ERICOPOLY

HELOCs are full recourse. This is why every time somebody talks about Wells Fargo's underwater HELOC exposure I wonder why the same people don't make a big stink over credit card exposure. After all, both are full recourse and neither is backed by an asset. Why should underwater HELOCs be written down any more than credit card debt, given the same delinquency numbers (if the same)?

There you go, that's true. On one of the conference calls, they claimed that they look for deals were the hurdle rate is 12% (my understanding is that 12% is what they would earn on the deal if no leverage were involved). ROIC is 21% on such a deal with 40% down payment and 6% interest rates.

Now, if we use the Depression as a yardstick to measure any "investment", then we might also use Weimar Germany too. Thus even the zero coupon bonds are a terrible investment. The question on the thread wasn't about whether or not a Depression was coming, it was about mild deflation or deflation. Perhaps this is a question about what to buy for the 1920s, a period of mild deflation. Or Japan -- had you bought real estate in Japan with cap rate north of 10% you'd be doing fine today. The trouble is, cap rates in Japan weren't at 10% in 1989 which is why the price fell. Like all things, it's best to buy when price to cash flow is well and truly out of whack with the median levels. It doesn't surprise me at all that the Fairfax officers didn't mention real estate as a good investment right now -- it's completely below their radar to be buying $60,000 houses for rentals. You might as well ask Buffett which microcap stocks he's considering for Berkshire's portfolio of stocks. EDIT: Actually, were it not for the government intervention in 2008 Berkshire would already be bankrupt by now. So that's not a good investment either.

I hope they aren't borrowing at 10% to buy assets at 12%.

Prem once said "we do not invest in gold" on a conference call -- was somewhere in the 2007-2009 time frame.

The returns are higher on owned homes vs rentals, because you don't have to pay tax on the implied rent you are earning. Everybody has to live somewhere -- owners pay rent... it's implicit. It's the rent they would otherwise be earning if they moved out and rented it... but then they'd have to pay rent themselves to live elsewhere. Back it up a step so they don't move out and live elsewhere, and one can see what is going on -- both your own landlord and tenant. This is why owning a home is the best investment most people will ever make. Hardly anyone can find another investment that would dependably throw off enough after-tax income to pay their rent, and 100% adjusting in lock step when rates rise. But it's not in vogue to call it an investment. Yes, you can stupidly take on too much mortgage, but you can also stupidly rent a place beyond your means that leaves you with no money left for savings. So that's a wash.

Are we going to see a rebirth in US labor and manufacturing if China cannot afford it's currency peg anymore? Or will it just go to some other low cost nation, perhaps with labor that already cheaper than China's at the present.

Nah I think you're right. There's the cost of leverage, which is already accounted for. But the risk associated with leverage is something else entirely - failure to pay, failure to refinance, failure to meet various covenants. So those are worthwhile considerations and would have an impact on valuation, but I'm not sure how big of an impact it would have. Failure to refinance and meeting covenants seem remote. Failure to pay is offset by their high credit customer base and proven chartering model. So it's a factor, but not a huge one IMO - maybe from 10x to 9x DCF? Assuming 10x DCF is the right price for the no leverage scenario. It would be interesting if they would start a new unleveraged version of themselves. Someone could conceptually go long on the new unleveraged one trading at 10x DCF and hedge by shorting the highly leveraged one also trading at 10x DCF. Whenever on the cusp of financial crisis, this trade should work every time.

Agreed! Where is inflation at this point however? Everyone talks about it but I can't see it in the numbers (I agree that it depends on how you count it). To me you need to have continuous money printing exceeding deflationary trends/money-credit destruction, wage inflation accompanied by spending/velocity/credit increase or a combination of both in the long run to get to inflation. One more thing that could produce the most significant form (house price) of asset inflation: Building them slower than growing demand. Are we building them slowly? Granted, hasn't yet led to anything but price declines, however we all concede that we started with too many. What happens to prices when we have too few? I'll wager that a significant portion of that personal debt per capita is at least partially if not substantially backed by real estate.

Then there are the people who say that companies have boosted their profit margins by cutting labor costs, and that the trend must revert to the mean. That wouldn't be deflationary, but it would hurt some equities (not the banks or home builders). EDIT: Here is the link to "The Invisible Stock Bubble": http://www.smartmoney.com/invest/stocks/the-invisible-stock-bubble-1305647031991/?cid=sm_dailyfinanceRSS Of course, if incomes are temporarily depressed doesn't that mean that the personal debt bubble isn't as big as people say?

Fairfax's deflation bet is pretty small (as percentage of equity) compared to their previous CDS bet (which was something like 10% of equity). This despite the fact that the CPI contracts are 10 yrs duration vs 5 yrs duration for the CDS. They are worried about their equity prices however, clearly. I think the first 25% drop in the Russell 2000 will get them back to about the point where they bought those hedges... and I think the index traded down to a point 30% below current levels last year, and they didn't close them out at those levels.

