Time to buy Fairfax again?

[giofranchi]

it seems unrealistic given the asset mix.

 

Obviously, the asset mix is not a static thing. It will change.

 

With regard to bonds yield is not all that matters, there is also capital appreciation. What would happen if the yield on 10-years US government bonds goes from 2.6% today to 1% like in Germany?

 

Gio

 

[thepupil]

well if that happened,  AGG would go up by it's duration of ~5% (X 1.5) + a little bit for convexity since duration increases as bonds fall, let's call it 10 or 12%. So Fairfax would have a massive one-time gain and then the prospective return going forward would be even worse.

 

But you know fairfax and what kind of bonds they hold better than i do so maybe there is more to it than that.

[ERICOPOLY]

Richard believes the appropriate valuation is to DCF their future six-sigma returns. 

 

Please don't misrepresent my position.

 

I haven't.

 

I said that if they are indeed value investors, and if they indeed have a portfolio of stocks and bonds that is marked-to-market below intrinsic value....  then it must be the case that FFH itself should not be trading at intrinsic value.

 

You disagreed, saying that we must DCF those future capital gains -- therefore, the only way for that to be the case is if the market puts the "magic hat" premium on FFH such that the current market price reflects the intrinsic value of the stocks and bonds that they currently hold.

 

It's okay though if you don't even realize the implications of what you said, but it's not a misrepresentation.

[giofranchi]

So Fairfax would have a massive one-time gain and then the prospective return going forward would be even worse.

 

As I have said, I think other distressed debt opportunities might present themselves, and FFH might shift their bonds portfolio from plain vanilla government bonds to distressed debt opportunities.

 

As I have said, at least in part it already happened in 2009 with a large purchase of municipal bonds in California. And look at what Dan Loeb has done with Greek and Portuguese government bonds. And Howard Marks is raising a huge sum of money to take advantage of distressed debt opportunities he sees coming.

 

As I have said, when distressed debt opportunities arise again, I think many will seek the (relative) safety of US government bonds, pushing their yields down (like Japan, like Germany… like even France!!). Then, FFH might sell their US government bonds and redeploy their proceedings in newly formed distressed debt opportunities.

 

Not saying it will happen… but I don’t think it is so far-fetched either!

 

Gio

 

[thepupil]

so let's look at the interim report

 

$3.7B in cash (0 duration, no risk)

$3.7B in short term treasuries and other government bonds (0 duration no risk)

$10.2B in bonds, about half of which are 5 yrs and in and 40% of which are 10 yrs and out <---so you'd get some big capital appreciation here with rates falling, let's say that $4B is all in 30 yrs with an 18 ish duration, if rates fell 1.5% these would go up by 33 ish% so you'd add a billion and change to book value, not exactly life changing.

 

And the prospective return going forward would be terrible!

 

so in my humble opinion, you want rates to rise to increase the return on the portfolio without moving down into equities or lower quality bonds. this would obviously hurt prices and decrease book on the way there but would be better for increasing ROE and P/B.

 

I understand asset mix is not static and i do not know anything about Fairfax's capital requirements or if all that cash and fixed income is excess capital that can be deployed into higher returning things eventually, but in its current form there is simply no way to earn 6% on the portfolio.

 

I don't know if they have the ability to shift to higher risk things like distressed debt and equities. I assume you do and that's why you have so much confidence, but in my opinion it's tough to bake in your assumptions future distressed opportunities. Anyways, that's all i have to say.

[giofranchi]

in its current form there is simply no way to earn 6% on the portfolio.

 

I agree. It is already 4 years now their return on investments has been much lower than 6%. In 2014 it was… negative!

But I guess that’s just how they structure their portfolio: periods of returns way below average have always been followed by periods of returns way above average. I can count only 5 years of average returns in their whole history.

 

in my opinion it's tough to bake in your assumptions future distressed opportunities.

