken or the same returns with less risk taken.

[ QUOTE ]
Like Warren Buffett, though, in general I do believe in putting your eggs in one basket and watching that basket closely

[/ QUOTE ]

Warren Buffet seeks to buy undervalued securities meaning that he believes that his total return for his investment will be great than alternatives for the equivalent amount of risk taken.

[ QUOTE ]
Most money managers have portfolios spread out over dozens and dozens, or even hundreds of stocks. The more stocks they are invested in, the closer their returns approach the market return. It is no wonder that, after fees, they are underperforming the market.

[/ QUOTE ]

The main reason they underperform the averages is due to the fact that their portfolios are not properly diversified which basically means that they are assuming more risk than they need to to acheive the returns they're receiving.

[ QUOTE ]
By concentrating your bets on only the very best of your investing ideas, you can acheived much better results.

[/ QUOTE ]

At what risk? It's always question you need to ask yourself IMO. What do I expect to make and what risk am I taking to acheive those returns. Furthermore is there better alternative where I could acheive the same return and take less risk or I can take the same amount of risk and acheive an even higher return in an alternative investment.

[ QUOTE ]
Think of it as the equivlalent of being tight and aggressive. The key of course is identifying and distinguishing the GREAT investing ideas from the merely good ones and of course the bad ones.

[/ QUOTE ]

But if you play poker really well and your bankroll is too small for the limits you're playing at you'll go broke nonetheless more than likely. The idea of concept of risk in buying stocks is similar to the fluctuations in bankrolls due to variance.

laserboy

06-16-2004, 05:39 PM

[ QUOTE ]
beta is one way, a very popular and accepted way.

[/ QUOTE ]

Beta measures the correlation of a stock price with the market. That's it. I am not sure how you can say investment risk can be encapsulated in such a limited statistic. To be quite honest with you, I am not sure how this is a useful statistic at all. I typically look at business metrics like earnings and free cash flow. When I speak of risk, I am speaking more about the liklihood that a company will increase its earnings in the future and subsequently pass those profits along to shareholders. There is no statistic that can measure this. It is a highly subjective endeavor.

[ QUOTE ]
Portfolio theory would argue against this. IMO what you're saying is that you can identify companies that the market has not priced correctly i.e. you can identify companies that are undervalued. That doesn't say anthing about the risk you're undertaking. The idea is to maximize your expected return for the risk that you are undertaking.

[/ QUOTE ]

This seems illogical to me. What if you have insider information that the company is going to be bought out at a 20% premium next week? Is this stock a more or less risky investment than an index fund? The more information you have on a company, the better you can assess the proper valuation of a company and the amount of "risk" involved. Again we may be talking about risk in different terms.

[ QUOTE ]
Ok but you aren't compensated by a risk premium for undertaking individual company risk that can be diversified away. It's easily provable mathematically that this can be accomplished. In actuality Ray Zee's comments in this thread were very accurate. My post was about quantifying the risk-reward parameters when someone does do this. If you could have higher returns by undertaking the same amount of risk from an alternative investment you'd be crazy not to select the alternative investment.

[/ QUOTE ]

Well I think we can all agree that all rewards being equal, the less risky investment is the superior one. My contention is that the concepts of risk and reward are entirely subjective measures. If you are using beta as your parameter of risk, then it should come as no revelation that SPY is going to be the less "risky" investment. Wow, big surprise.

[ QUOTE ]
Sure it does. Stocks in the aggregate have had a historical risk premium of about 5.5% above bonds. Let's say that you have a holding period of ten years. Therefore the risk free rate is the rate paid by the appropiate government bond for the duration of your holding period which in this case is a 10 year Treasury. The 10 year is currently yielding approximately 4.8%. Therefore your expected returns for SPY is 10.3% for the holding period. If you leverage 2-1 it's 20.6%. Alternatively if you put it all on an individual company or a select few companies where your expected return for the same 10 year holding period is 20.6% then it is the same. The question is which choice involved less risk. My post was about how to answer that question according to portfolio theory.

[/ QUOTE ]

Again our views of risk and reward are skewed quite differently. For instance, I think our economy as whole is being artifically propped up by a massive debt bubble and unsustainable corporate earnings "growth". Investing in the S&P500 right now would be the height of folly to me, the very definition of a risky investment. Yet there are many small caps out there that I feel very confident in placing my money for the next ten years. This is all based on my research. Of course this is all extremely subjective. Some people are better at assessing risk and reward than others. That's why Warren Buffett is Warren Buffett, and I on the other hand still work for a living. /images/graemlins/smile.gif

[ QUOTE ]
The main reason they underperform the averages is due to the fact that their portfolios are not properly diversified which basically means that they are assuming more risk than they need to to acheive the returns they're receiving.

[/ QUOTE ]

Wall Street professionals underperform the market for MANY reasons, but I won't get into that for now.

The problem I have with "diversification" is that when given the opportunity, most people pick stocks like they are filling out an all-star ballot and they put about as much thought into them. They pick and choose stocks the "like" without taking the time to quantify why they like them. If they took the time to do so, they would find that there are VERY FEW truly great investment opportunities out there. Warren Buffett recently said there is not a single thing on the market he finds interesting. For the most part, I tend to agree with him. If Warren Buffett can't find any interesting stocks out there, how is the average investor expected to find 20 good stocks?

