Present Value of Future Cash Flows

[luutzu]

So an assumption about Buffett and a Uni subject - Is that all you bring to the table? I reckon you have at least read RM's book? What else?

Buffett is on TV, and you have to trust people on TV, haven't you seen the Lego Movie?

It's easy to fall in love with complicated formulae, and maybe it was just me but I was quite proud to be able to work out a beta result for CAPM. And with all these cash flows, how good it was to plug in a whole bunch of numbers and get a nice and neat "value"... and if I had stayed with the Masters degree, I would actually come to know what alpha is and how to model all these nonsense too.

Like i said, if you're certain of the accuracy of all the factors in your model, it's the way to go.
Just I doubt anyone could accurately estimate a company's CURRENT earning power, let alone know that and able to predict other influences on the business.

 

[craft]

Buffett is on TV, and you have to trust people on TV, haven't you seen the Lego Movie?

It's easy to fall in love with complicated formulae, and maybe it was just me but I was quite proud to be able to work out a beta result for CAPM. And with all these cash flows, how good it was to plug in a whole bunch of numbers and get a nice and neat "value"... and if I had stayed with the Masters degree, I would actually come to know what alpha is and how to model all these nonsense too.

Like i said, if you're certain of the accuracy of all the factors in your model, it's the way to go.
Just I doubt anyone could accurately estimate a company's CURRENT earning power, let alone know that and able to predict other influences on the business.

Now you have lost me - are you making some assertion now about us using beta and CAPM? and whatis the reference to Buffett meant to mean?

 

[luutzu]

Now you have lost me - are you making some assertion now about us using beta and CAPM? and whatis the reference to Buffett meant to mean?

What i am saying is that Buffett is a very smart man, and as far as I can see, sounds like a very decent man too.
When a person that smart, and that successful, tells you this is how he value a company... maybe there's something to it.

The guy doesn't use a computer, doesn't even have a calculator... and he's able to decide within a few minutes the value or businesses worth billions.

How could he or anyone do that?

By having a computerised brain that can do your calculations, predict growth rates and forecast the future? That he's a "six sigma event" or take over companies and run its management?


These models you guys are using is based on the assumption that you could accurately predict the future... and by using a couple growth assumptions, a few scenarios, you get a few nice rounded figures and congratulate yourselves for knowing how this and that will impact this and that.

That's nice but that's no more simplistic than a "simple" perpetual annuity approach.

But unlike the simple model of valuing a financial asset, your models are more dangerous because, among other things, you might not be aware that your growth forecasts and assumptions are influenced by general market or economic sentiments.. that and you seem to think that the reported earnings, averaged out or whatever, is the earnings and all you need to do to get an edge is to accurately predict the future and its impact on your company.

Me, i preferred to pay for what i see now... and if the future is brilliant, i'll pay an appropriate price for that brilliance when it happen... and if what i see now is pretty good but the world is saying it's going to end, well i don't know if that's the case so let's not bet on it.


And it could be just me again, but I find it pretty hard to know a business well enough to even have a reasonably confident idea of its earning power.

-----------

And yes, beta and CAPM is rubbish.
It's a bunch of complicated maths built on false foundation.

It's insane to say that in order to make more money, you have to take on more risk.

That's like telling a person that to make money, they have to go to the casinos, and bet all on black or something.

It occurs to a lot of ordinary people that gamblers lose money because they took on more risks by playing against the casino, and those who doesn't take risks gambling get to keep their money.. and those that put more thought into their work, take a few a risk as possible, actually make money out of it - at least more often than a risky gambler.

But let say CAPM is right on this risk-reward assumption. They're wrong in how they define risk.

An average person like myself would have thought that a risky business is one that doesn't make money, not ones whose market price fluctuate more than the averages of other businesses in a representative index.

 

[craft]

What i am saying is that Buffett is a very smart man, and as far as I can see, sounds like a very decent man too.

Yes he is - you should listen to him carefully.

Everything written by Buffett in the Berkshire Hathaway Annual reports indicates that he simply uses the Present Value of Future Cash Flows to estimate Intrinsic Value.

Also see this link for some more information.

https://www.aussiestockforums.com/forums/showthread.php?t=20847&p=670773&viewfull=1#post670773

When a person that smart, and that successful, tells you this is how he value a company... maybe there's something to it.

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

Actually I'm wondering if a person who mis-asserts and mis-represents as the basis for telling us how to suck eggs warrants a point by point response (maybe somebody else has the patience) or if it would just be an exercise in futility.

And yes, beta and CAPM is rubbish.
It's a bunch of complicated maths built on false foundation.