 

The fact we will see many distressed debt opportunities in the future, given the unprecedented levels of indebtedness the whole developed economies find themselves right now, is imo a very probable scenario… And Howard Marks seems to agree.

The fact FFH might be able to take advantage of those opportunities when they arise is another matter… But they have already done so in California 6 years ago… Why shouldn’t they be able to replicate?

 

Gio

[tombgrt]

Good points thepupil. And to achieve a minimum 15% CAGR over a rolling 10-year period, I guess FFH has to achieve 20-25%+  CAGR (I am to lazy to make a decent calculation now) in the next 6 years. Or am I not thinking long term enough? ;x

[thepupil]

3 yrs of 0% return followed by 3 yrs of 20% return for 15 years is 8% CAGR if i did my math right.

 

throw in a 100% return in year 3 just for fun and it's a 13% CAGR. getting to 15% with prolonged periods of 0% is REALLY hard.

 

i understand the whole "shoot for the moon and if you miss you'll land amongst the stars" thing in that you'd be very happy if Fairfax grew by 12% and that would provide a satisfactory return. 

 

But in my view, Happiness = Reality - Expectations

 

It is easier to decrease expectations than to change reality, and the 15% CAGR expectation is certainly in need of a big decrease

[giofranchi]

getting to 15% with prolonged periods of 0% is REALLY hard.

 

I haven't said 'prolonged periods of 0%'. Actually, the years in which returns were negative are only 3 in their whole history.

 

15% is the favorable outcome I might expect... favorable but not impossible.

The average insurance company is the negative outcome I might expect.

Call whatever is in between the most probable outcome I might expect.

 

But let me ask you: do you know many companies with at least the possibility to compound at 15%, and led by trustworthy people like PW, Bradstreet, Barnard, etc., in a business you understand and which is practically not subject to structural changes?

 

If yes, I am all ears... Try to convince me! I would be very glad if you'd succeed! ;)

 

Gio

[Guest longinvestor]

3 yrs of 0% return followed by 3 yrs of 20% return for 15 years is 8% CAGR if i did my math right.

 

throw in a 100% return in year 3 just for fun and it's a 13% CAGR. getting to 15% with prolonged periods of 0% is REALLY hard.

 

i understand the whole "shoot for the moon and if you miss you'll land amongst the stars" thing in that you'd be very happy if Fairfax grew by 12% and that would provide a satisfactory return. 

 

But in my view, Happiness = Reality - Expectations

 

It is easier to decrease expectations than to change reality, and the 15% CAGR expectation is certainly in need of a big decrease

Laying out expectations of 15% CAGR is something Prem has done, Buffett never does. This is a significant point of departure.

[RichardGibbons]

Richard believes the appropriate valuation is to DCF their future six-sigma returns. 

 

Please don't misrepresent my position.

 

I haven't.

 

I said that if they are indeed value investors, and if they indeed have a portfolio of stocks and bonds that is marked-to-market below intrinsic value....  then it must be the case that FFH itself should not be trading at intrinsic value.

 

You disagreed, saying that we must DCF those future capital gains -- therefore, the only way for that to be the case is if the market puts the "magic hat" premium on FFH such that the current market price reflects the intrinsic value of the stocks and bonds that they currently hold.

 

It's okay though if you don't even realize the implications of what you said, but it's not a misrepresentation.

 

Sorry, where did I say that I think they're going to have six-sigma returns?  Where did I say that I think that I believe their portfolio is trading below fair value?  Heck, maybe I think that they are likely to under perform the market.

 

Please don't misrepresent my position.

[ERICOPOLY]

Richard believes the appropriate valuation is to DCF their future six-sigma returns. 

 

Please don't misrepresent my position.

 

I haven't.

 

I said that if they are indeed value investors, and if they indeed have a portfolio of stocks and bonds that is marked-to-market below intrinsic value....  then it must be the case that FFH itself should not be trading at intrinsic value.