Most investors only get a handful of truly exceptional investing ideas in their lifetime, what Peter Lynch would describe as "back up the truck" situations. Just like you only pick up AA once every 220 hands. From my view it is important to maximize these opportunities.

I think by now you get the point that I am not too fond of Portfolio Theory or for that matter Efficient Market Theory. Not that I don't see the merits of them. The average dolt on the street (this includes most investment "professionals"), I would agree, has no business picking their own stocks, like a child has no business handling sharp silverware. You can be very successful using an index fund approach to investing and I know many very smart people who subscribe to these beliefs.

But I think it is important to realize that there are individuals who are better at quantifiying risk and reward than the "market" is. You can say that Warren Buffett is taking an inordinate amount of "risk" with each investment that he makes, but after a while it becomes time to concede that his assessment of "risk" is perhaps a little better than yours.

Ray Zee

06-16-2004, 11:28 PM

if you put your money into many very risky or high beta stocks. you will approach the risk level of a no risk stock yet get compensated with the risk premium. that in itself leads one to tend to have multiple baskets to watch.

then again betting it all one one stock is not so bad if you have a stock that say hasnt priced in an event, like a takeover. you may think in the short term that may happen because of what you know. the downside is small yet the upside is huge. in a case like that betting it all can be justified.

another case-- your house is about to be reposessed and you will lost 100,000. might pay to take a shot with your whole portfolio to protect against that loss.

again-- you have a chance to become a partner in a good deal but are short some amount to enter. so it may pay to risk alot for the chance to get into the partnership.

adios

06-17-2004, 07:52 AM

[ QUOTE ]
Beta measures the correlation of a stock price with the market.

[/ QUOTE ]

Nope not exactly. The correlation divided by the variance of the individual stock.

[ QUOTE ]
That's it. I am not sure how you can say investment risk can be encapsulated in such a limited statistic.

[/ QUOTE ]

I'd recommend Damodaran's book Investment Valuation and Cornell's book The Equity Risk Premium.

[ QUOTE ]
To be quite honest with you, I am not sure how this is a useful statistic at all.

[/ QUOTE ]

I'd recommend the books. The idea is that the risk you undertake in buying a stock that you are compensated with a risk premium by is a function of the overall market risk.

[ QUOTE ]
I typically look at business metrics like earnings and free cash flow. When I speak of risk, I am speaking more about the liklihood that a company will increase its earnings in the future and subsequently pass those profits along to shareholders.

[/ QUOTE ]

This is assessing the what the valuation of a company should be. A company's value is more or less the present value of future earnings. It isn't risk, it's your estimate of what the valuation should be.

[ QUOTE ]
There is no statistic that can measure this. It is a highly subjective endeavor.

[/ QUOTE ]

The stock price is the market's estimate.

[ QUOTE ]
Again we may be talking about risk in different terms.

[/ QUOTE ]

We are.

[ QUOTE ]
Well I think we can all agree that all rewards being equal, the less risky investment is the superior one.

[/ QUOTE ]

And how would you evaluate this?

[ QUOTE ]
My contention is that the concepts of risk and reward are entirely subjective measures.

[/ QUOTE ]

It's not though, an equity risk premium is paid for undertaking market risk. You seem to not address this fact. What the forward looking equity risk premium is, is another matter but historically stocks have out performed bonds and the difference is the equity risk premium.

[ QUOTE ]
If you are using beta as your parameter of risk, then it should come as no revelation that SPY is going to be the less "risky" investment. Wow, big surprise.

[/ QUOTE ]

Plenty of stocks have a lower beta than SPY.

[ QUOTE ]
Again our views of risk and reward are skewed quite differently. For instance, I think our economy as whole is being artifically propped up by a massive debt bubble and unsustainable corporate earnings "growth". Investing in the S&P500 right now would be the height of folly to me, the very definition of a risky investment. Yet there are many small caps out there that I feel very confident in placing my money for the next ten years. This is all based on my research. Of course this is all extremely subjective.

[/ QUOTE ]

It's your estimate of their valuation. My post addresses this precisely in how to evaluate whether or not the alternative are better. You should be thanking me /images/graemlins/smile.gif.

[ QUOTE ]
But I think it is important to realize that there are individuals who are better at quantifiying risk and reward than the "market" is.

[/ QUOTE ]

Your statement should read:

But I think it is important to realize that there are individuals who are better at estimating valuations than the "market" does.

[ QUOTE ]
You can say that Warren Buffett is taking an inordinate amount of "risk" with each investment that he makes, but after a while it becomes time to concede that his assessment of "risk" is perhaps a little better than yours.

[/ QUOTE ]

Should read:

You can say that Warren Buffett is taking an inordinate amount of "risk" with each investment that he makes, but after a while it becomes time to concede that his assessment of VALUATION is perhaps a little better than yours.

btw it's not my concept of risk, it's the widely accepted viewpoint of what stock market risk is.

adios

06-17-2004, 12:12 PM

.................

laserboy

06-17-2004, 04:53 PM

[ QUOTE ]
Nope not exactly. The correlation divided by the variance of the individual stock.

