Saturday, November 28, 2009

Getting it wrong over coffee

Money management can be one of the most interesting careers in the world. At best it gives you lots of unstructured time to think about how the world really operates – and to make bets based on your hypotheses.  You get to conduct uncontrolled but real time experiments in the social sciences.

But never forget this is social science – not physics – and a little dogmatism about your rules or positions can result in getting it spectacularly wrong.

This was one of the more interesting places I have got it wrong lately – and so I thought I would write it up. It was also a surprising case of getting it wrong – because despite the analysis being totally stuffed I managed to scrape a (small) profit out of the trade.

The case involved three coffee companies (Diedrich, Green Mountain and Peet’s Coffee). I will explain what these companies do later – and how they are involved – but first I want to digress a little on the oddity of American coffee chains.

Starbucks failure in Australia

As an Australian the success of Starbucks (or Peet’s in San Francisco) puzzles me. Australia is blessed with a plethora of interesting and sometimes quirky, often very stylish cafes.  For those who are interested here is a photo of my local (Bronte) strip of coffee shops.  They are better than any strip I know of in New York, SFO or Chicago.  The coffee is better too – and that is widely commented on by visitors to our shores. 

It’s not that Australia is even a particularly coffee addicted country.  Per capital consumption is not huge by developed country standards and is lower than the United States.  We just have a better coffee scene.

It puzzles me why.  There are awful lot of things that the United States does (substantially) better than Australia so this is an oddity deserving some explanation. 

American coffee chains (particularly Starbucks but also some of the donut variety) have tried to break in and mostly failed.  Starbucks closed most of its stores last year after multi-million dollar losses.  They couldn’t cope with the Australian competition.  Which is odd because in most things the US is a far more competitive market than Australia – and US senior management tend to be more battle hardened than their Australian peers (see my piece on the use of American CEOs in the Australian context).  Its just the competition in cafes is far more fierce here.

I would love to be corrected – but I think the reason has to do with our wage structure.  American low-end wages are very low indeed whereas Australia has minimum wages at quite high levels.  Hiring unskilled labour to run a coffee shop according to a formula (and devoid of in-store entrepreneurial talent) works in America but does not work in Australia.  Also entrepreneurial talent in America has too many opportunities to waste itself in a coffee shop – whereas small-time competent entrepreneurs will open a small coffee shop in Australia.  [Maybe one of the good things about America is that it uses entrepreneurial talent well.]

Anyway this leads me to believe that some businesses that look bullet proof (running simple chains of shops which sell an addictive and brand loyal product) might actually be more vulnerable to quite strange social and economic trends than you would think.  Coffee shops look impregnable (and Peet’s typically trades with a 30 times price earnings ratio) but coffee shops can be beaten back by small entrepreneurial talent for reasons that are hard to articulate. 

The Green Mountain Keurig Cup

Another thing that coffee shops are vulnerable to is easy-to-make-at-home quality coffee.  Nescafe and other instant brands distribute a large amount of caffeine – as for that matter does the Coca-Cola company.  But against a well made espresso (diluted with frothy milk to taste) it just can’t cut it.  Our funds management business requires its morning coffee (which we buy from an excellent and entrepreneurial local cafe).  That said – American coffee is just not that special – and perhaps is more vulnerable than you think. 

One thing it is vulnerable to is the Keurig Cup (or K-cup).  K-cups are a plastic and coffee device you put in a special espresso machine and it makes you – instantly and with minimum mess and fuss – a very good espresso with very few traps for the unwary.  Its certainly a better drink than instant – and matches in quality a better cafe (though there are some tricks with milk frothing that the k-cup does not match).  

K-cups are a truly amazing business.  Once you have sold the machine you get to sell the cups (an addictive product no less) ad-infinitum with astonishingly fat margins.  And you don’t need to rent expensive real estate to do it.  This is like cigarettes – but with a growing market and without the litigation (and without actually killing your customers).

The first time I saw a k-cup was when a pretty young woman in a department store offered me a free coffee (brewed in a k-cup).  I noticed the razor-and-blades business model and determined that I did not want to be a sucker to that machine.  Alas – and to my endless shame as a stock picker – I did not even consider buying shares in the company which owned the k-cup business.  After all as a stock-picker you would really want to be on the receiving end of a razor-and-blade business selling a new addictive product.

That company is Green Mountain Coffee Roasters – and – since I first saw the machine the stock is up over five thousand percent.  A five thousand percent gain would – of course – somewhat have improved my finances and the finances of my clients...

The attraction of the k-cup business is not unnoticed by the market – Green Mountain trades at a PE ratio around fifty.  Still that is less than one times earnings on the price it was when I first saw a k-cup.