 

You disagreed, saying that we must DCF those future capital gains -- therefore, the only way for that to be the case is if the market puts the "magic hat" premium on FFH such that the current market price reflects the intrinsic value of the stocks and bonds that they currently hold.

 

It's okay though if you don't even realize the implications of what you said, but it's not a misrepresentation.

 

Sorry, where did I say that I think they're going to have six-sigma returns? 

 

You alluded (in a hypothetical) to paying a DCF for expectations of six-sigma returns.  You stated that if you believed that they were going to make 1,000% per year then it would be appropriate to value their expected six-sigma returns using DCF.

Is it an exaggeration to mark 1,000% a year as "six sigma"?  Is it only two sigma?  How many sigmas is 1,000% a year?

 

 

 

It was in an example you provided.  I was still referring to the Fairfax of your example.  You said it would be appropriate to discount the future cash flows if you expected them to make 1,000% a year.

 

Suppose you were confident that Fairfax, without insurance operations, was able to reliably compound its book value at 1000% a year through investing.  Saying, "that stock is worth book value" makes no sense because its future free cash flows are worth far more than that.

 

 

 

[RichardGibbons]

Richard believes the appropriate valuation is to DCF their future six-sigma returns. 

 

Please don't misrepresent my position.

 

I haven't.

 

I said that if they are indeed value investors, and if they indeed have a portfolio of stocks and bonds that is marked-to-market below intrinsic value....  then it must be the case that FFH itself should not be trading at intrinsic value.

 

You disagreed, saying that we must DCF those future capital gains -- therefore, the only way for that to be the case is if the market puts the "magic hat" premium on FFH such that the current market price reflects the intrinsic value of the stocks and bonds that they currently hold.

 

It's okay though if you don't even realize the implications of what you said, but it's not a misrepresentation.

 

Sorry, where did I say that I think they're going to have six-sigma returns?  Where did I say that I think that I believe their portfolio is trading below fair value?  Heck, maybe I think that they are likely to under perform the market.

 

Please don't misrepresent my position.

 

First, you disapprove of valuing their securities at market.

 

Now, you claim you can't tell if they overperform or will underperform the market.

 

Which Richard will we hear from next?

 

I also haven't said that I disapprove or approve of valuing their securities at market, or claimed that I can't tell if they will overperform or underperform the market.

 

Oh, I think I figured out why you're so confused.  You believe that someone saying "I didn't say X" means "I believe the opposite of X".  That isn't the case, OK?  Saying "I didn't say X" doesn't imply anything about the my beliefs about X.

 

The Richard you'll hear from next is the one who doesn't want you to misrepresent his position, because he's proven you wrong already, is bored of the discussion, but doesn't want people to believe lies about his beliefs.  Same one as the last 3 messages.

 

Please don't misrepresent my position.  <== This is what I'm saying (just in case it isn't clear)

 

[RichardGibbons]

Also, please don't misrepresent my position.

[ERICOPOLY]

I also haven't said that I disapprove or approve of valuing their securities at market, or claimed that I can't tell if they will overperform or underperform the market.

 

You say you didn't object to valuing securities at market?

 

The mark-to-market values on the books reflects the market's DCF for the businesses underlying those securities.  Your objection to assigning market value to them implies that you expect those market values to be incorrectly discounting the future cash flows derived from holding those investments.

 

You specifically wrote it is "a pretty bad idea" to use those values.  You said that instead of using the mark-to-market values, you wanted to include a DCF derived premium because this is an investment business and you lectured about how the cash flows from an investment business need to be DCF'd.  Well, how are you expecting the cash flows to be different from the ones predicted by Mr. Market’s DCF unless you are predicting that the stock pickers at the investment company will either over perform or underperform the market?

 

Unless a "Magic Hat" premium is thrown in there because these guys are seers of the future or something like that.  Such a premium has to be agreed to by Mr. Market or it will vanish after you've paid it.  That's a tall order to expect the general consensus to remain that HWIC will just reliably beat the market to such a certainty as to warrant a DCF on that "alpha".  My objection is that you think an investor should go ahead and do that, you gave the reason that it's a business.  You are therefore advocating for the Magic Hat premium.