It's insane to say that in order to make more money, you have to take on more risk.

That's like telling a person that to make money, they have to go to the casinos, and bet all on black or something.

It occurs to a lot of ordinary people that gamblers lose money because they took on more risks by playing against the casino, and those who doesn't take risks gambling get to keep their money.. and those that put more thought into their work, take a few a risk as possible, actually make money out of it - at least more often than a risky gambler.

But let say CAPM is right on this risk-reward assumption. They're wrong in how they define risk.

An average person like myself would have thought that a risky business is one that doesn't make money, not ones whose market price fluctuate more than the averages of other businesses in a representative index.

Where in the world did we use or defend CAPM and Beta?

 

[DeepState]

1. Before Tax NPV $2.78 (discount rate 15.00%)
After Tax NPV $2.88 (discount rate 10.50%)

As I expected the valuations should either be exactly the same or at least very close.

I am not sure why the After Tax NPV is 3.6% or $0.10 higher. I assume that it has to do with the difference between 28.5% and 30% and the effect on the discount rate....

2. The only "grey area", and one that I am stilling working through myself is the terminal value.

If you are using a perpetuity (or a number based on a proportion of a perpetuity) that includes a discount rate which is equal to the cost of capital, wouldn't this effect the IRR? Should the discount rate on the perpetuity be the hurdle rate? There's a big difference for example 1 / 0.11 and 1 / 0.15 even if it is a far dated cash flow at the end of the calculations.

I guess what I am saying is that if your discount and hurdle rate are the same thing, then perhaps this isn't an issue?

From Craft:

I always use before tax and interest (i.e. eliminate the financial structure) to compare. I want to establish the best business not who's got the best tweaked financial structure - that's a minor(unless its tweaked to aggressively) and later consideration.

The question to me is whether because of the PPL levy I should push up my pretax hurdle rate as less will now stick to my ribs and more to the governments


1. Although post-tax FCF is 0.7 x pre-Tax FCF, it is not the same thing as discounting post-tax FCF by an interest rate which is 0.7x the pre-tax interest rate. Compounding effects do surprising things. To see the fuller effect, do not assume the company stabilizes after five years and watch the valuations diverge materially. The discount rates required to make these NPVs have the anticipated relationship changes depending on your reinvestment assumption.

One small wrinkle, I think Craft produced a valuation using pre-tax FCF and gross dividends using the pre-tax discount rate as is appropriate. He found the ratio of valuations differed by a ratio of 1.5% attributable entirely to the value of the PPL. The gross dividends in the growth phase, however, are not distributing full franking credits that accrued. Allowing for this, the after tax valuation would then be a little further lower than the pre-tax valuation to allow for franking not distributed (and never will be distributed under these assumptions).

2. In the ultimate extension of valuation theory, you can vary your interest rate assumptions as things get more distant. This is sometimes justified by observing that the bond yield curve, for example, has a non-linear shape. Given the discount rate applicable to cashflows from bonds varies depending on the time they are expected to be received varies...why not equity valuations? Errr.

If you apply a lower rate of capitalization to residual income than your hurdle rate, what you are saying is that, at terminal date, the market will value it as you have assumed...using a different discount rate than you want. Nothing stops you from doing that, but I'm sure it's obvious to you that it is less conservative than just discounting all cashflow at your higher discount rate. Changing this terminal discount rate will definitely impact your IRR as the terminal value will change with all else constant.

If you are using residual assets as terminal value, this might disfavor low capital businesses depending on how you might otherwise value their ongoing cashflow. Again, in theory, the right thing to do is to figure out return on capital and fade it in to some cost of capital figure over some time frame. Except, in low capital businesses, your assets are largely your people. Hence, you need to...get this...convert your people into capital-like assets and account for them as such in ROIC type calculations. In reality, terminal growth rates of 2-3% are used and you just test it to see if it is in your zone of cheapness.

Craft: This is a problem with using pre-tax DCF. On a net of tax basis, no adjustment would be required for your discount rate as the impact of the PPL is just an item that reduces discretionary FCF. The switch between the tax mix worsens your after tax NPV as you would expect if the discount rate is held constant. On a pre-tax basis, you continue to have the same FCF and, if you are thinking in pre-tax terms, this should just stay the same....but you know it isn't when it comes down to it. So I guess you should add some sort of 'fudge factor' to make allowance for it.