The k-cup license holders and Diedrich Coffee

Four companies have licenses to produce and sell k-cups.  Only one of those companies (Diedrich) is listed.  A license to sell k-cups is a license to share to some extent in Green Mountain’s very rapid growth however the terms with which you share are unknown.  As a default position I would expect that Green Mountain – the technology holder – would extract its pound-of-flesh for granting such a license and whilst a license holder might grow very fast you would not expect it to be outrageously profitable.  However – as I said the terms of the license are unknown and have never been published. If Green Mountain – in the infancy of the k-cup business – gave out the licenses on stupid terms then owning one of the four licenses to produce k-cups would be getting most of the benefit of being Green Mountain – but without even the expense of subsidizing the espresso machines or of hiring the pretty-young-woman sales people to sell new espresso machines.

It seemed however unlikely to me that Green Mountain – who have a truly fantastic product – would ever have signed a contract with Diedrich – or anyone else for that matter – which was truly unfavourable.  After all Green Mountain held the aces.

Moreover Diedrich has truly strange accounts.  It was an unsuccessful owner of a coffee shop chain (Gloria Jeans) which they sold off and the balance sheet has about 50 million in accumulated losses.  Sales were flying – but half those sales were back to Green Mountain (for whom they out manufactured product).  Moreover the accounts were truly strange in other ways (it was for instance very difficult to get a handle on tax expense).  There was a truly frightening article that appeared on Seeking Alpha which went through the accounts in great detail – and – if you took all the suggestions in that article at face value – then you would probably conclude that Diedrich was a fraud.  (The Seeking Alpha article has since been removed.)  

None of that stoped Diedrich being a truly explosive stock – indeed before I had even looked at the stock (and in about six months) it has gone from 21c to just over $30.  You don’t need too many of them in your lifetime…

Now this move looked truly bizarre given (a) the fact that it did not even own the k-cup technology and (b) the bizarre accounting as outlined in the Seeking Alpha article. 

I thought (possibly incorrectly) that Diedrich was probably a stock-promote rather than a business as per the suggestion in the Seeking Alpha article – but I did not bother to do any of the work to prove or disprove that case.  That work came later.

The Seeking Alpha article was anonymous and hard to verify (as discussed below).  

The danger of shorting frauds

The reason I did not do any of the work to confirm or deny the Seeking Alpha article was that shorting frauds (especially well publicised frauds) is possibly the most dangerous thing you can do on Wall Street.  Take for example a fraudulent oil company in Africa – but one that really has some oil.  The company has oil – so it can show flows to visiting analysts.  It has pictures and local politicians are excited to hob-nob with the management.  But there is no real way of telling whether they have 1 million barrels or 10 million barrels.  They may have 3 million barrels (which would be valuable – but not earth shattering) but tell the market they have 20 million barrels.  There really are very few ways to tell – and if they are determined they can exaggerate at will.  Moreover suppose you have worked it out (or at least have suspicions) – it doesn’t help you.  If someone is going to fake the existence of 17 million barrels of oil there is not much that stops them from faking the existence of 170 million barrels of oil.  If you are short and they “announced” a 100 million barrels (or a few trillion cubic feet of gas) and the market believes them then you are stuffed.  The stock could go for a monstrous run – and you will be forced to cover at a shocking loss. 

The problem with shorting frauds is that there is nothing to keep the fraud grounded – and hence there is nothing to limit your losses.  If you short a real company (say Dell) and Dell sells less computers than it anticipated it will tell the market.  The stock will probably go down.  If it sells more computers than anticipated at better margins the stock will go up.  But it is vanishingly unlikely to sell ten times as many computers as anticipated.  You can be short and whilst you might lose money if you are wrong you will not do serious damage.  Frauds – because there is nothing to keep the claims grounded – can cost you an enormous amount as a short.

There was a possibility that Diedrich was a fraud – but it was a high-growth with a substantial short interest.  There was simply no way that I was going to short it because the potential losses looked vast.  Given that I did not bother to fact-check the Seeking Alpha article.

The rare safe time to short a fraud

There is only one time that it is relatively safe to short a fraud – and that is when a real company is suckered into buying it.  That happens – and the cannon-example was when Mattel purchased The Learning Company.

The Learning Company (TLC) was (and remains) a children’s educational game software maker.  It has real games and real sales.  In that sense it was similar to my African oil company with a few million barrels.  It however had accounts which were peculiar – and several short sellers were convinced it was mostly fraudulent.  [I had never heard of the company so I cannot vouch for their views.] 

Eventually Mattel purchased TLC.  The purchase was made from weakness.  Mattel used to have a great franchise in toys for young boys (think Matchbox cars).  There is simply no way I could interest my nine year old in Matchbox cars anymore – not when they have a Playstation or a Wii.  Boys toys were dead and computer games killed them. 

Mattel was desperate to join the computer game revolution – and it paid a fortune for what really was a dodgy property.  The losses were in the billions.  Wikipedia tells the story quite well.  It was one of the worst acquisitions ever on Wall Street – and not only did it cost Mattel over 3.5 billion (mostly in stock) but it revealed the underlying (and fundamentally incurable) weakness in Mattel’s business.   It also cost the CEO (Jill Barad) her job.  Jill made her career promoting and marketing Barbie dolls and her claim to fame was that she was brilliant – possibly amongst the best ever – at marketing bimbos.  (A company insider once noted to me that that included herself.)  Jill was however absolutely useless at assessing a computer game company.  [Personally though I think she should have been kept on as CEO but limited to Barbie…  She really did deserve that job and did it well.] 