[/ QUOTE ]

Explain to me how this information has any bearing on the future performance of a stock (please do not say "this is an accepted statistic"). As in roulette, I believe that past results have no bearing on future outcomes. When evaluating individual stocks, I do not look at charts or stock price movements. I certainly do not look at the performance of "the market".

[ QUOTE ]
I'd recommend the books. The idea is that the risk you undertake in buying a stock that you are compensated with a risk premium by is a function of the overall market risk.

[/ QUOTE ]

I understand the concept of a risk premium. I do not understand how beta or "the market" come into play when evaluating the merits of an individual stock. I will take a look at the books though. Thanks. I recommend anything Charlie Munger has ever said or written about Portfolio Theory or CAPM.

[ QUOTE ]
This is assessing the what the valuation of a company should be. A company's value is more or less the present value of future earnings. It isn't risk, it's your estimate of what the valuation should be.

[/ QUOTE ]

Discounted cashflow analysis does allow for risk premiums. The greater the risk, the greater the "discount" to future cashflows. Typically these types of risks are drawn from business factors like unforseen competition or swings in commodities pricing. These are the types of risks I evaluate when purchasing stocks, not past stock price volatility.

[ QUOTE ]
Your statement should read:

But I think it is important to realize that there are individuals who are better at estimating valuations than the "market" does.

[/ QUOTE ]

What is the point you are trying to make here? Every investment decision takes into account an assessment of risk and reward. When Warren Buffett purchases a stock, he has assigned it a valuation that is based on among other things an appropriate risk premium. He then decides whether the risks are worth the rewards, allowing for a healthy "margin of safety". If you are saying he is a successful investor because of his ability to value stocks with absolutely no regard for things like "beta", then I agree with you 100%.

Ultimately I think I approach investing from a completely different viewpoint than you. Purchasing a stock is purchasing ownership in a business. Nothing more. I do not view it as a floating variable that moves up and down due to the whims of the market (though in many respects that's what it is). There is nothing I despise more than Wall Street and modern investing culture. As Benjamin Graham said "Investing is most intelligent when it is most business like."

Waterproof

06-17-2004, 08:37 PM

Interesting thread, but it reminds me sooo much of my finance classes... Anyway, I know that this isn't exactly the point of the post, but if you truely want to gamble in the market for whatever reason (stock looks undervalued, future buy out in 2 weeks, etc.) Why not buy options or LEAPS?

You don't have to risk your whole portfolio on one great idea. If you are right (which you would have to be VERY confident in to make this size of a wager), you get paid. If you are wrong, there is a limited downside.

These investments are not long-term, which gives you an added advantage. You like the stock today, but it's a dog in two years- exercise and/or sell. You like the stock today, you love it in two years... you are still able to purchase the underlying security for around the same cost because you have successfully hedge against its appreciation. Either that, or you can take your profits and purchase more options.

I know the cost of options/LEAPS eats into profits, but the added security that they would provide in a situation like this seems well worth it.

Any particular stock you were thinking of? /images/graemlins/smirk.gif

Ray Zee

06-17-2004, 11:37 PM

investing is all about costs. if they are high you lose for sure. thats why the buy andhold strategy is always been best, as you dont turn over your portfoio. mutual funds dont beat the market for this very reason, and costs, and churning portfolios to be in high fashion stocks and out of the ones that did badly last year.
options, leaps and other derivative type things are for the people selling them and the brokers.

Carl_William

06-18-2004, 01:54 AM

About 7 or 8 years ago, Michael B. Higgins wrote a book:

Beating the Dow: A High-Return, Low-Risk Method for Investing in the Dow Jones Industrial Stocks with as Little as $5,000
Michael B. O'Higgins, John Downes

Basically this method consisted of buying equal amounts of the five Dow 30 Industrial stocks with the highest yields. Then examining the yields every so often and rebalancing if the yields change. O’Higgins noticed that the Dow 30 stock with the second highest yield usually did better than the four other stocks – so he suggested to buy a larger amount of this stock (than the other four).

Some of the brokerage firms started a mutual fund based to some degree on O’Higgins method. They did this by loading the fund with the ten Dow 30 Industrial stocks that had the highest yields (5% for each of these stocks), and then loading the other 50% of this mutual fund with an S&P 500 index fund. These funds had a nickname: “Dogs of the Dow.”

The bottom line is that: O’Higgins created a method which at that time required investing in only five stocks. Assuming this method worked, one problem with it requires rebalancing the stocks every so often as the yields change. This could be very expensive if a conventional broker is involved – thus it is imperative to use a super discount brokerage firm when using this method.

Carl_William

06-18-2004, 02:13 AM

"All the Money on One Stock?" What does ths mean? I take it that it means with what is says. Also I assume that we are talking about honesty -- that is no insider information. Well....

Usually, no prudent gambler or investor would put all of his money on one stock, one horse or one bet. Of course this happens with some people -- and most of these people lose it all (or most of it in case of a stock). In the past, I have known many poker players who continually tried to double their bankrolls when on win streaks -- they all end up on the rail.

One exception is Mr. Rooney who put a big bet on a horse ; his horse won the race; and he bought the Pittsburg Steeler Pro Football team.

