| User | Post |
|
11:59 pm
January 3, 2012
|
Jae Jun
| | |
| Admin
| posts 1464 |
|
|
jalleninvest said:
Graham used twice the AAA bond 12 month yield in some of his articles. the 10 year AAA yield right now is about 2.22%. I've also seen Graham talk about the AA bond yield, which is about 3.40% at the moment.
Really? I didn't know. Can you reference some links for me? At the moment, it seems safe just to use the long term bond yield which is about 4.5% ish.
One other amusing feature of Graham's approach was insisting on simplicity. Any analysis that relied on more than simple math was suspect in his approach, and towards the end of his career was an advocate of very basic criteria. Of course, that happened to be the early 70's when buying just about anything cheap worked out quite well…. and a lot was cheap then.
That's what I'm finding out. My screens based on Graham's simple approach has been doing well compared to those where more research were put in.
|
|
|
1:16 pm
January 3, 2012
|
jalleninvest
| | Coronado, CA | |
| Member | posts 22 |
|
|
Post edited 5:25 am – January 3, 2012 by jalleninvest
I tend to thnk about intrinsic value as only one of several estimations of "value," not unlike what a real estate appraiser does when arriving at an opinion of value.
A real estate appraiser looks at three methods of valuing a property. The replacement cost approach tries to figure out what it would cost to duplicate the property at current construction etc rates. The income approach arrives at a valuation based on the income a property would generate, using present market rents and costs, then applying a capitalization rate appropriate for the type of property and location. The market value method looks at comparable properties, and figuring out adjustments for size, location, quality, view, amenities etc. All of this effort is expended to arrive at an "opinion of value." In some situations, the income approach is most relevant; in others, comps are most important. In some markets, there is no point in building new if you can buy cheaper.
Graham used twice the AAA bond 12 month yield in some of his articles. the 10 year AAA yield right now is about 2.22%. I've also seen Graham talk about the AA bond yield, which is about 3.40% at the moment.
One other amusing feature of Graham's approach was insisting on simplicity. Any analysis that relied on more than simple math was suspect in his approach, and towards the end of his career was an advocate of very basic criteria. Of course, that happened to be the early 70's when buying just about anything cheap worked out quite well…. and a lot was cheap then.
|
|
|
1:14 am
December 22, 2011
|
Jae Jun
| | |
| Admin
| posts 1464 |
|
|
Google somehow does an excellent job of concealing the good pages.
The equation with growth is to show that it should be removed from the equation.
The author goes on to replace g with what he simply calls speculative value. That way, you work with what you know and have the unknown value on the right side of the equation.
He then goes on to challenge what that g figure could be by reverse engineering the market price and expectations.
|
|
|
6:19 pm
December 21, 2011
|
somrh
| | |
| Member | posts 336 |
|
|
Jae, it looks the equation they're using (derived in CH 4… the preview didn't include the derivation) has growth as an input variable. So you're still stuck with trying to estimate what the market thinks the growth rate will be.
|
|
|
12:02 pm
December 21, 2011
|
Jae Jun
| | |
| Admin
| posts 1464 |
|
|
At the moment, it is just set to 4.4%
Which is in line with the rate over a long period. You could adjust it but it doesn't really change the valuation a whole lot.
I prefer not to play around with difficult to determine variables. Makes it seems like Im focusing too much on getting a number that suits my taste as opposed to challenging the market price based on my valuation.
|
|
|
11:51 am
December 21, 2011
|
Graeme
| | Austin, Texas | |
| Member | posts 183 |
|
|
Do you alter your denominator in you Graham formula to represent the current bond rate (3%) or do you keep it at 6%?
|
|
|
11:11 am
December 21, 2011
|
Jae Jun
| | |
| Admin
| posts 1464 |
|
|
@ Nell,
You have a good point about systemizing holding positions. I need to come up with a spreadsheet or something to maintain my positions.
I tend to stray from position sizes.
@ Graeme & Somrh,
Here is a preview of the book from Google Books.
Go to Chapter 6 which is the chapter on choosing a discount rate.
http://books.google.com/books?…..mp;f=false
|
|
|
10:13 am
December 21, 2011
|
somrh
| | |
| Member | posts 336 |
|
|
The book looks interesting. I'll admit I'm skeptical that you can infer a discount rate though. Even using a simple model that assumes constant growth indefinitely to give you the equation:
p = d / (r – g)
where p is the price, d is the current dividend, r is the discount rate and g is the growth rate you still have 2 unknown variables. The price should reflect both. If you have a sense of what the markets think the growth rate is, you can estimate the discount rate but how do you know what the markets think the growth rate is?
nell asked:
1. How much capital should one invest in a "risky" company?
I'm sympathetic to the idea that risk can't be quantified. I think there are things you can quantify but you have to make all sorts of dubious assumptions to do so. That's not to say that there isn't value in that but it has limitations. So I'll skip your other questions.