A short seller’s dream is when a real company – preferably one which is a little dopey and in an old industry (matchbox cars as compared to computer games) buys something hot, sexy and fundamentally dodgy. 

Peet’s bids for Diedrich

I thought I had my own repeat of Mattel and The Learning Company.  Peet’s – a well run but simple coffee chain around the Bay Area bid cash (debt funded) and stock to buy Diedrich.  Ah – here it was – a simple company with a simple model – whose margins seemed to be declining (for reasons I have not fully identified) which bids for a well promoted company with massively complex (and potentially misleading) accounts. 

To say I was excited is understating it.  This is precisely the sort of thing that excites a short seller.  At worst what would happen was that Peet’s – a company with less than 200 coffee shops (and owning only the brand and fittings) would have roughly 140 million debt and a business (Diedrich) that – depending on the contract with Green Mountain – might not be worth very much at all. 

In the bad-case Peet’s stock – trading at 30 times earnings – could lose 80 percent of its value.  [That was what happened to Mattel.]

So I did some work.  I went back to the Seeking Alpha article and tried to verify every claim in it.  Alas I could not.  There were a few simple (and innocent) explanations for some of the red-flags highlighted in the Seeking Alpha article.  However with respect to some of the claims in the article I had to acknowledge that the author had a point.  He wasn’t right with respect to everything – but he was right on some points.  Diedrich’s accounts were (and still are) pretty aggressively stated.

I shorted some Peet’s.  Not a lot – but maybe a third of the final position I hoped to have.  I intended to do a little more work (especially as the debt covenants were published).  Besides – in the Mattel case there were many quarters as The Learning Company disaster unfolded.  There were plenty of opportunities for a short seller in Mattel to test their thesis on the way down.  This was a small position – but as it unfolded I hoped to make it a big position. 

Now the observant will notice there is a missing detail.  I still do not know the terms of the contract between Green Mountain and Diedrich.  If the terms are highly favourable to Diedrich then shorting Diedrich (or the new owner of Diedrich) is spectacularly misguided.  In that case Diedrich is just a cheap way into Green Mountain’s (fantastic) business.

Not knowing the terms of that contract was the key weakness in all of my analysis.  And I knew it. 

Peet’s had a conference call when they announced the purchase.  They indicated that the terms of the contract were acceptable but they refused to detail what they were.  Paraphrasing Mandy Rice Davies “well they would say that wouldn’t they”. 

What I was looking for in future quarters (as a guide to increasing the position) was evidence that the Green Mountain contract was favourable to Green Mountain. 

Well I was wrong

I started this by stating that most of the analysis was wrong.  And it was – in a very specific way.  My biggest concern was always the contract Green Mountain had with Diedrich.  And only one party outside Diedrich knows that contract and understands its economics intimately.  That is Green Mountain. 

And they proved me wrong.  Green Mountain overbid Peet’s for Diedrich - $30 per share – all cash. 

Oops – now I know I am wrong.  Green Mountain wants to bring that contract back in house.  They want it in-house for about a quarter of a billion dollars in cash!  Remember ultimately all that Diedrich has is a few tax losses, a small manufacturing facility for K-cups, a couple of second-rate coffee brands and THAT CONTRACT. 

Peet’s overbid – bidding $32 a share ($20 in cash the rest in stock).  Green Mountain upped its bid to $32 per share ($265 million) all cash.  Remember this stock was trading earlier this year under 25 cents!  This is one hell-of-a-valuable contract.

Needless to say events have shown I was wrong about Diedrich with respect to the only thing that mattered (the worth of the k-cup contract with Green Mountain).  I covered my short and thanked luck that it was not (much) worse.

Of course this bidding war drove down the value of Peet’s stock.  So I was wrong and made a small profit.

Money management – lucky or smart?

Given I was wrong I can hardly say I was smart.  But Peet’s was not a bad candidate for a short.  It has a very thin balance sheet (almost all profits have been used to buy back stock).  It does not own the sites of its stores.  It was paying cash so it was going to incur debt with only intangible assets (its brand really) to back that debt.  And it had a 30 PE. 

Given all of that I could be wrong in lots of ways but still make a profit. 

But I should not get carried away.  I was wrong – proven wrong – and the profit is nice but it shouldn’t be used to ignore the fact that my hypothesis was simply inconsistent with observation.

Back to scratching around (often fruitlessly) for things that might make money.




Wednesday, November 25, 2009

The Ides of March and the Fed exit strategy

The dollar bears (and inflation mongers) are all over the blogosphere and press at the moment. It is almost a consensus that the US dollar is doomed and that gold is a great investment. Sure there are doubters – but few are as vocal as Paul Krugman who refers to inflation as the phantom menace.