FatOtt

06-21-2004, 10:30 PM

[ QUOTE ]
Again the CAPM indicates the required return for investing, CAPM is a function of beta and assuming the risk indicated by the beta, then the market portfolio can be utilized to yield a return of

Return_Indicated_by_Stock_Beta_Acheivable_Using_SP Y

(remember SPY is the proxy for the market portfolio) where all the firm specific risk is diversified away.

If Return_On_Stock_Due_to_Being_Undervalued is the actual return that the investor will receive due to the stock being undervalued then the return expected is:

Return_On_Stock_Due_to_Being_Undervalued * R Squared

Therefore

Return_Indicated_by_Stock_Beta_Acheivable_Using_SP Y
= Return_On_Stock_Due_to_Being_Undervalued * R Squared

Solving for Return_On_Stock_Due_to_Being_Undervalued:

Return_On_Stock_Due_to_Being_Undervalued
= Return_Indicated_by_Stock_Beta_Acheivable_Using_SP Y / R Squared

As an example:

If a stock's beta indicates a return of 10% is required for the holding period desired and the stocks R Squared is 20% then the stock has to be undervalued enough to have a 50% return for the holding period:

.10 / .20 = .5

Similarly if a stock's beta indicates a required return of 10% and has an R squared value of 80% then the required return for the holding period is 12.5 %:

.10 / .80 = .125.

[/ QUOTE ]

None of this makes any sense, with one example being this:
[ QUOTE ]
If a stock's beta indicates a return of 10%

[/ QUOTE ] You can't look at a stock's beta in isolation and say anything at all about required return. Beta just tells you the degree to which the individual stock is sensitive to the market factor.

You can't specify what would or would not be an acceptable risk/return tradeoff for a given stock without specifying some investor's (or the market's aggregate) utility function, including the degree of risk aversion. You can't calculate what the expected return of a particular stock is without first specifying the expected market return and risk-free rate, in addition to the firm's (expected) beta.

As a simple example, if you have someone who is incredibly risk-averse, increasing a stock's expected return by 50% would not be acceptable if it increased the risk (however measured) by 1%. If you have someone who is not at all risk-averse, they'd be willing to buy a stock when it was 1% undervalued even with a huge increase in risk.

FatOtt

06-21-2004, 10:43 PM

A couple more things:

If you believe all the assumptions of CAPM (which you appear to, based on your post), this doesn't make sense either:

[ QUOTE ]
When overweighing one's portfolio, one is probably better off buying a large cap, well known company, with lots of revenues, decent margins that has been beaten up unjustly (determining when this happens is another matter altogether).

[/ QUOTE ]

Of course, under your assumptions, no stock would ever be beat up unjustly, so your strategy would be sub-optimal.

[ QUOTE ]
MREITS typically have low betas implying low risk but they also have low R squared values so buying a portfolio of these types does not diversify away firm specific risk.

[/ QUOTE ]

Buying a portfolio of REITS does not diversify away firm-specific risk, but it doesn't have anything to do with their R2 values from a market regression - it's because the firm-specific risk that they possess is common across all firms. It's possible to gather up a bunch of low-R2 firms and be diversified as long as they don't share a common risk factor.

[ QUOTE ]
It's hard to beat SPY for diversification.

[/ QUOTE ]

It's easy to beat SPY for diversification, especially because the firms in the S&P 500 are much larger than the average firm. That is, even though you own many firms, you own a bunch of firms that share a common factor: size. (If you believe the Fama-French 3 factors.) Try the Vanguard Total Market Fund or ETF for better diversification across US equities, maybe even adding foreign securities and/or bonds for even more diversification.

[ QUOTE ]
The goal is to be compensated appropriately for all risk undertaken. In my mind this is my most money managers do not beat the averages over the long run, they don't get compensated for all the risk that they undertake in their portfolios.

[/ QUOTE ]

One major reason why most managers fail to beat the market return is the same reason why most individual stocks fail to beat the market return: stocks have asymmetric payoffs. That is, with limited liability, you can earn anywhere from -100% to infinity. That's going to (and historically has) result(ed) in the mean individual stock return being much higher than the median stock return. That is, most individual stocks fail to beat the market return. As an example, consider a market of 5 stocks, 4 of which have a 5% return and one of which has a return of 100%. The total market return is 24%, but 4 of the 5 stocks underperformed.

I don't think there's any evidence that portfolio managers aren't being compensated for risk. In fact, if you look at the research, the primary drivers of underperformance in mutual funds are the expense fees (obviously) and the cash holdings that typically underperform the market. If you look at the actual stock picks made by mutual fund managers, they tend to outperform the market.

adios

06-22-2004, 05:58 PM

[ QUOTE ]
Of course, under your assumptions, no stock would ever be beat up unjustly, so your strategy would be sub-optimal.

[/ QUOTE ]

What assumptions are those? I'm using the CAPM to determine the cost of equity.

What I've tried to do in this thread is indicate a way that one would analyze the risk-reward parameters in putting all of the money on one stock. In analyzing the risk-reward parameters I'm using a comparison to an alternative. That alternative is using a market proxy such as SPY to determine the same risk reward parameters. In short I'm making a comparison and I don't think there's an inconsitency.