The general practice that I have seen is that these calcs are done on a full equity basis. That is, you need to reorganize the company accounts to produce the after tax FCF as if the company had no leverage. Mostly, that involves adding back the cost of interest payments and adjusting tax paid accordingly. This accounts for differences in capital structure across companies. You then discount that figure with what you would like the company to return if you owned the entire share base which can include considerations of your own capital structure - ie. you can factor in your own leverage etc. Because franking is a factor in Australia, you would also discount the value of franking to whatever degree you think is right for you (usually it is 60-100% to allow for the fact that not everyone values franking).


Cheers

 

[luutzu]

Sorry to say, but you're wrong. And if Buffett does what you are doing, he too is wrong.
And just to name drop, I remember reading/hearing an interview where Charles Munger said he never saw Buffett do a discounted cash flow model like the ones you guys are into.

I agree that yes, a company's value is the discounted present value of future earnings [net op. cash flow]... but that does not mean I, and I bet you not Buffett or anyone who's a bit sensible, would take that to mean going ahead and predict what that future cash flow will precisely be, from now until eternity.

And if you can't possibly predict what that growth or earning to be over 5 or ten years, why in the world would you find it more precise to plug in assumed growths and impact of this and other policies and factors.

I'm no mathematician but i'm pretty sure that the more variables you use, the greater the IMprecision will be. Unless, of course, you get all those variables correct.

And we are talking about variables that depends on other variables, most of which the company has no control over, most of which no politicians or treasurer or Nostrodamus could foresee let alone estimate its ultimate impact on a company's growth.

Maybe you guys can predict the future, I know I can't.

---

I thought in your reply you believe in beta and CAPM.

 

[McLovin]

I agree that yes, a company's value is the discounted present value of future earnings [net op. cash flow]...

Given your vast repository of knowledge, I'm surprised you think earnings = net operating cash flow. Crack open those text books again, my friend.

I really think you should read up on what you're saying because you have made a lot of very simple mistakes. Nothing wrong with that of course, I make them all the time, but you're on here telling everyone else they're wrong.

You never answered my question about how you're system would have fared in 2007, or with mining services stocks a couple of years back. The idea of jusy accepting today's earning at face value and extrapolating them to the end of time, and checking back in a couple of years if things don't work out would have had you shuffling down the footpath in your bathrobe and slippers.

 

[DeepState]

1. When a person that smart, and that successful, tells you this is how he value a company... maybe there's something to it.

The guy doesn't use a computer, doesn't even have a calculator... and he's able to decide within a few minutes the value or businesses worth billions.

How could he or anyone do that?


2. These models you guys are using is based on the assumption that you could accurately predict the future... and by using a couple growth assumptions, a few scenarios, you get a few nice rounded figures and congratulate yourselves for knowing how this and that will impact this and that.

That's nice but that's no more simplistic than a "simple" perpetual annuity approach.

But unlike the simple model of valuing a financial asset, your models are more dangerous because, among other things, you might not be aware that your growth forecasts and assumptions are influenced by general market or economic sentiments..

... that and you seem to think that the reported earnings, averaged out or whatever, is the earnings and all you need to do to get an edge is to accurately predict the future and its impact on your company.


3. Me, i preferred to pay for what i see now... and if the future is brilliant, i'll pay an appropriate price for that brilliance when it happen... and if what i see now is pretty good but the world is saying it's going to end, well i don't know if that's the case so let's not bet on it.


4. And it could be just me again, but I find it pretty hard to know a business well enough to even have a reasonably confident idea of its earning power.


5. An average person like myself would have thought that a risky business is one that doesn't make money, not ones whose market price fluctuate more than the averages of other businesses in a representative index.

Hi Luu


Here are my views on your comments:


1. He is able to make quick decisions because he knows several industries intimately and knows the characteristics of the companies he is looking for. Then...he values them. For him, price is very important. Valuation relative to price is a measure of his safety margin. So valuation is important.

Let's take a look at Page 17 of his latest annual report (Source: BRK-US)



Notice how he:
+ normalizes earnings
+ forecasts productivity and crop prices
+ produced expectations for the future which were realized, but where earnings are definitely different to current, let alone normalized earnings.
+ uses a dcf approach.


2. No-one claims to be able to predict with extreme accuracy. But if you happen to know that the forward curve for iron ore is going through the floor, wouldn't it be vaguely sensible to make even some allowance for it in your valuation of FMG-AU? Anything would be more accurate than doing nothing.

Or, to make the point clearer, let's say FMG-AU was within 2 years of mine life left before it is moth-balled. And you'd like to capitalize a perpetuity?