Anyway I thought I would have a brief analytical foray.

Over this crisis the Federal Reserve increased the money supply by literally trillions of dollars and several hundred percent. In an ordinary world this would have caused very rapid inflation – but suffice to say that this is still not an ordinary world. Short term interest rates are stuck at zero and medium term rates are pinned very close to zero. People (and financial institutions) are willing to hold (and not spend) inordinate amounts of money. Liquidity preference is very high and it has absorbed all the extra money the Fed has printed.

Obviously some day liquidity preference will wane (though Japan tells us that day might be very far into the future). When the liquidity preference has waned all that money that was printed will be excess and might turn into inflation.  The Fed will need an exit strategy. Inflation hawks are already worried that inflation is inevitable. Gold is a favoured investment of many – including some hedge fund managers who I think have very fine minds.

Ok – I think the inflation hawks are wrong and I think gold will be a lousy investment. But I need to explain why.

The Fed printed money to buy lots of riskier assets. Often these were assets owned by banks and the banks borrowed the cash from the Fed secured by these assets. About a trillion of the assets were qualifying mortgages guaranteed by Fannie and Freddie. Whatever, there is a huge increase in money supply (liabilities of the Federal Reserve) offset by an equally huge increase in assets held by the Federal Reserve.

The Fed has an exit strategy – a natural exit strategy. When people’s liquidity preference wanes they will want to hold risk-assets rather than cash. And the Fed owns trillions of dollars of risk assets. The exit strategy is simply to sell those risk assets and take back (and destroy) the cash that they created during the ciris.

The right speed to do this is also – in some sense – naturally determined. The right speed is at the speed the general public wants to hold mortgages and other risk assets again.

Now all of this presupposes something utterly critical. It presupposes that the assets on the Federal Reserve balance sheet (loans to banks secured by risk assets, Fannie and Freddie mortgages and paper) are good. If the assets held by the Federal Reserve are worthless then the Fed cannot use those assets to buy back the excess money supply precisely because those assets were a wipe-out.

The Fed has an exit strategy – but only if the asset side of its balance sheet is solid.

And this brings us back to the middle of March when the system was collapsing around us. The Fed was the only source of liquidity – they were lending to big banks secured by almost anything the big banks had unencumbered. They even lent to big banks secured by (of all things) recreational boat loans.

If everyone was insolvent then those loans to banks will be no good. The Fed will have no capacity to withdraw the excessive money supply because the asset side of their balance sheet will be stuffed.

However if the crisis was a liquidity crisis and not a solvency crisis then, come the time to exit quantitative easing, the Fed will have a sufficient balance sheet to do its part.

The determining factor as to whether the end game here is inflation is whether March was a liquidity or solvency crisis.

This blog has been consistent. I do not believe the US banking system was insolvent in March. I never did believe it. And I thus believe the Fed has an exit strategy. Given that belief I think that gold is a lousy investment from here (though thankfully I have not been short it).

More generally I think there are serious problems with believing both of the following:

(a) the financial crisis was essentially a solvency crisis – the banks were mass insolvent in March and a bailout was inevitably going to impose very large long term costs and

(b) that inflation is not a large risk.

However Paul Krugman seems to believe both these things, and he has a Nobel Prize in economics and is clearly smarter than me. So please take all of this with a grain of salt.

Sunday, November 15, 2009

The missing details: Bronte Beach edition

Saturday afternoon and I had volunteer lifesaving duty.  My (broken) collarbone is knitted enough to be able to go in for a swim in modest surf – but if there were a difficult rescue I would pass the duty onto someone else.  Really I am a pair of eyes – the job is to watch and assess – not to make a hero of myself.

I was sitting chatting with Rod, a fellow lifesaver at the North end of the beach watching quite a large crowd and getting modestly annoyed when the (fibreglass) board riders were sailing too close (or into) the flagged bathing area.  (Swimmer’s heads tend to come off badly when hit with a fibreglass surf board.) 

At the very south end of the beach is a rip (a current that goes out to sea) and some lifesavers were standing around chatting around the rip.  This is the same rip where the Muslim men were rescued last November

There was someone swimming in the rip – with quite good – even stylish strokes.  But he was getting nowhere.  Rod and I were debating whether he was even likely to get into trouble.  The stroke was – as I said – strong – but given the current what he was doing was futile.  We watched for about a minute when I decided to walk down the other end of the beach and see what the other lifesavers wanted to do about it.  I was not worried.

As I walked the guy stopped swimming – just gave up – and started to drift out to sea at about 1.5 metres (5 feet) per second.  I got to the lifesavers about the time I thought it was actually going to be necessary to go in and get the guy – but the professional lifeguard on the beach had run down, got a rescue board and was already on his way to effect the rescue.  These are the same lifeguards from the TV series

The victim was still treading water, the surf was not rough – and I suspect if he knew what he was doing (that is knew to swim across the current) he could have rescued himself.  But I was still a little peeved at myself for missing the easiest of board rescues (and the kudos/self congratulations that would go along with it).