Market effeciency is a topic well worth discussing but that's not what this thread was about. It's about evaluation of risk-reward parameters. Here's something from Damodaran's book that is relevent regarding market effeciency:

Myth 5: To make money on valuation, you have to assume that markets are inefficient

Implicit often in the act of valuation is the assumption that markets make mistakes and that we can find these mistakes, often using information that tens of thousands of other investors can access. Thus, the argument, that those who believe that markets are inefficient should spend their time and resources on valuation whereas those who believe that markets are efficient should take the market price as the best estimate of value, seems to be reasonable. This statement, though, does not reflect the internal contradictions in both positions. Those who believe that markets are efficient may still feel that valuation has something to contribute, especially when they are called upon to value the effect of a change in the way a firm is run or to understand why market prices change over time.

Furthermore, it is not clear how markets would become efficient in the first place, if investors did not attempt to find under and over valued stocks and trade on these valuations. In other words, a pre-condition for market efficiency seems to be the existence of millions of investors who believe that markets are not.
On the other hand, those who believe that markets make mistakes and buy or sell stocks on that basis ultimately must believe that markets will correct these mistakes, i.e. become efficient, because that is how they make their money. This is a fairly self-serving definition of inefficiency – markets are inefficient until you take a large position in the stock that you believe to be mispriced but they become efficient after you take the position.
We approach the issue of market efficiency as wary skeptics. On the one hand, we believe that markets make mistakes but, on the other, finding these mistakes requires a combination of skill and luck. This view of markets leads us to the following conclusions. First, if something looks too good to be true – a stock looks obviously under valued or over valued – it is probably not true. Second, when the value from an analysis is significantly different from the market price, we start off with the presumption that the market is correct and we have to convince ourselves that this is not the case before we conclude that something is over or under valued. This higher standard may lead us to be more cautious in following through on valuations. Given the historic difficulty of beating the market, this is not an undesirable outcome.

We've discussed many times on this forum that stocks fly under the radar screen so to speak. There are many stocks that don't have significant analyst coverage and personally I believe that they present the best opportunities but I digress.

[ QUOTE ]
Buying a portfolio of REITS does not diversify away firm-specific risk, but it doesn't have anything to do with their R2 values from a market regression - it's because the firm-specific risk that they possess is common across all firms. It's possible to gather up a bunch of low-R2 firms and be diversified as long as they don't share a common risk factor.

[/ QUOTE ]

Did I say otherwise? The fact that R2 values are low means that they're not correlated well to the market. Of course the have firm-specific risk that is common which was my point. To reiterate my point. The fact that R2 values are low means that there is diversifiable risk that is not being compensated by a risk premium. Selecting a portfolio of stocks in MREITs does not diversify away the individual company risk i.e. the firm specific risk still exists. Therefore a portfolio of MREITs is taking on unecessary divirsifiable risk that is not compensated by a risk premium and if indivual MREIT R2 values are low, the uncompensated risk i.e. where a risk premium is not being paid, is substantial most likely.

[ QUOTE ]
It's easy to beat SPY for diversification, especially because the firms in the S&P 500 are much larger than the average firm. That is, even though you own many firms, you own a bunch of firms that share a common factor: size. (If you believe the Fama-French 3 factors.) Try the Vanguard Total Market Fund or ETF for better diversification across US equities, maybe even adding foreign securities and/or bonds for even more diversification.

[/ QUOTE ]

Diversification in terms of assuming U.S. stock market risk i.e. the valuation of U.S. stocks in the aggregate was what I was referring to.

[ QUOTE ]
One major reason why most managers fail to beat the market return is the same reason why most individual stocks fail to beat the market return: stocks have asymmetric payoffs. That is, with limited liability, you can earn anywhere from -100% to infinity. That's going to (and historically has) result(ed) in the mean individual stock return being much higher than the median stock return. That is, most individual stocks fail to beat the market return. As an example, consider a market of 5 stocks, 4 of which have a 5% return and one of which has a return of 100%. The total market return is 24%, but 4 of the 5 stocks underperformed.

[/ QUOTE ]

So? Supports my point totally.

[ QUOTE ]
I don't think there's any evidence that portfolio managers aren't being compensated for risk.

[/ QUOTE ]

I didn't say they weren't being compensated for risk. I said they were taking on unnecessary risk for their expected returns.

[ QUOTE ]
In fact, if you look at the research, the primary drivers of underperformance in mutual funds are the expense fees (obviously) and the cash holdings that typically underperform the market. If you look at the actual stock picks made by mutual fund managers, they tend to outperform the market.

[/ QUOTE ]

If that's the case then they're expected returns are probably less than the market returns but I believe their holding periods are relatively short and they trade fairly actively. I submit that the typical mutual fund market approach is such that all risk taken is not compensated by a risk premium.

adios

06-22-2004, 06:29 PM

.....

FatOtt

06-23-2004, 03:20 PM

[ QUOTE ]

What I've tried to do in this thread is indicate a way that one would analyze the risk-reward parameters in putting all of the money on one stock.