 

 

I think Dhandho, just like every other business, should trade at multiple that is dependent on its discounted future free cash flows.  If you think that Dhandho is going to greatly outperform the returns of the market, then you're saying is the equivalent to saying that Dhandho should trade greatly below the value of its future cash flows.  And if you believe in DCF for other companies, but not for investing companies, you're saying that Dhando should be cheaper than those other companies that you do value using DCF.

 

To me, that's a pretty bad idea.  (This is the funny thing about this discussion.  If you follow through your reasoning to the natural conclusions, you very quickly reach a contradiction that should indicate to you that your premise is faulty.  Well, unless you want to throw out DCF entirely.)

 

 

 

[ERICOPOLY]

I also haven't said that I disapprove or approve of valuing their securities at market

 

Right, so why does it "make no sense" to value the securities at market if you don't disapprove of valuing their securities at market?

 

To me this looks like two different Richards are posting their thoughts.

[ERICOPOLY]

Richard believes the appropriate valuation is to DCF their future six-sigma returns. 

 

Please don't misrepresent my position.

 

I haven't.

 

I said that if they are indeed value investors, and if they indeed have a portfolio of stocks and bonds that is marked-to-market below intrinsic value....  then it must be the case that FFH itself should not be trading at intrinsic value.

 

You disagreed, saying that we must DCF those future capital gains -- therefore, the only way for that to be the case is if the market puts the "magic hat" premium on FFH such that the current market price reflects the intrinsic value of the stocks and bonds that they currently hold.

 

It's okay though if you don't even realize the implications of what you said, but it's not a misrepresentation.

 

Sorry, where did I say that I think they're going to have six-sigma returns? 

 

You alluded (in a hypothetical) to paying a DCF for expectations of six-sigma returns.  You stated that if you believed that they were going to make 1,000% per year then it would be appropriate to value their expected six-sigma returns using DCF.

Is it an exaggeration to mark 1,000% a year as "six sigma"?  Is it only two sigma?  How many sigmas is 1,000% a year?

 

 

 

It was in an example you provided.  I was still referring to the Fairfax of your example.  You said it would be appropriate to discount the future cash flows if you expected them to make 1,000% a year.

 

Suppose you were confident that Fairfax, without insurance operations, was able to reliably compound its book value at 1000% a year through investing.  Saying, "that stock is worth book value" makes no sense because its future free cash flows are worth far more than that.

 

[RichardGibbons]

Suppose you were confident that Fairfax, without insurance operations, was able to reliably compound its book value at 1000% a year through investing.  Saying, "that stock is worth book value" makes no sense because its future free cash flows are worth far more than that.

 

Try to reconcile that comment "makes no sense" with this one:

 

I also haven't said that I disapprove or approve of valuing their securities at market

 

Right, so why does it "make no sense" to value the securities at market if you don't disapprove of valuing their securities at market?

 

To me this looks like two different Richards are posting their thoughts.

 

Yeah, that's because there's a difference between "valuing their securities at market" and valuing their security (i.e. shares of FFH) at the value of the securities they hold.  Those really aren't the same thing.

[RichardGibbons]

You alluded (in a hypothetical) to paying a DCF for expectations of six-sigma returns.  You stated that if you believed that they were going to make 1,000% per year then it would be appropriate to value their expected six-sigma returns using DCF.

Is it an exaggeration to mark 1,000% a year as "six sigma"?  Is it only two sigma?  How many sigmas is 1,000% a year?

 

It was in an example you provided.  I was still referring to the Fairfax of your example.  You said it would be appropriate to discount the future cash flows if you expected them to make 1,000% a year.