As for "risky" companies, I wouldn't "invest" in one. I might speculate. And my preference for speculation is options. They limit my downside risk, they lower how much capital I devote to the position. And they leverage my returns. I figure if I'm going to gamble I should at least make it interesting.
|
|
|
9:18 pm
December 20, 2011
|
Graeme
| | Austin, Texas | |
| Member | posts 183 |
|
|
Post edited 1:18 pm – December 20, 2011 by Graeme
Jae,
Can you give us the teaser trailer on how they reverse engineer what the market is giving as a discount rate?
|
|
|
11:39 am
December 19, 2011
|
nell
| | |
| Member | posts 100 |
|
|
Risk should be viewed not only for one investment but also in a portfolio context:
1. How much capital should one invest in a "risky" company?
2. How many failed bets can my portfolio accept while still preserving my capital? (This relates directly to margin of safety – 50% margin of safety means 50% of my portfolio bets can go to nil and i would still preserve my capital if all other bets reach intrinsic value.)
3. How many stocks should I hold in a portfolio? If i have 5 uncorrelated bets and a 50% chance to loose in every bet, i would have a 3% probability to loose all my capital.
|
|
|
11:36 am
December 19, 2011
|
Jae Jun
| | |
| Admin
| posts 1464 |
|
|
I just read about this exact topic on the weekend in a book I'm reading.
"Accounting for value"
The book goes about discussing how to determing a discount rate by first reverse engineering what the market is expecting the discount rate to be.
Not the easiest book to read and I'm finding it difficult to recall what I had just read in the past few days.
But I can vouch that it is an interesting discussion.
|
|
|
3:17 pm
December 18, 2011
|
Graeme
| | Austin, Texas | |
| Member | posts 183 |
|
|
somrh said:
Here's a fun question though. How much does your valuation vary depending upon what discount rate you choose?
The fluxuation of a margin of safety is huge once you start playing around with discount rates. Having a rate of 9%, 12% or 15% can change your intrinsic value target dramatically. A company could go from "currently selling at intrisic value" to "45% undervalued" based on your rate.
Those are the numbers I use tho: I try to break down all the companies I'm looking at into a safe (9%), medium (12%) and risky (15%) category based on size, business model, profitability, track record etc.
But man…sometimes it's temping to value something with the 9% when you know it should probably get the 15%.
|
|
|
2:45 pm
December 18, 2011
|
ankitgu
| | |
| Member | posts 49 |
|
|
I agree with Jae on this topic quite a bit.
I don't think I worry too much about discount rates… I know that I won't invest if the return will be under something like 5-6%, I'd rather just hold onto cash most likely in a money market fund protected by FDIC insurance. Discount rates and interest rates that drive them will vary over time, and if I'm going to hold something for a long period of time, then it will be subject to varying rates.
I think this is a part of the reason why we need to focus on a huge margin of safety. By purchasing at a 50% discount to an intrinsic value, we can get a lot of protection against macroeconomic factors among other issues.
If you have an investment that is very much sensitive to interest rates, you might want to look at purchasing a hedge of some kind. The length of time of these hedges and price at which you buy them may depend on how much money you move, because some instruments may be limited to institutional investors who can purchase it in very large transactions. Small investors will be able to purchase options against publicly traded indexes though. If you need to make a 20 year long investment and want to hedge against interest rate fluctuation, you might want to use TLH or TLT options. There might be some maintenance required in order to maintain a hedge, which could create additional work though.
In general, we should probably look for investments that have a large enough margin of safety that have investment merit on their own. If we're getting so precise that we need to hedge out interest rate movements, then it's probably too close of a call.
|
|
|
5:57 am
December 17, 2011
|
nell
| | |
| Member | posts 100 |
|
|
@somh:
Typically that's done by choosing a higher discount rate. Or do you have in mind something along the lines of this? (As a side note, I'm still not sure how one would go about implementing it.)
Two methods..
1) Calculate your risk-free intrinsic value and take a big margin of safety for any risks that are inherently difficult to predict. (Simple, easy method -> I prefer this one..)
2) Use a scenario analysis that weighs different scenarios (good, normal, bad times) and adjust cashflows accordingly to the scenario. You should get three different intrinsic values.
IV = Pr (S 1) * IV (S 1) + Pr (S 2) * IV (S 2) + Pr (S 3) * IV (S 3)
Pr: Probability 0..1, where Pr(S1) + Pr(S2) + Pr(S3) = 1
S: Scenario 1..3
IV: Intrinsic Value
Approach 2. can be extended to a complete Monte-Carlo Analysis that modern finance offers as a decision support tool. I dont believe that risk can be quantified and i prefer approach 1.
My two cents..
|
|
|
4:16 am
December 17, 2011
|
somrh
| | |
| Member | posts 336 |
|
|
Graeme wrote:
If you're using that as your desired rate of return, at what point do
you say "ok, bonds are only yielding 2%. Am I crazy to be looking for my
old desired rates of return."