Or CSL-AU just licensed Gardasil a vaccine for HPV, something that every girl and now, pretty much every boy will need to take. It had not achieved sales but has been approved for sale by the FDA, received CE mark and also TGA approval. We are to make no allowance for this?

DCFs do not need to be complicated. They permit the flexibility to take such things as above into account. A perpetuity requires few assumptions. Sometimes, where earnings are stable and growth is steady, a perpetuity type valuation might be appropriate. However, only in extreme situations would an assumption of the same earnings being generated in perpetuity be any decent sort of representation for the company. DCFs require more assumptions, but you'll find only a few matter and people focus on those.

If you are going to run an annuity or run a DCF, a good place to start is some form of normalized earnings. Hence the assumption load is pretty much identical for both practices. Thereafter the assumptions differ. If your future earnings stream is anything other than some form of simple compounding, then a DCF is superior to PE etc.

It is reasonably argued that any of those assumptions are hard to derive accurately. But to suggest there is no better predictive ability than perpetuating a zero growth rate is unreasonable. That zero growth rate assumption, simple and innocent as it may seem, will be wildly inaccurate for almost all companies.

It will lead you into situations like expensive depleting mines and prevent you from buying cheap rockets at their take-off point. That would not seem like a useful tool to apply to the task of investment.

While a DCF will be wrong, on average, it will be less wrong than a perpetuity calculated as you have done.

DCF is just a tool. It may, for some, be an edge. Or it may support other decision processes to ensure that there is at least some measure of valuation support. In either case, we would want the best tools available. Perpetuity at zero growth would not form part of that tool kit.


3. The market is forward looking. By the time you have revalued the firm using historical figures, the price has already moved to largely reflect it although some residual exists. Of all the fundamentals you can think of, companies move most on the basis of near term changes in forecast earnings. You will not participate.


4. You need to do more research until you get comfortable, or otherwise realize you have no idea what this thing is worth and try to make money in ways other than knowing what a company's fundamental value is. This is, actually, not entirely unreasonable. But this point is besides the issue to hand which is the relative worth of valuations using annuities or DCF. If you have no idea what earnings is, neither approach is likely to be of much use. Most people do have some idea. And some idea is usually better than no idea.


5. Enron, Merrill Lynch, Lehman...all made money. Whoops.

Watsapp...no earnings...USD 19bn...Whoops.

Risk is buying/holding a company for more than it is worth. In order to assess this risk, you need to have a reasonable shot at figuring out what it might be worth. Perpetuity and zero growth almost gives you zero chance of achieving this. It's underlying assumptions are significantly flawed from the outset, which is made worse because there are so few assumptions available to be made in this approach.


Overall, valuation is hard and seeing the future is hard. The difference in effort between producing an annuity style valuation and DCF is usually not that large in practice given sensitivities are concentrated into a small number of variables. Yet an annuity which simply perpetuates a current earnings stream to infinity is approximating virtually no company on the market at all. Flawed as this practice may be given uncertainty, a DCF can make at least some allowance for what is known in the market and adjust the earnings/FCF streams accordingly. Such a valuation approach is unlikely to be inferior to a perpetuity at zero growth.

 

[craft]

1. Although post-tax FCF is 0.7 x pre-Tax FCF, it is not the same thing as discounting post-tax FCF by an interest rate which is 0.7x the pre-tax interest rate. Compounding effects do surprising things. To see the fuller effect, do not assume the company stabilizes after five years and watch the valuations diverge materially. The discount rates required to make these NPVs have the anticipated relationship changes depending on your reinvestment assumption. If the dividend is fully franked I don't see this - care to make the point with an example so I can grasp what you are getting at?

One small wrinkle, I think Craft produced a valuation using pre-tax FCF and gross dividends using the pre-tax discount rate as is appropriate. He found the ratio of valuations differed by a ratio of 1.5% attributable entirely to the value of the PPL. The gross dividends in the growth phase, however, are not distributing full franking credits that accrued. Allowing for this, the after tax valuation would then be a little further lower than the pre-tax valuation to allow for franking not distributed (and never will be distributed under these assumptions). Agree with this - its basically the same as the levy issue.

Craft: This is a problem with using pre-tax DCF. On a net of tax basis, no adjustment would be required for your discount rate as the impact of the PPL is just an item that reduces discretionary FCF. The switch between the tax mix worsens your after tax NPV as you would expect if the discount rate is held constant. On a pre-tax basis, you continue to have the same FCF and, if you are thinking in pre-tax terms, this should just stay the same....but you know it isn't when it comes down to it. So I guess you should add some sort of 'fudge factor' to make allowance for it.