 

So, up there you say "in a hypothetical". So do you understand what a hypothetical means?  Can you make the leap to figure out why taking a hypothetical and using it as evidence to say that's what I think Fairfax should be worth, is misrepresenting my view?

 

To make it clear, any hypothetical examples I use aren't necessarily representative of how I view reality.  For instance, if I say, "hypothetically, if China attempted to land an army in California, USA would probably try to stop them", I'm not saying "hey, China's attempting to land an army in California!"  And if you say "Richard's saying China is attempting to land an army in California", you are misrepresenting my views.  (To be explicit, as far as I know, China is not attempting to land an army in California.)

 

Is that clear?  (Please tell me that you've figured out that the is conversation has long since past the stage where it became pointless and boring....)

 

 

[ERICOPOLY]

Suppose you were confident that Fairfax, without insurance operations, was able to reliably compound its book value at 1000% a year through investing.  Saying, "that stock is worth book value" makes no sense because its future free cash flows are worth far more than that.

 

Try to reconcile that comment "makes no sense" with this one:

 

I also haven't said that I disapprove or approve of valuing their securities at market

 

Right, so why does it "make no sense" to value the securities at market if you don't disapprove of valuing their securities at market?

 

To me this looks like two different Richards are posting their thoughts.

 

Yeah, that's because there's a difference between "valuing their securities at market" and valuing their security (i.e. shares of FFH) at the value of the securities they hold.  Those really aren't the same thing.

 

That's true but it's off-topic.

 

In the first quote I provided above, you said that valuing it (hypothetical Fairfax) at book value made no sense.  Thus, in that quote you are valuing the securities on the balance sheet above market. 

 

That conflicts with the second quote where you say that you don't do that.

[RichardGibbons]

Suppose you were confident that Fairfax, without insurance operations, was able to reliably compound its book value at 1000% a year through investing.  Saying, "that stock is worth book value" makes no sense because its future free cash flows are worth far more than that.

 

Try to reconcile that comment "makes no sense" with this one:

 

I also haven't said that I disapprove or approve of valuing their securities at market

 

Right, so why does it "make no sense" to value the securities at market if you don't disapprove of valuing their securities at market?

 

To me this looks like two different Richards are posting their thoughts.

 

Yeah, that's because there's a difference between "valuing their securities at market" and valuing their security (i.e. shares of FFH) at the value of the securities they hold.  Those really aren't the same thing.

 

That's true but it's off-topic.

 

In the first quote I provided above, you said that valuing it (hypothetical Fairfax) at book value made no sense.  Thus, in that quote you are valuing the securities on the balance sheet above market. 

 

That conflicts with the second quote where you say that you don't do that.

 

No, I'm not.  I'm valuing the security (hypothetical FFH) above book value.  I say nothing about the value of the securities on their book.  (And even if you assume book value is consisting entirely of securities at their current market price, I'm still saying nothing about the value of the securities on their book.)

 

 

[ERICOPOLY]

You alluded (in a hypothetical) to paying a DCF for expectations of six-sigma returns.  You stated that if you believed that they were going to make 1,000% per year then it would be appropriate to value their expected six-sigma returns using DCF.

Is it an exaggeration to mark 1,000% a year as "six sigma"?  Is it only two sigma?  How many sigmas is 1,000% a year?

 

It was in an example you provided.  I was still referring to the Fairfax of your example.  You said it would be appropriate to discount the future cash flows if you expected them to make 1,000% a year.

 

So, up there you say "in a hypothetical". So do you understand what a hypothetical means?  Can you make the leap to figure out why taking a hypothetical and using it as evidence to say that's what I think Fairfax should be worth, is misrepresenting my view?

 

 

This is the post you are objecting to:

 

Take that to it's logical conclusion -- people aren't going to pay "intrinsic value" for Fairfax when that value depends on them being six-sigma "seers" of the stock and bond markets.  It's a ludicrous expectation to hold for Mr. Market.

 

Eric,

if I had said that, I would have predicted an expansion in the multiple FFH will be priced tomorrow. Have I said that? No!