Considering that real yields for shorter term (5-10 years) treasuries are negative you might not be crazy.
nell wrote:
I think it is foolish to modify a discount rate to account for risk ..A contribution for risk should be included in our cashflow estimates (-> dcf context)
Typically that's done by choosing a higher discount rate. Or do you have in mind something along the lines of this? (As a side note, I'm still not sure how one would go about implementing it.)
…
Here's a fun question though. How much does your valuation vary depending upon what discount rate you choose? Since I have some time this week I may finally finish writing that article.
|
|
|
2:42 am
December 17, 2011
|
nell
| | |
| Member | posts 100 |
|
|
There is no "correct" discount rate, but it should be your personal return hurdle where you feel comfortable assuming that specific risk in equity ownership in general (like lending someone, you dont know, money).. My personal return hurdle is 10% for all equity investments.. I think it is foolish to modify a discount rate to account for risk ..A contribution for risk should be included in our cashflow estimates (-> dcf context)
|
|
|
2:43 pm
December 16, 2011
|
Graeme
| | Austin, Texas | |
| Member | posts 183 |
|
|
somrh, I totally agree in that intrinsic value can't (or at least, shouldn't) be seen as a moving target. That just doesn't make any sense. And so Jae I totally agree with your "keep in simple" method of giving a lower risk premium to large caps and a larger one to small caps. But at what point (or is there a point) where the borrowing ability of soverign states or even corporate bonds change our underlying "risk free" rate. If you're using that as your desired rate of return, at what point do you say "ok, bonds are only yielding 2%. Am I crazy to be looking for my old desired rates of return."
|
|
|
11:01 am
December 16, 2011
|
Jae Jun
| | |
| Admin
| posts 1464 |
|
|
good question Graeme
as somrh alluded to, it's more black magic than anything else.
I prefer to just keep it simple.
It will get laughed and scoffed at by finance schools and wall street.
earnings yield or required return = risk free rate + risk premium
Using the 3% as the example.
Large caps = 3% + 5-6% = 8-9%
Small caps (includes illiquid, OTC, PINK) = 3% + 9-12% = 12-15%
That's just a rough guide. Depending on the environment and how cheap stocks are, the risk premium can be lowered by a percentage point or so.
|
|
|
6:36 am
December 16, 2011
|
somrh
| | |
| Member | posts 336 |
|
|
A worthwhile quote from Benjamin Graham, The Future of Common Stocks:
It seems logical to me that the earning/price ratio of stocks generally should bear a relationship to bond-interest rates. If this thesis is accepted in its simplest form we must conclude: If one dollar of Dow earnigns were worth $17 when bon yields were 4.4%, that one dollar is now worth only 52 per cent of $17, or $8.80, with AAA bonds at 8.5 per cent. This in turn would suggest a currently justified multiplier of, say, nine for the normal current earnings of the Dow.
So I think that's why Graham used AAA bonds as an input variable which you'd have to look at current rates for.
But it does suggest that "intrinsic value" is actually a moving target (I've frankly never liked the phrase "intrinsic value"… not in axiology and even mroe so dislike its use in value investing circles but it is the lingo… how can it be intrinsic if it depends upon interest rates?). If I calculate, say, the DCF for a stream of cash flows, I need to choose some discount rate. The price I arrive at will depend upon my choice of a discount rate. What's the correct discount rate?
Or consider the following question. Are bonds in a bubble right now? I know I mentioned I have some reservation in using the term bubble but I think we can say that, for example, stocks might be overpriced by comparing their returns to bond returns. But what do we compare bond yields to? And there's a lot of talk of, say, "credit bubble". I have sympathy for Fama who questions what that even means (in spite of the fact that I don't think much like him at all!)
One way to procede would be to ask the question: what rate of return would I like to receive? – and then use that for a discount rate. But I don't know.
As for arriving at a margin of safety, I'm an advocate of using uncertainty analysis to help in this regard. And as soon as I'm not done distracting myself with other things I'll finish putting together some techniques I developed (by developed, I mean borrowed from other disciplines) to apply to finance.
|
|
|
2:47 pm
December 15, 2011
|
Graeme
| | Austin, Texas | |
| Member | posts 183 |
|
|
Post edited 7:01 am – December 15, 2011 by Graeme
Part of value investing 101 is to look for earnings yields that are significantly higher than AAA bond yields. We actually use that AAA bond metric for a lot of our valuations and margin of safety calls, because, if you're not beating the bond yield why aren't you just buying bonds.
For the most part we use 6% as our metric for a long term bond yield. But with a pretty brutal market out there at what point do we start using lower bond yields in our calculations. My question is: do you guys personally change your margins of safety or what you look for in an earnings yield based on current 20 year bond numbers? I think today the 30-year US Treasury bond was just over 3%. That's a lot different than 6%. Graham advocated to look for earnings yields twice the bond rate. Does that mean we are now satisfied with 6% earnings yield as opposed to 12%?
I'm curious to see what you guys think!
(and ignore the apostrophe in the title)
|
|
Home