[/ QUOTE ]

But you haven't done that anywhere. You've included no discussion of any risk-aversion measures that would determine whether someone would be inclined to own one stock. That decision is based on the individual's utility curve and risk aversion. The equation that you wrote down doesn't make any sense at all.

[ QUOTE ]
If Return_On_Stock_Due_to_Being_Undervalued is the actual return that the investor will receive due to the stock being undervalued then the return expected is:

Return_On_Stock_Due_to_Being_Undervalued * R Squared

[/ QUOTE ]

No, if the Return_On_Stock_Due_to_Being_Undervalued is X, then the return expected is X + beta*[expected market return - risk free rate] (in a CAPM world).

Your r-squared measure doesn't have anything to do with expected returns. It tells you about the variance of the expected return (how much the actual return is likely to track the expected return of beta*market premium, not the variance of the market return), not the mean of the expected return (the difference between first and second moments.)

[ QUOTE ]
The fact that R2 values are low means that there is diversifiable risk that is not being compensated by a risk premium. Selecting a portfolio of stocks in MREITs does not diversify away the individual company risk i.e. the firm specific risk still exists. Therefore a portfolio of MREITs is taking on unecessary divirsifiable risk that is not compensated by a risk premium and if indivual MREIT R2 values are low, the uncompensated risk i.e. where a risk premium is not being paid, is substantial most likely.

[/ QUOTE ]

The uncompensated risk for a basketful of REITS is likely to be high, but it doesn't have anything to do with whether the firms have a high or low r-squared value. You can construct a basket of low r-squared stocks that has a high residual risk component just as easily as you can construct a basket of low r-squared stocks that has a low residual risk component. It all comes down to how much the stocks in your portfolio covary, not with how much each stock covaries with the market (which is your r-squared measure).

[ QUOTE ]
Quote:
It's easy to beat SPY for diversification, especially because the firms in the S&P 500 are much larger than the average firm. That is, even though you own many firms, you own a bunch of firms that share a common factor: size. (If you believe the Fama-French 3 factors.) Try the Vanguard Total Market Fund or ETF for better diversification across US equities, maybe even adding foreign securities and/or bonds for even more diversification.

Diversification in terms of assuming U.S. stock market risk i.e. the valuation of U.S. stocks in the aggregate was what I was referring to.

[/ QUOTE ]

Diversification in terms of assuming aggregate U.S. stock market risk is much better acheived by using a total market fund than by using SPY. Again, because SPY consists of the largest firms in the U.S. equity market (with some exceptions like Berkshire Hathaway), buying SPY does not diversify away the size risk factor the same way that the total market fund would. SPY is not even close to the most effective way to acheive a diversified fund representing the aggregate U.S. equity market.

[ QUOTE ]
I didn't say they weren't being compensated for risk. I said they were taking on unnecessary risk for their expected returns.

[/ QUOTE ]

These two things sound like exactly the same point. Namely, that managers are taking risks in their stock selections and aren't being compensated sufficiently. There is no evidence of that in the existing financial research that I'm aware of.

[ QUOTE ]
I submit that the typical mutual fund market approach is such that all risk taken is not compensated by a risk premium.

[/ QUOTE ]

There is no evidence of this that I'm familiar with. Can you provide any?

Overall, the r-squared number from a CAPM/Market Model regression doesn't tell you that much - certainly not the things that you claimed it does in your initial post in this thread.

adios

06-23-2004, 04:31 PM

[ QUOTE ]
But you haven't done that anywhere. You've included no discussion of any risk-aversion measures that would determine whether someone would be inclined to own one stock. That decision is based on the individual's utility curve and risk aversion.

[/ QUOTE ]

I don't need to. The assumption is that the investor knows what the value should be and needs to evaluate the risk-reward paramaters so that they can make an investment decision based on their individual utility function.

[ QUOTE ]
The equation that you wrote down doesn't make any sense at all.

[/ QUOTE ]

I disagree but sorry you don't understand.

[ QUOTE ]
No, if the Return_On_Stock_Due_to_Being_Undervalued is X, then the return expected is X + beta*[expected market return - risk free rate] (in a CAPM world).

Your r-squared measure doesn't have anything to do with expected returns. It tells you about the variance of the expected return (how much the actual return is likely to track the expected return of beta*market premium, not the variance of the market return), not the mean of the expected return (the difference between first and second moments.)

[/ QUOTE ]

Expected may be a poor choice of words perhaps required is better. Anyway r-squared is an indicator of the amount of undiversifiable and diversifiable risk. I didn't say r-squared was used to determine the return value. Intuitivly I think it's clear that the more unnecessary, uncompensated risk that is undertaken the higher the required returns should be.

[ QUOTE ]
The uncompensated risk for a basketful of REITS is likely to be high, but it doesn't have anything to do with whether the firms have a high or low r-squared value.

[/ QUOTE ]

Au contraire, from Damodaran's book, Investment Valuation chapter 8:

The third statistic that emerges from the regression is the R squared (R2) of the regression. While the statistical explanation of the R squared is that it provides a measure
of the goodness of fit of the regression, the economic rationale is that it provides an estimate of the proportion of the risk of a firm that can be attributed to market risk; the balance (1 - R2) can then be attributed to firm-specific risk.