You are right. There are some intangibles there which are worth paying a premium for. However, you'd have that value even if they didn't employ leverage. Once they leverage it, should "fair price" come at the same multiple to cash flow as the low-risk form of the company? I have yet to see anyone's price target of SSW discuss this. Perhaps this is because in the real world there is no such thing as risk-adjusted fair value? Maybe it's all in my head?

If you look at the level of debt in the private sector there should be more de-leveraging to go (at least for a few more years). If the government goes for austerity then there could be a wave of deflation. Wouldn't you agree? Anything is possible. However I'm wondering about that idea of looking at debt in the private sector. If I have $300,000 of mortgage debt and my house is underwater by $100,000... and I walk away from the house, isn't that delevering? I can still go out and spend like I always did... nay, I can spend more because renting is cheaper.. And supposing the MBS that carries my mortgage is already discounted in the market place by that $100,000... then what happens to the economy? I walk away and the value of the MBS doesn't change, but now I spend more? I realize that this doesn't explain the entire picture, I'm just wondering if we've already deleveraged at least a good portion of that consumer debt... this due to so many of them trading in the marketplace in these bundled securities, and at least substantially discounted by the market.

Most businesses are valued far in excess of liquidation value... so I'm still confused at that example. Alright, suppose I start a holding company that raises cash at 6% to buy SSW preferred stock which is yielding 9%. What multiple to book value should it trade at? To answer your question, most businesses trade at premiums to liquidation value because there is something intangible of value there -- for example, MSFT makes ridiculously high returns on equity without using any debt. That "something" has value. But the ship and the service of providing crews are the only value -- there is no other intangible. They merely take that value and lever the crap out of it.

The moment that happens they'll just rent the unit that is being sold. These rents are far lower than the mortgage they were trying to pay. Surprises me that people aren't noticing that foreclosures simply mean people with extra money to spend on consumption. Except for the squatters that got 18+ months of sitting in the home without making payments of course. But generally speaking, the rents are nowhere near the 2006 level of mortgage payments. It seems to me like the prior rent level must have been artificially low, due to the oversupply of housing during the construction boom (people who should have remained renters bought new homes, artificially depressing rents). Now the opposite situation over the next few years if construction remains depressed -- a housing shorting, artificially boosting rents. Of course, people argued that over supply was the main impetus behind the housing collapse, but refuse to accept that a supply shortage is looming. They want to instead point out that Japan house prices just kept going lower, despite their population of home dwellers doing the same (which doesn't seem to matter to these bears, for whatever reason supply/demand isn't in vogue).

I'm not sure which part of my post you're referring to here - the comparison with MSFT ? I'm not sure about REITs. It's not my specialty, which is why I don't participate on those threads. I see the analogy, but I think commenting will only lead to confusion. SSW seems to be valued on a few different metrics - #1 dividend, #2 cash flow, #3 assets - debt. I break out 3 because book value or intrinsic value don't really describe it. I think a lot of SSW's high valuation in '07 was due to the replacement cost of their container ships. Well, they're not as valuable now because the yards are building them more cheaply, so really #1 and #2 are what drive the price today. Yes, I think that if you didn't use leverage you would create the same cash flows, but you wouldn't create the same value for shareholders. Let's say fair value is 3,000 based on 10x DCF. If you go the pure equity route, you save on the interest costs - 6% * 2,500 debt = 150, so DCF = 450, but you have to put up more on the equity side. So instead of 100mm shares we'd have 240mm shares (assumes 40/60 equity/debt split). Price per share = 4500 / 240 = 18.75. Same numbers with the current debt equity split is 300 / 30 = 30 per share, as you pointed out. The equity picture is actually worse because there's a tax shielding effect associated with debt cost (i.e. when you finance with debt, you don't pay tax for that finance cost. You will pay tax on the additional profits associated with the 100% equity approach, so DCF is closer to 400 but whatever the point is made.). This is a simple model, but I think it fairly illustrates why debt it used the way that it is. If I were the Washington family, I'd probably prefer a 100% debt capital structure, but clearly that won't fly because banks won't lend that kinda dough at those rates without having equity participants around to absorb the blow if things go sour. I'm not criticizing their choice of using debt, which seems to be what you are justifying. I agree with you it seems like a good risk-adjusted idea. I'm criticizing the analysis that tries to put a multiple on their distributable cash flow without taking into account the level of leverage. It just doesn't make sense, because one winds up with situations where one is expecting somebody to buy the shares at a price far in excess to liquidation value. Another version of the company with the same assets but without using leverage is also going to sell for the same multiple to cash flow, yet at 1x liquidation value? EDIT: The person buying the highly levered version of the company ought to see outsized returns for the extra risk, but that's not going to happen if he buys it for the same multiple to cash flow as the no leverage version.

Look what they are saying about HPQ: http://stocks.investopedia.com/stock-analysis/2011/Worries-About-Tomorrow-Sink-Hewlett-Packard-Today-HPQ-IBM-DELL-CSCO-EMC0519.aspx?partner=YahooSA Even if the company can't grow much more than 1%, it looks no worse than fairly priced Fairly priced at $36 with non-GAAP EPS of $5 given a scenario of 1% growth? Since when did fair price for the components of the DOW imply earnings yield of 13.9% under the expectation of growth instead of decline?