The general practice that I have seen is that these calcs are done on a full equity basis. That is, you need to reorganize the company accounts to produce the after tax FCF as if the company had no leverage. Mostly, that involves adding back the cost of interest payments and adjusting tax paid accordingly. This accounts for differences in capital structure across companies. You then discount that figure with what you would like the company to return if you owned the entire share base which can include considerations of your own capital structure - ie. you can factor in your own leverage etc. Because franking is a factor in Australia, you would also discount the value of franking to whatever degree you think is right for you (usually it is 60-100% to allow for the fact that not everyone values franking).


Cheers

Yep agree that eliminating the interest and adjusting for the tax shield (NOPLAT) is a purer valuation of equity. Personally I'm still more happy to do most of my work at the EBIT level as I'm first and foremost interested in business analysis and I hadn't felt the need to continue to NOPLAT. The PPL levy issue made me consider this again though as did your point above about franking credits that May not be distributed during a growth phase.

Thanks for your post - always thought provoking.

 

[luutzu]

Given your vast repository of knowledge, I'm surprised you think earnings = net operating cash flow. Crack open those text books again, my friend.

I really think you should read up on what you're saying because you have made a lot of very simple mistakes. Nothing wrong with that of course, I make them all the time, but you're on here telling everyone else they're wrong.

You never answered my question about how you're system would have fared in 2007, or with mining services stocks a couple of years back. The idea of jusy accepting today's earning at face value and extrapolating them to the end of time, and checking back in a couple of years if things don't work out would have had you shuffling down the footpath in your bathrobe and slippers.

I have just recently read the textbook actually. And over time, a company's net operating cash flow should approximate its reported net earnings. Do you want me to quote the textbook and page number?

I started investing around 2004, did OK but didn't really know what i was doing although any fool could have done well during those years... though 4 of my 5 stocks were taken over - Colorado, Coles, PHG [scaffolding company], Rio Tinto, but BHP weren't allowed by the European Union... so maybe i might have known a little of what i was doing.

and got completely smacked with the GFC while fully invested but my excuse was I was building a house, got a job to pay the loans... I know I wouldn't have seen it anyway if i were in the market.

----

I never said accepting today's earnings at face value and project it.
I'm saying that I would try to understand the company's business, its financial position, performance etc.. and from these try to see what its current earning power is... this could be the current year's, it could be what it had earned a few years' back before the boom... from these, I then discount into infinity... then repeat this next time there are material changes I can see happening to the company.

Give you an example:




I don't need crazy forecasts to tell you in two seconds that company will either go broke or raise more equity in 2008-2009.

 

[McLovin]

I have just recently read the textbook actually. And over time, a company's net operating cash flow should approximate its reported net earnings. Do you want me to quote the textbook and page number?

Net operating cash flow excludes non-cash items (like depreciation and amortisation) that net earnings do not. Your text book is wrong if it's saying the net ocf will approximate net earnings over time.

Anyway, you've got your system. Good luck with it.

I'm taking a break for a while.

 

[DeepState]

I have just recently read the textbook actually. And over time, a company's net operating cash flow should approximate its reported net earnings. Do you want me to quote the textbook and page number?

Luu

Although net earnings will converge to operating cashflow as reinvestment in the business ceases and the asset base is fully depreciated, that doesn't happen for businesses which are growing/shrinking their capital base for at least part of the forecast period. Using the same discount rate, the discounted value of FCF will not equal that of net earnings.

McLovin was correct to say you can't just substitute one for the other. It only becomes right under a very special condition of zero reinvestment through the entire forecast horizon.

Cheers

 

[DeepState]

I have just recently read the textbook actually. And over time, a company's net operating cash flow should approximate its reported net earnings. Do you want me to quote the textbook and page number?

Luu

Although net earnings will converge to operating cashflow as reinvestment in the business ceases and the asset base is fully depreciated, that doesn't happen for businesses which are growing/shrinking their capital base for at least part of the forecast period. Using the same discount rate, the discounted value of FCF will not equal that of net earnings.

McLovin was correct to say you can't just substitute one for the other. It only becomes right under a very special condition of zero reinvestment commencing from a fully depreciated asset base through the entire forecast horizon. Alternatively, it occurs when reinvestment matches depreciation over the horizon.

Cheers

 

[luutzu]

Luu

Although net earnings will converge to operating cashflow as reinvestment in the business ceases and the asset base is fully depreciated, that doesn't happen for businesses which are growing/shrinking their capital base for at least part of the forecast period. Using the same discount rate, the discounted value of FCF will not equal that of net earnings.