Ex post we realised there were a few missing details:

First – the lifesavers at the South end of the beach simply did not notice the guy caught in the rip.  Maybe they noticed his fine swimming stroke and assumed he was not a “customer”.  Maybe they were looking at pretty women in bikinis.  Maybe they were just preoccupied.  Whatever – they did not see.

Second – the customer was from Bavaria.  He was a tourist.  He had once swum competitively (hence the stylish swimming stroke) but he had never swum in the surf.  He simply did not understand his predicament and he had no idea how to get out of it.

Third – the customer was wearing cut-off cotton jeans – not a nylon swimming costume.  That makes it just so much harder – and an amazing proportion of our rescues are of people who go in fully or partially clothed.  [The fully clothed are often Muslims.]

Fourth – the customer had had a couple of beers.

If I had known these four details I would not have walked to the other end of the beach – I would have run as fast as I could.  Those details – none of which were readily apparent – changes the interpretation of the guy in a rip from “interesting and slightly comic” to “life-and-death”. 

The existence of a problem was obvious to me – and I (incorrectly) presumed that it was similarly obvious to my fellow lifesavers.  I just assumed because I had noticed everyone had noticed – and hence I acted almost apathetically to the danger.  Moreover I assumed away my four missing details because the customer had a fine swimming stroke which created an illusion that all was under control. 

That is I suspect a very human mistake…



Saturday, November 7, 2009

The media market has a conservative bias

Here is a sequence of numbers to bring tears of joy to a stockholder and tears of rage to a liberal pundit.* 


















It’s the quarterly operating profit of the cable network programming for News Corp in millions of dollars as reported since September 2005.  It’s not all Fox News – but Fox News is the main driver.

I love reading Talking Points Memo, the Daily Kos, Paul Krugman and Brad Delong – but its quite clear that the mass audience and the dollars are elsewhere. 

And whilst there are some nice new liberal media sites (including many I read) and I think people like Josh Marshall have reinvented part of American journalism that is all a delusion.  The media market has a conservative bias. 

Just to make the point further I have met a few media barons – including briefly the Sun King himself.  My impression of media barons is that whilst they have political views (often quite strong ones) there real bias is to things that are profitable.  Rupert is in my home town this week (Sydney) and he is personally expressing views associated with asylum seekers in Australia that are associated with the left of Australian politics.  They are not views expressed in his local newspapers

Therein is the rub.  He is quite happy to have his newspapers express views contrary to his own when it sells papers.  The media market determines media bias – and – as the above string of numbers show – the media market has a conservative bias and that bias is getting stronger.  Media bias follows the money-making bias of media owners.  People who proclaim liberal media bias are just not following the dollars.

I hope – sincerely hope – that Josh Marshall, Markos Moulitsas and others of the new media liberal elite can make a go of it.  But the conservative side generates operating profits of half a billion per quarter and that gives them a longevity (and power) that the new media – for all its obvious intelligence – can only watch in gob-smacked wonder. 





*In this case I am both a liberal pundit and a stockholder.  I don’t know whether to cry or to cry. 

Friday, November 6, 2009

Fannie Mae’s results – oh, and what if Bank of America reported the same way…

There have been some mathematical corrections to this post discussed in the comments. My pencil notes had the numbers right. By the time I got to writing it out errors had entered. Sorry.

Fannie Mae just put out awful looking results based primarily on massive (and increasing) credit loss provisions. Indeed their provisions this quarter were the largest thus far in the cycle.

Its worth looking a little closer because – like it or not – all Americans are owners of Fannie – both the downside (their current book) and the upside (if any) through taxpayer ownership of the common stock.

The nature of credit loss provisions

Each quarter almost every financial institution takes some charges when loans they have made settle at less than 100c in the dollar. At the moment charge-offs are at historic highs.

Every quarter a company makes an estimate of future losses – a “provision” if you will.

Provisions by definition are estimates – whereas charge-offs are real and mostly final.

The difference between provisions and charge-offs goes to a “reserve for future losses” or more commonly just “reserves”.

Most financial institutions are taking more provisions than charge-offs – in other words they are building reserves. This is necessary because there are a lot of delinquencies and a lot of loans in the foreclosure process and – just frankly – a lot of loans that common sense tells you will end in charge-off.

Most institutions build reserves relatively slowly. Bank of America for instance – in broad numbers – has had 13 billion of provisions per quarter for the last three quarters and charge-offs of 6,8 and 9 billion respectively. If the charge-offs skyrocket (say to 20 billion) at bank of America then it will find itself under-reserved – and will wind up having to report very big losses. However if charge-offs slowly level off around 13 billion per quarter then BofA will – ex-post – look OK.

The honest answer in the case of BofA is that we really do not know where charge-offs will wind up but we can make educated guesses. In the last conference call BofA thought charge-offs would peak about the first quarter of 2010. If they are right then their current reserving is right and BofA is probably a steal as a stock right now. If however charge-offs continue to rise for another 18 months peaking out at say $35 billion per quarter then BofA will need to be recapitalised further and may wind up as government property.