[ QUOTE ]
You can construct a basket of low r-squared stocks that has a high residual risk component just as easily as you can construct a basket of low r-squared stocks that has a low residual risk component. It all comes down to how much the stocks in your portfolio covary, not with how much each stock covaries with the market (which is your r-squared measure).

[/ QUOTE ]

Since an investor can diversify away individual company risk, when an investor accepts firm specific risk it matters a lot. It's silly to state that accepting firm specific risk doesn't matter.

[ QUOTE ]
Diversification in terms of assuming aggregate U.S. stock market risk is much better acheived by using a total market fund than by using SPY. Again, because SPY consists of the largest firms in the U.S. equity market (with some exceptions like Berkshire Hathaway), buying SPY does not diversify away the size risk factor the same way that the total market fund would. SPY is not even close to the most effective way to acheive a diversified fund representing the aggregate U.S. equity market.

[/ QUOTE ]

Whatever but betas are computed based on changes in the valuation of the stock market as a whole of which SPY is an excellent proxy for.

[ QUOTE ]
These two things sound like exactly the same point. Namely, that managers are taking risks in their stock selections and aren't being compensated sufficiently. There is no evidence of that in the existing financial research that I'm aware of.

[/ QUOTE ]

If one is not compensated with a risk premium for all the risk undertaken it's silly to dismiss this.

[ QUOTE ]
These two things sound like exactly the same point. Namely, that managers are taking risks in their stock selections and aren't being compensated sufficiently. There is no evidence of that in the existing financial research that I'm aware of.

[/ QUOTE ]

Why wouldn't this be an obvious?

[ QUOTE ]
There is no evidence of this that I'm familiar with. Can you provide any?

[/ QUOTE ]

I'll see what I can do.

[ QUOTE ]
Overall, the r-squared number from a CAPM/Market Model regression doesn't tell you that much - certainly not the things that you claimed it does in your initial post in this thread.

[/ QUOTE ]

It tells you the it provides an estimate of the proportion of the risk of a firm that can be attributed to market risk. See reference to Damodaran's book above.

FatOtt

06-23-2004, 04:51 PM

[ QUOTE ]

Anyway r-squared is an indicator of the amount of undiversifiable and diversifiable risk

[/ QUOTE ]

Maybe stating this simply will help.

R-squared from the CAPM model tells you the proportion of a firm's total risk that covaries with total market risk. 1 minus r-squared tells you the proportion of the firm's total risk that is orthogonal to total market risk. It does not tell you the level of firm-specific risk, only the proportion.

If a firm has a high R-squared, you don't know whether the total firm risk (firm-specific plus market related risk) is high or low.

The R-squared measure will never be part of an expected return calculation because it doesn't have anything to do with expected return.

The only thing R-squared can tell you is how confident in your estimation you can be. That can only be included in an equation when you specify both a utility function and a risk-aversion parameter.

[ QUOTE ]

If Return_On_Stock_Due_to_Being_Undervalued is the actual return that the investor will receive due to the stock being undervalued then the return expected is:

Return_On_Stock_Due_to_Being_Undervalued * R Squared

Therefore

Return_Indicated_by_Stock_Beta_Acheivable_Using_SP Y
= Return_On_Stock_Due_to_Being_Undervalued * R Squared

[/ QUOTE ]

Again, there is no basis at all for the equations you wrote above. Find a reference from Damodaran that attempts to calculate expected or required returns that looks anything like it.

[ QUOTE ]

If one is not compensated with a risk premium for all the risk undertaken it's silly to dismiss this.

[/ QUOTE ]

This is begging the question. Where are you finding any support for managers not being compensated with a risk premium?

adios

06-23-2004, 05:30 PM

[ QUOTE ]
R-squared from the CAPM model tells you the proportion of a firm's total risk that covaries with total market risk. 1 minus r-squared tells you the proportion of the firm's total risk that is orthogonal to total market risk. It does not tell you the level of firm-specific risk, only the proportion.

If a firm has a high R-squared, you don't know whether the total firm risk (firm-specific plus market related risk) is high or low.

[/ QUOTE ]

It tells you the proportion of your investment that is being compensated by a risk premium and the proportion of your investment that isn't being compensated by a risk premium.

[ QUOTE ]
Again, there is no basis at all for the equations you wrote above. Find a reference from Damodaran that attempts to calculate expected or required returns that looks anything like it.

[/ QUOTE ]

Sure there is. If I have $1000 to invest and I can invest in a well diversified portfolio where all specific company risk was diversified away with a beta of 1.0 and a risk free rate of 4.5% with a risk premium of 5.5% then my required return would be 10%. However, if I invested it in a firm with a beta of 1 and an R Squared of .2, I'd only be compensated by a risk premium for the proportion of my investment attributable to R squared, in this case 20%. The other 80% is more or less a crap shoot of 0 EV. Since my required return is 10%, I better be able to believe that the required return for investing in only one stock is equal to the required return of the well diversified portfolio. Therefore since only 20% is compensated by a risk premium I should expect a return of 50%.

I tell you what I'll submit this to Damodaran and see what he says if anything and also point him to this thread.

[ QUOTE ]
This is begging the question. Where are you finding any support for managers not being compensated with a risk premium?