 

Yeah, sorry.

 

Richard said it.

 

Richard believes the appropriate valuation is to DCF their future six-sigma returns. 

 

Perils of multitasking with you guys.  I apologize if I misrepresented your position for Richard's.

 

Nobody has ever literally accused you of believing that Fairfax will deliver six-sigma returns.

 

I told Gio that it was Richard when I said "Richard said it".

 

By saying "it", I'm not saying that Richard said he is expecting six-sigma returns from Fairfax.

 

In that statement I'm saying it was Richard that said he'd pay for expected "alpha".  And you said so.  Go back to that hypothetical 1000% example and you say precisely that.

 

That's why I'm saying I was referring to the hypothetical example.  That hypothetical example was where you stated you would DCF the future alpha if you believed there to be future alpha.  And so I mixed up Gio's message with yours and mistakenly accused Gio of paying upfront for alpha.  Then he said he didn't say that, so I said something to the effect of "sorry, oh yes, that was Richard".  Not the exact statement "six sigma", but rather the viewpoint that expected alpha should be bought and paid for upfront.

 

It is your viewpoint however that expected "alpha" should be bought and paid for upfront, right?  That's how I've understood all of your comments.

 

Therefore I have disagreed with all of your comments. 

 

[ERICOPOLY]

And even if you assume book value is consisting entirely of securities at their current market price, I'm still saying nothing about the value of the securities on their book.

 

1) 

The securities in the hypothetical FFH's portfolio are already DCFd by Mr. Market. They represent the future cash flows expected by Mr Market discounted to the present

2) 

That collective DCF is the only thing that comprises "book value" for the hypothetical FFH.

3)

What is your DCF based on that brings you to pay a premium to this value if you are agnostic as to the value of the securities on this book?

4)

If you say "managerial outperformance", why isn't that tied to your expectation of the value of the securities on the books?  I personally don't expect outperformance from a portfolio of securities that I hold no opinion on, so when you say that you do expect outperformance from securities that you hold no opinion on I wonder how you can hold that view.  And if you think it's because the investors are really awesome that you expect outperformance, I find it unlikely that you have no opinion as to whether the securities are undervalued -- how are you expecting outperformance otherwise.

5) 

And so when you say you are paying for outperformance... I don't see how you can detach that comment from the implication that the assets are undervalued.  So when you say the former, I see the latter as well.

 

So is there some special corner case that you are arguing about where you've found a company that you'll pay a premium for even though nothing leads you to believe that the assets are undervalued?  That would be interesting for you to explain further.

[ERICOPOLY]

There is also the example earlier on where I said that Fairfax should never trade for intrinsic value if they are indeed value investors.

 

You disagreed.

 

However think more about that example for just a minute.  The market (in order to get the DCF right) would need to know how undervalued it's securities were.  But in order to know that, it would have to disagree with itself about the very value of those assets.

 

To me, that is an unreasonable expectation for Mr. Market.

[ERICOPOLY]

Or some examples...

 

Example 1:

In the case where you "know" the current portfolio of hypothetical FFH to be 50 cent dollars on average.  Does this knowledge affect your DCF calculation?

 

Example 2: 

In the case where you "know" the current portfolio of hypothetical FFH to be 75 cent dollars on average.  Does this knowledge affect your DCF calculation versus Example 1?

 

Example 3: 

In the case where you "know" the current portfolio of hypothetical FFH to be 100 cent dollars on average.  Does this knowledge affect your DCF calculation versus Example 1 & 2?

 

 

I provided these examples to hopefully demonstrate that whatever you give for DCF, it does in fact say something about your opinion of the value of the securities in the portfolio. 

 

That's because the DFC depends not just on how successful future investments are, but the current ones as well.

 

This is why I think there are two Richards commenting here:  one uses a method that's output depends on knowledge of the value of the securities as input, and the other claims to have no opinion on the value of the securities.