I am inclined to think that BofA’s current educated guess (charge-offs peaking early next year) is a little optimistic – but not very optimistic and I am happily long Bank of America common shares. This is – as I stated – an educated guess. Other people I respect have different educated guesses. The (very smart) Chris Whalen has a completely different view arguing (amongst other things) that the liabilities for fraudulently sold securitisations at Countrywide and Merrill will produce losses large enough to render BofA insolvent. I think he is spectacularly wrong – but difference of opinion makes a market.

In BofA’s case 13 billion per quarter is sort of a magic number because it happens to approximate the pre-tax, pre-provision profitability of the bank. Provided actual end charge-offs remain around or below 13 billion per quarter BofA will be able to earn its way of its mess. If charge-offs go to 25 billion per quarter they can’t earn their way out – and hence just the implicit government guarantee they currently have will not be enough to save them.

I note that current charge-offs are comfortably within the 13 billion per quarter so all is well for the moment. As to the future – all we can take are educated guesses. And that is all bank provisioning is. In BofA’s case the 13 billion (plus or minus a couple) of provisions taken each quarter seems a little optimistic to me – and you can understand why when the gun is pointed at the executives head they manage to (miraculously) pick their provisions to roughly match their pre-tax, pre-provision profit. But as I have noted I think the provisions in BofA’s case are only slightly optimistic – and the end charge-offs won’t go very far above 13 billion per quarter.

Analysing the Fannie Mae result in this light

Fannie Mae – as stated - took an enormous loss this quarter. The key to this loss was a credit charge of $22 billion. This credit charge can be broken into two broad categories – which are (a) the actual charge-offs taken, (b) the addition to reserves.

Like BofA, Fannie (and Freddie and just about everyone esle) needs large reserves because – frankly losses and delinquency are still getting worse. The amount you need to add to reserves is an estimate. If your reserves are large enough (which doesn’t seem to be the case in any financial institution I look at outside the GSEs) then you don’t need to add and you might even be able to run the reserves down a little.

In Fannie’s case this quarter there is one more thing complicating the reserves versus charge-offs picture. Fannie changed the way it accounts for one of its loan modification programs (the “Home Affordable Modification Program” or the HAMP) such that when loans are acquired from securitisation trusts for modification they are written down to market. This loss (which Fannie calls a “loss on acquisition”) is not a final loss (as per a normal charge-off) but rather an estimate of the future charge-offs they would take on those loans.

So lets break up the credit charge.

The provision for credit charges was 21.96 billion – which I will round to 22.0 billion – given that the nearest 100 million seems close enough. The charge offs were 10.9 billion (see table 10 in the 10Q). Note 3 to that table tells us that of that 10.9 billion 7.7 billion came from the “loss on acquisition” on the HAMP. The actual loans that were charged-off (final) were 3.2 billion. They were probably a bit higher because there were some HAMP charges taken last quarter and maybe some were finalised this quarter.

But nonetheless the way to think about this is that final losses this quarter were 2.2 billion. Provision for future losses (HAMP losses and provision build) were 19.8 billion. Similar ratios have applied every quarter since Fannie Mae went into conservatorship.

Now I am going to make the obvious point. Bank of America provides roughly 1.5 to 2 times its charge-offs each quarter. Fannie Mae provides 7 times (and has been closer to 10 times in past quarters).

If Bank of America were to provide at the same rate its quarterly losses would be 50-80 billion and it would be completely bereft of capital – it would be totally cactus. It would be – like Fannie Mae – a zombie government property.

What I think is going on…

I think what is going on here is a different standard for Bank of America. And for Wells Fargo. And for Citigroup. And for PNC and for every other major bank in America. There is also a different standard for Goldman Sachs. That standard is different to Fannie Mae. BofA (like everyone else) gets to choose its reserving ratios – and to be a little optimistic. Fannie Mae chooses ratios that are so-off-the-scale high that it is different.

Remember provision build is an estimate not a fact – and Fannie is estimating extraordinarily bearishly and Bank of America’s estimates are slightly generous. But regulators are controlling Fannie in such a way that keeps it down. They are allowing Bank of America to act as if all is well whilst Fannie Mae appears to be a complete zombie. Which I think corresponds roughly to the new policymaker consensus that what is good for big banks is good for America.

It is clear why BofA has chosen the 13 billion of provisions per quarter – which is that it roughly corresponds to their pre-tax pre-provision income. Moreover – in my view the 13 billion per quarter is not far wrong so the decision is defensible.

It is not clear why Fannie has chosen to reserve quite so aggressively. My guess is that there is no active conspiracy – but the pressure to make extraordinary provisions at Fannie is very high for a variety of non-commercial reasons. These provisions are defensible only if you believe the housing market gets substantially worse from here. That seems to belie the evidence on the ground – at least for now. Housing markets in the core bubble states have clearly stopped deteriorating. Current provisions (including mark to market provisions on the HAMP) are now 6 years current charge-offs. They are only 18 months or so at most banks including BofA.