McLovin was correct to say you can't just substitute one for the other. It only becomes right under a very special condition of zero reinvestment through the entire forecast horizon.

Cheers

Thanks.

Yea, i don't mean that op cf will equal earnings, but i expect it to converge close to earnings but delayed. How many years behind earnings i guess will depends on the company i'm looking at.

I'm not quite sure what i'll do with net op cf figures just yet. Intention was that it be one of the few factors to see if management is playing around with reported earnings... could also have some predictive use of future earnings.

-------

I don't think our approaches are that different.... just that you guys try to put a more concrete estimates on growth etc while i simply approximate and already include it in the C [earnings].

I think you've misunderstood that I make a zero growth assumption. Growth is already implied in the C/i... that I expect it to grow at least by i. And if next year or there about it grows i+x, or if its growth is i-x... that growth has nothing to do with my expectations, it's just what the business does.


Then there's the pure asset play, or dividend play and has little to do with earnings or growth.
Say a historically well run company, with strong and conservative capital structure, able management, doing great business but its industry is stablising [not booming]... if i feel its future is relatively good considering, that it won't go broke as far as i can tell, that it's still making sales and have enough cash... and paying a dividends yield of 8 or 9% on the market price. I wouldn't see much wrong with buying it and hope for the best.

Your farm example there agrees with my understanding more than you think.

 

[luutzu]

Net operating cash flow excludes non-cash items (like depreciation and amortisation) that net earnings do not. Your text book is wrong if it's saying the net ocf will approximate net earnings over time.

Anyway, you've got your system. Good luck with it.

I'm taking a break for a while.

Bruce Lee to Young Grasshopper:
"Don't think! Feeeeelll... It is like a finger pointing away to the Moon. [pag!] Don't con.cent.trate on the fing.ger or you will miss all that heavenly glor...rryyyy. Do you understand? [Grasshopper bow... pag!] Don't ever take your eyes off your opponent, even when you bowwwww..."

thanks man. You guys are alright.

 

[DJG]

8-9% dividend yield - you obviously don't expect much..

 

[DeepState]

Thanks.


I think you've misunderstood that I make a zero growth assumption. Growth is already implied in the C/i... that I expect it to grow at least by i. And if next year or there about it grows i+x, or if its growth is i-x... that growth has nothing to do with my expectations, it's just what the business does.

Your farm example there agrees with my understanding more than you think.


Post #364

The calculation is to find the value of an expected cashflow [ie Earnings], at a required interest rate, extending that to infinity [or a couple hundred years, depend on how precise you want]

ie. PV = lim(n to infinity) C/i x [1 - (1/(1+i)to n ] (i checked with a textbook)

When you extend n far enough into the future, (1/(1+i)to n approaches zero... leaving you with PV = C/i


Post #365

But no, that's where i think there's a misconception about required rate of return and growth.

A required rate of return is return on capital, on equity, not growth in sales or growth in earnings by 8pc into infinity.

7 or 8%, depends... could be 6 or 9...

But no, that's where i think there's a misconception about required rate of return and growth.

A required rate of return is return on capital, on equity, not growth in sales or growth in earnings by 8pc into infinity.

So if the company earn 8% next year... and kept it back, i expect it will reinvest that 8% and compound it at 8 or higher... if not, return to me, its owner in terms of dividends or whatnot so i can find other opportunities.

No company can grow sales, let alone earnings, by 8% or 10 or 20... and i don't expect companies i look at to do so

Post #386

I remember reading/hearing an interview where Charles Munger said he never saw Buffett do a discounted cash flow model like the ones you guys are into

Luu

Your understanding of required rate of return and growth are misconceived.

If you expect growth to equal your discount rate (" Growth is already implied in the C/i... that I expect it to grow at least by i") and discount it by the same discount rate and are using a discount rate like 8%, your company is worth infinity. Your earnings will become the world economy as I said before.

You have misunderstood and misapplied a perpetuity model. C/i Is the present value of C into perpetuity if C does not change...zero growth. Check your text book and read it again. ("PV = lim(n to infinity) C/i x [1 - (1/(1+i)to n ] (i checked with a textbook)")

Buffett allows for growth. Your model does not. That's about the biggest difference imaginable for valuation purposes. There is no more important factor than the assumed difference between perpetuity growth and discount rate. "Your farm example there agrees with my understanding more than you think."...but your understanding is fundamentally inaccurate, he forecasts path ways that fade as opposed to extrapolates forever, allows for growth rates different to discount rates....each of these violates your understanding as documented in your posts and actions as similarly detailed.