Myth, I won't argue with your thesis or your position on cap gains vs. your age, but I do want to look at the cap gains numbers. Assuming the 5% yield sticks and the $1.50 is right, then in 2013 we're looking at $31.50/share incl. dividends. Which is 85% in less than two years - about 36% per year. Maybe I'm dead wrong on the div and the gain, but I think I'm in the 80% right zone, which still puts me at 28% per annum. Maybe that's below the expected return in this market, but I'll take what I can reasonably assess as probable. I'll note that during feast and famine, Seaspan's operations and customers have performed exactly as expected. Their financing got screwed, though, so they're not impervious to flaw :D PS. This is one of the things that keeps me from investing in a company like MSFT.. 85% value creation for them means that they need to create about $160bn in present value gains. It just seems like such an improbably large number. Maybe I need more imagination or something, but I see SSW creating $1bn more quickly than MSFT creating $160bn. There's somewhat of a difference here. SSW gets it's earnings by borrowing money to buy assets. I'm not sure you can value it strictly on cash flows. I think it needs to be valued on it's assets to some degree + some premium for the additional cash flows they get for providing crews and management services. I bring this up because I struggled to justify to myself whether or not the price is being influenced by rational people right now, and whether I would be correct in my assumption that a higher payout would lead to the $25+ price I was expecting. In other words... if a commercial REIT buys up a lot of properties at 14% yields, using 6% interest loans and 25% down. How much of a premium to book value should it trade at? Similarly, another REIT buys up properties at 14% yields and uses cash only. The former has more intrinsic value than the latter, assuming all ends well. Do they both trade at intrinsic value? Should they both trade at the same multiple to liquidation value, with the former generating better returns? Do these things get valued only on cash flow, or is risk-adjustment important? I think at the expectations for SSW share prices in the high $20s for example, that's roughly 10x distributable 2012 cash flow. Couldn't you get the same return by investing in ships WITHOUT using any leverage? So in what way is that a rational means of valuing SSW? Who here thinks a highly leveraged company should trade on the same multiple to distributable cash flow as a completely debt free version of the same assets?

You are ignoring the value of the cash flow. I guess a utility yielding 14% is not worth purchasing if the price gains are merely nominal? I mean, that's really the issue here. You are asserting that real estate is poor investment but not discussing the fundamentals. Just speculating on price movement.

Even the 1920s were deflationary: ftp://ftp.bls.gov/pub/special.requests/cpi/cpiai.txt I came up with a net 1.17% decline in the CPI using the data from 1920 to 1929. EDIT: sorry about that link... something is forcing the *http* onto the front of an *ftp* URL.

1930s were deflationary. ftp://ftp.bls.gov/pub/special.requests/cpi/cpiai.txt We start out with deflation even BEFORE the 1929 crash -- so it isn't like a huge new trend that suddenly happened with the crash, rather it was a magnification of that trend: 1927 -1.7% CPI 1928 -1.7% CPI 1929 0% CPI Deflation accelerates with widespread panic -- no doubt the cascading bank failures had something to do with this (no FDIC protection either) 1930 -2.3% CPI 1931 -9.0% CPI 1932 -9.9% CPI 1933 -5.1% CPI Then, after that period is over, what happens for the rest of the decade? 1934 +3.1% CPI 1935 +2.2% CPI 1936 +1.5% CPI 1937 +3.6% CPI 1938 -2.1% CPI 1939 -1.4% CPI 1940 +0.7% CPI Is it presently the beginning of 1930 and we're about to see 9+% deflation over the next two years? Or is it the end of 1933 and we're going to see inflation for the next 4 years, followed by two mild years of deflation and then followed by another decade of inflation?

Yes, however I can't find any MSFT puts expiring in the 1930s. Here is a bit of background though -- we can wonder/speculate if a repeat of such a banking crisis is likely today: http://www.fdic.gov/bank/analytical/firstfifty/chapter3.html

His price target is DOW 1,000 -- in other words, the DOW will fall 92%. So let's just look at some DOW components: MSFT: will fall to $16b market cap? Wow, net-net. Who would have imagined that? This guy is truly visionary.

It's the Windows laptop market that is getting hurt the most. Things that will reinvigorate sales are new models incorporating solid state drives and power sipping processors. Trouble is, I hunted around for a model with a solid state drive and could only come up with Samsung's Series 9. SSDs are still quite expensive relatively, but that should come down rapidly. SSDs fix many of the Windows complaints, such as boot time and resume time. When "asleep", the power is actually completely off, so you can swap out the battery for example. Then it takes only three seconds to resume. In other words, you would very infrequently go through a full boot sequence, which is 60% faster anyhow. Airplane travel gets better when sleep is full power off.

Perhaps I over did it a bit. Well, good I hope we make more money. There must be a lot of profit in those ships. That's a huge price reduction.