[/ QUOTE ]

If someone underperforms the market isn't it reasonable to assume that they're not being compensated fully by a risk premium?

FatOtt

06-23-2004, 06:05 PM

I think I've found at least one serious flaw in your calculation - I think I understand what's going wrong a little better now.

[ QUOTE ]
If I have $1000 to invest and I can invest in a well diversified portfolio where all specific company risk was diversified away with a beta of 1.0 and a risk free rate of 4.5% with a risk premium of 5.5% then my required return would be 10%. However, if I invested it in a firm with a beta of 1 and an R Squared of .2, I'd only be compensated by a risk premium for the proportion of my investment attributable to R squared, in this case 20%. The other 80% is more or less a crap shoot of 0 EV.

[/ QUOTE ]

You're assigning a 0 EV to 80% of the firm, which is incorrect. There's no firm-specific risk premium attached to that 80%, but you still expect to earn the risk-free rate and the market risk premium. Calling it 0 EV is wrong - maybe you meant it's "0 EV above the expected market return", but I think that changes all of your calculations.

[ QUOTE ]

I tell you what I'll submit this to Damodaran and see what he says if anything and also point him to this thread.

[/ QUOTE ]

As we seem to not be making much progress, that sounds like a good way to come to some kind of agreement. If he agrees that the r-squared value is involved in any kind of required or expected return calculation (without introducing some assumptions about utility functions and risk-aversion levels), I will definitely reconsider what I've said.

[ QUOTE ]
If someone underperforms the market isn't it reasonable to assume that they're not being compensated fully by a risk premium?

[/ QUOTE ]

No, I tried to explain this before. The underperformance is primarily driven by:
1. Cash holdings. Funds hold cash in order to meet anticipated redemptions or if they can't find any desirable investments. These cash holdings obviously underperform the market and have less risk than the market. It's not fair to penalize the managers for the return on their cash while still saying their cash positions are as risky as the market.
2. Firm expenses like management salary, trading costs, administrative expenses, etc.
3. More than half of the mutual funds will underperform the market for the same reason that more than half of the stocks will underperform the market. It is not a symmetric distribution.

Additionally, as I said before, the stocks held by managers tend to outperform the market.

None of this supports your theory of managers picking stocks that don't adequately compensate them for the risk they take, yet it does explain why most mutual funds underperform.

adios

06-23-2004, 06:36 PM

[ QUOTE ]
You're assigning a 0 EV to 80% of the firm, which is incorrect. There's no firm-specific risk premium attached to that 80%, but you still expect to earn the risk-free rate and the market risk premium.

[/ QUOTE ]

No you don't. One is only compensated by a risk premium for the portion of the risk that cannot be diversified away. Firm specific risk can be diversified away thus you're not compensated for that risk by a risk premium. What you wrote is just wrong. Your expectation is 0 for risk that can be diversified away.

[ QUOTE ]
As we seem to not be making much progress, that sounds like a good way to come to some kind of agreement.

[/ QUOTE ]

Ok

[ QUOTE ]
If he agrees that the r-squared value is involved in any kind of required or expected return calculation (without introducing some assumptions about utility functions and risk-aversion levels),

[/ QUOTE ]

To repeat again I think I've made it clear that r-squared isn't part of a required or expected return calculation. See previous posts.

If one wants to evaluate whether or not an investment is appropriate for their utility function doesn't one first have to ascertain the risk-reward parameters of an investment in order to make that evaluation? What I'm addressing is the determination of the risk-reward parameters.

[ QUOTE ]
No, I tried to explain this before.

[/ QUOTE ]

Sure it's reasonable it's a simple math problem.

[ QUOTE ]
The underperformance is primarily driven by:

1. Cash holdings. Funds hold cash in order to meet anticipated redemptions or if they can't find any desirable investments. These cash holdings obviously underperform the market and have less risk than the market. It's not fair to penalize the managers for the return on their cash while still saying their cash positions are as risky as the market.

[/ QUOTE ]

If they're market timing, their goal is to acheive a return greater than or equal to the market return, and their underperforming then they're not allocating their resources in an effecient way which means they're not being compensated by the equity risk premium appropriately. If there stated goal is to achieve a return less than the market return by accepting less risk that's another story.

[ QUOTE ]
2. Firm expenses like management salary, trading costs, administrative expenses, etc.

[/ QUOTE ]

Ok again if their goal is to equal or do better than the market return then they aren't allocating their resources effeciently.

[ QUOTE ]
3. More than half of the mutual funds will underperform the market for the same reason that more than half of the stocks will underperform the market. It is not a symmetric distribution.

[/ QUOTE ]

So what? They all could do as well as the market which means they're portfolios are sub optimal in terms of risk-reward parameters.

[ QUOTE ]
Additionally, as I said before, the stocks held by managers tend to outperform the market.

[/ QUOTE ]

Which means the weighting of stocks in their portfolios is done in such a way that their returns are less than the risk premium being paid by the market.

[ QUOTE ]
None of this supports your theory of managers picking stocks that don't adequately compensate them for the risk they take, yet it does explain why most mutual funds underperform.

[/ QUOTE ]

I just explained how it did.