Am I being too harsh?

Is it too harsh to apply the same provision to charge-off ratio to Bank of America as it is to apply it to Fannie Mae? Well if the credit was deteriorating faster at Fannie that BofA I would be too harsh. But if the credit were deteriorating faster at BofA then I would be too generous. The best test of that is non-performing loans.

At year end BofA non-performing loans were 18.2 billion. They were 31.9 billion by the end of the third quarter – a rise of 75 percent.

Fannie Mae NPLs were 111.8 billion at the end of the year (20.4 on balance sheet, 98.4 off balance sheet). They were 197.4 billion at the end of the third quarter – a rise of 76 percent.

75 percent versus 76 percent – I will call that a wash.

Indeed almost however I cut it the situation is getting worse for BofA at roughly the same rate as it is for Fannie Mae.

Except for one thing. The government wants BofA alive. Lots of people want Fannie Mae dead.

My views

Bank of America survives now but for the good grace of the quasi-government guarantee. So do all banks. But Bank of America is – in my view (a view open for dispute) ultimately solvent. Its provisions are optimistic – but not (in my view) excessively so. If the cash losses per quarter rise to (say) 30 billion dollars then BofA will die and will cost the taxpayers a lot of money. I think that is unlikely but it is not impossible. Provision additions are always just an educated guess – not a science.

If the same standard were applied to Fannie Mae as bank of America Fannie would still have needed government assistance. It started with less capital and more levered than BofA. But the position would not look anything like as bad as it does.

You can of course interpret this to suggest that the Fannie Mae standard should be applied to BofA – and indeed to the rest of the financial system. You would (in my educated guess) be wrong. But I would have little ground to dispute it.

Disclosure: Long preference shares of the GSEs, long Bank of America. Could be wrong about both.

Saturday, October 31, 2009

Zion sent their lawyers to get us. It is like being flogged with Jericho lettuce. I drop one on them. They can’t psychologically handle it.

Oh no! Zion sent their Lawyers to get us.

It's like being flogged with Jericho lettuce!

The Feral Fundamentalists have

Come to savage us!

They must be ravenous!


Meddling Mediocrity, from the Televangelist Aristocracy,

Rip off merchants from Hal Lindsey Ministries,

But Old Dozy knows when I've got 'em,

They fail to reply when I drop one on 'em.

It's somethin' they can't psychologically handle.

Them and their band of shareholder wealth vandals.*

Last week I had an exchange with Zion Oil and Gas’s lawyers.  Zion it seems objected to my characterisation of the Ma’anit-Rehoboth #2 Well as dry.  They accused me of deliberately misinforming the market and of stock manipulation.  They threatened to report me to regulators.

I asked whether the well did show hydrocarbon flows – and if so how much?  After all they have been up and down this well with equipment many times and if there were hydrocarbons they would detect gas in those trips (so-called “trip gas”).  Eventually they said through their lawyers that they had found hydrocarbons in this well.  (Note that their position appears to have changed since early this week – as this weeks drilling report denies the finding of hydrocarbons.)

The three letters that they sent me are reproduced here (1), (2) and (3). 

Zion are currently issuing shares under a rights issue.  Selectively informing me of a hydrocarbon find is of course an offense.  Not informing the market of such a find during a rights issue is similarly an offence.  Likewise would be failing to inform the market that the well was substantially dry.  Whatever, I agreed with them the regulators should be informed.  They had been keen to turn me in.  So I sent them this reply:

Dear David [Aboudi – Zion’s lawyer in Israel]

It is clear that you are intending to report me to the regulatory authorities for my blog post on Zion oil and gas.

I think we should proceed quickly.  I have copied this letter to Stephen J. Korotash---Associate Regional Director of the SEC office in Fort Worth in charge of enforcement.  This is the appropriate regional office with jurisdiction over Zion.  I have previously copied him all three of your emails to me and my blog post.

Could you suggest a time that is appropriate for a conference call?

I have not invited Tim Johnson who is the US Attorney for South Texas which has venue over Houston based issuers, however if you wish to include him his email is [withheld]

I look forward to our discussion.

Thanks in advance.


John Hempton

I have heard nothing more from them.

They drill for oil in the Promised Land.  I sit at the arse-end of the earth.  But good religious folk like them should know it is rude not to reply. 

I am waiting for the next warm Jericho lettuce flogging.




PS.  I send a draft of this post to company for comment.  They have since made much clearer statements as to the hydrocarbons in this well.  They are testing zones of porousity so they still have hope – but they note that:

As yet, no hydrocarbons have been ‘Produced to Surface’…

[However] with regard to our log analysis, an independent log analyst noted that the Ma’anit-Rehoboth #2 well does have a specified amount of potential “net pay”…

The analyst was careful to comment that the results of his analysis … should not be considered ‘quantitative’ due to the effects of borehole washouts on the input logging measurements used for his analysis.