If you want to persist with perpetuities for all companies despite the fact that this is far more inaccurate than even simple allowances for known factors that will effect earnings in the coming period, what you actually need is the Gordon Growth Model. PV = c x (g/(i-g)) where g is the growth rate. When g = 0, meaning no growth, the Gordon Growth model collapses into what you have been using to justify your valuations. That's a pretty big error. If you think that the rate of growth equals your discount rate ("Growth is already implied in the C/i... that I expect it to grow at least by i. " and where i is your "required interest rate") then if i=g and the present value goes to infinity.

"No company can grow sales, let alone earnings, by 8". Yet when you are using an required interest rate of "7 or 8%, depends... could be 6 or 9..." it basically tells you that the growth rate that you assume must be different to the required discount rate and, yet, your current formulation makes no allowance for differences between your assumed growth rate and required discount rate. Didn't this occur to you?

Luu...these are big errors. One possible reason your valuations make no sense to you and why you find no worth in pursuing more expansive approaches is that your approach is built from fundamental misconceptions from the outset.

Cheers

 

[Ves]

1. Although post-tax FCF is 0.7 x pre-Tax FCF, it is not the same thing as discounting post-tax FCF by an interest rate which is 0.7x the pre-tax interest rate. Compounding effects do surprising things. To see the fuller effect, do not assume the company stabilizes after five years and watch the valuations diverge materially. The discount rates required to make these NPVs have the anticipated relationship changes depending on your reinvestment assumption.

One small wrinkle, I think Craft produced a valuation using pre-tax FCF and gross dividends using the pre-tax discount rate as is appropriate. He found the ratio of valuations differed by a ratio of 1.5% attributable entirely to the value of the PPL. The gross dividends in the growth phase, however, are not distributing full franking credits that accrued. Allowing for this, the after tax valuation would then be a little further lower than the pre-tax valuation to allow for franking not distributed (and never will be distributed under these assumptions).

Hi RY

Thank you for all of your posts. I have read them, but so far haven't consumed them fully enough to warrant me to respond to a lot of the content.

However, this part quoted above interests me. So I experimented with the model in one of my previous posts (the one you quoted the results of with the pre-tax and post-tax calculations and franking credits, with the two discount rates).

And it appears that you are right, the higher the amount of earnings reinvested that produces returns above the cost of capital / discount rate, the bottom line after-tax NPV was higher than the before tax NPV. However, there is a point where if the payout ratio of dividends is increased the Before-tax NPV is higher than the After-Tax NPV.

It is possible that my model is based on an erroneous assumption (ie. the discount rates of 15% pre tax and 10.% after tax are not suitable) and that is the main cause of the differences in the observed results.

But what you say about compounding does make sense too. Although, I have not admittedly got my head around the whole thing yet.

 

[craft]

1. Although post-tax FCF is 0.7 x pre-Tax FCF, it is not the same thing as discounting post-tax FCF by an interest rate which is 0.7x the pre-tax interest rate. Compounding effects do surprising things. To see the fuller effect, do not assume the company stabilizes after five years and watch the valuations diverge materially. The discount rates required to make these NPVs have the anticipated relationship changes depending on your reinvestment assumption.

Hmmm You are right. Some sort of geometric thing going on with compounding maybe?

Do you know if there is a formula to calculate the right conversion given the reinvestment rate? My brain can't stretch far enough to work it out backwards.

Big thanks.

 

[luutzu]

Luu

Your understanding of required rate of return and growth are misconceived.

If you expect growth to equal your discount rate (" Growth is already implied in the C/i... that I expect it to grow at least by i") and discount it by the same discount rate and are using a discount rate like 8%, your company is worth infinity. Your earnings will become the world economy as I said before.

You have misunderstood and misapplied a perpetuity model. C/i Is the present value of C into perpetuity if C does not change...zero growth. Check your text book and read it again. ("PV = lim(n to infinity) C/i x [1 - (1/(1+i)to n ] (i checked with a textbook)")

Buffett allows for growth. Your model does not. That's about the biggest difference imaginable for valuation purposes. There is no more important factor than the assumed difference between perpetuity growth and discount rate. "Your farm example there agrees with my understanding more than you think."...but your understanding is fundamentally inaccurate, he forecasts path ways that fade as opposed to extrapolates forever, allows for growth rates different to discount rates....each of these violates your understanding as documented in your posts and actions as similarly detailed.

If you want to persist with perpetuities for all companies despite the fact that this is far more inaccurate than even simple allowances for known factors that will effect earnings in the coming period, what you actually need is the Gordon Growth Model. PV = c x (g/(i-g)) where g is the growth rate. When g = 0, meaning no growth, the Gordon Growth model collapses into what you have been using to justify your valuations. That's a pretty big error. If you think that the rate of growth equals your discount rate ("Growth is already implied in the C/i... that I expect it to grow at least by i. " and where i is your "required interest rate") then if i=g and the present value goes to infinity.

"No company can grow sales, let alone earnings, by 8". Yet when you are using an required interest rate of "7 or 8%, depends... could be 6 or 9..." it basically tells you that the growth rate that you assume must be different to the required discount rate and, yet, your current formulation makes no allowance for differences between your assumed growth rate and required discount rate. Didn't this occur to you?

Luu...these are big errors. One possible reason your valuations make no sense to you and why you find no worth in pursuing more expansive approaches is that your approach is built from fundamental misconceptions from the outset.

Cheers

My required rate of return and the company's ability to meet or surpass it are independent.

If i am able to find investment opportunities that could return me that required rate, and keep on doing it, then yea, I might take over the world. But that's just what I planned to do, not what the companies i bought into could or would do.

Say I've done a great amount of work, know a very good amount about the company, its assets, position and all that... and figured it, right now or within a year or two, could earn $100 a year, very steadily, quite easily... and could continue to do that for a foreseeable future.

That means this company is paying its owner a coupon, an annuity, of $100 per year until the end of the world as far as i am concern.

My required rate, or my cost of capital or my opportunity costs etc... say that rate is 8% a year.

What would an asset that earns $100, return at my rate and do so for ever be worth?

100/0.08 = $1,250.


What is the current price its owners are asking? the max i will go is around $1250, but depends on the strength and my best guesses, I could pay a bit higher [lowering my required rate because of more certainty or higher probability of real earning being more than $100]...

If it's asking at my value, I probably wouldn't pay for it, I'd preferred to pay lower [that's my margin of safety].

--------

Say I bought it... and over the next five years, there are no inflation etc... and the company pay all its earnings as dividend to me, the only owner. The company is then still valued by me to be $1250.

It doesn't need to have its sales growth or its earnings growth by 8% to keep me happy... it just need to return at least that yield from the initial capital i put into it [the price i paid, or at least the initial value].

So the excess profit of 8% that i required, if returned to me, will become my problem and I will have to find another opportunity that could do 8% minimum.

If the company that i now own then earn 10% on its capital/equity, I hope it would keep the dividends and keep growing at that... and hope it continue to do so until its growth and its opportunities no longer able to do that... then return as dividends.


---

When you use the projected growth, stages of growth... It does not make sense to me because:

1. You are saying that the company's present value is from its earnings stream, and its earnings will grow at g rate up to 5 or 10 years, then at g2 until enternity... and by discounting these future earnings [growing at g, then g2]... and discount at rate r1, then r2... the present value of the company is X.


That, to me, is paying for a future that you cannot predict with any accuracy.

You guys are basically making assumptions about future earnings, then discount that future earnings growth assumptions back and value those assumptions.

What if your assumptions are too optimistic? What if it is right, but the timing is slightly off?

------

So by me saying that growth is already implied in C, and growing at least by rate i.... I don't mean to say that C is to grow at rate i forever.

I mean that in the next couple of years, it could grow at i... then after that, I'll take a look again to see if it could still grow at i... and depends on my costs and greater expectations, depends on the company's earning capacity... if the C is bigger then, it will be more valuable... if lower, less valuable [assuming i is still the same]

--------

SUMMARY:

It makes no sense to me to say... this company is now earning $100... but this is to grow, given the industry, given its history, given inflation... will growth by 10% p.a. So its prevent value is to discount $110 back one year at r1, plus discount [110x1.1] 2 years... and so on until the 10th year where it will grow at g2, but by then the costs of capital will be r2 and so you discount and so on...


It makes more sense to me, and probably make me richer, to ask this: from what i can figured, this company is earning $100 right now... i require 8%, is the present value [the price] being offered allow me to achieve that?

And again, I don't come to $100 earnings based on reported earnings.
That figure is where i will concentrate my efforts... the future can wait.

---

"The Warriors of old act only when it is profitable to do so;
"He only engage in battle when victory is all but guaranteed.
- Sun Tzu.

You cannot make plans that says: I will whoop their hind if they first go to A, then act as B, then they will move to C then victory will be mine. But if they move to A, then for strange reasons ignore B and move to B+1, then i'm wrong.

That's exactly what you guys are doing with detailed discounted cash flow modellings.