He noted that the existence of any hydrocarbon-bearing, open-hole fracture porosity in the formations inferred from the effects of borehole washout on the conventional wireline log data analyzed was tenuous at best, as such reservoir properties are impossible to identify or quantify directly from conventional log data alone.

The analyst recommended testing the seven zones…

You will appreciate that, until such time as we recover hydrocarbons at the surface (or not), we are not able to give any estimates of what (if anything) we believe we may recover.

Given no hydrocarbons have been produced to the surface and the indications are tenuous at best I will now amend my original post to the well being probably dry.  Their legal threats demanding I withdraw the assertion the well was dry seem hollow.

Moreover their lawyer in a letter to me (and copied to the Zion’s CEO) said:

Our client has clearly indicated in its public filings relating to the Ma’anit-Rehoboth #2 well that the well logs indicate the presence of hydrocarbons in identified 'zones of interest'.

There is an inconsistency between Zion’s latest statement to the market and their lawyers statements to me.  That was a sustained exchange so the disclosure to me was not an accident.  However Zion’s comments during the rights period now appear to be appropriate – I think in no small measure due to this blog.


PPS.  Zion are Dallas (not Houston) based.  The right USDOJ official would be James Jacks.  That is good – he is probably more aggressive than Tim.


*Apologies to the former Australian Prime Minister and the master of insultMr Paul Keating – and Company B.  The Paul Keating original is about being flogged with “warm lettuce”.

Wednesday, October 28, 2009

The new GSE as zero meme – laying the assumptions bare – and a modest plan for Obama

There have been a few broker notes out suggesting that the GSE preferred stock is going to zero.  The preferred stock itself has been dreadful lately – retreating almost to our original purchase price. 

I think the broker notes are wrong – but lets do this formally because if you look at the assumptions in my model and the assumptions in the broker notes you can make up your own mind.  [I will lay out their assumptions and my assumptions clearly – you decide.]

The first “GSEs are zero” broker note was produced by Keefe Bruyette & Woods (one of the few brokers left covering the stocks).  I have reproduced the note here (and claim fair comment use for doing so). 

The core assumption is that the GSEs are closed – and that they are put into very rapid run off – and that they do not earn much money during this run off period.  Here is the revenue model for Freddie Mac.

(You will need to click all the tables in this note for details.)  



There are implicitly a lot of assumptions here. 

The first core assumption is that the net interest yield (after hedging costs) on the retained portfolio will be about 1 percent over the long run.  I agree.  In the bad-old-days Fannie Mae used to report about 120bps, Freddie Mac used to report about 80bps.  When they restated their results Fannie restated the results down and Freddie restated them up.  The right number was about half way between the Freddie and Fannie numbers – so 100bps is as good an estimate as any.

The second core assumption is that the short run hedged interest margin is also 1%.  This is flat wrong.  Fannie and Freddie are getting absolutely record interest spreads at the moment – absolutely shooting the lights out.  I detailed this here.  This model assumes that Freddie has net interest income of $8 billion this year – which is rather difficult because they are currently getting over $4 billion per quarter.  The high current net interest margin is a function of three things:

  • Firstly – and most obviously – the lack of competition in the mortgage market.  That is not going away in the short term – and it would be crazy to assume that net interest margins compress to 1 percent rapidly.

  • Secondly the Fed is being more than generous with the shape of the yield curve.  That is going to end – but possibly not that rapidly.

  • Thirdly – and this is important – there were several charge offs of derivatives which were used to hedge the net interest margin when the businesses went into conservatorship.  Those hedges are still there (but they have been written off up front rather than amortised over the life of the product).  As a result reported net interest margins will be higher in the short term. 

All up I would expect the net interest margin over the first two years to be maybe 12 billion dollars cumulative higher than this model.  Indeed as those numbers are currently being reported it is perverse to argue otherwise.

The third core assumption is that the company is put into massive and sudden runoff.  This is a political decision that – as far as I can tell – has not been made.  You can see this in the numbers because the owned portfolio (and for that matter the guaranteed portfolio) is assumed to drop 20 percent per year from now.  This is a far more aggressive assumption than the government is currently indicating for Fannie or Freddie.  Indeed last month Freddie’s owned portfolio actually rose a little.  Moreover no government official has so far indicated that Fannie or Freddie will have to get out of the guarantee business – and this model assumes that they must leave the guarantee business.

In my long series I made it clear that the value in the preference shares depended critically on the companies being allowed to stay in business – at least for a few years.  That is true of the value of almost every bank in America – in that the whole sector is dependent on the pre-tax, pre-provision profits from their current business to cover their credit losses.  If it were not for pre-tax, pre-provision profits even big (and sacred) companies like GE would not really be viable. 

If we just assume the portfolio remains flat for three years we can add another 10 billion to Freddie’s pre-tax, pre-provision profits.

Now it is possible that the Government might choose to put the GSEs into rapid run-off (and there are Wall Street firms who crave the interest rate hedging business and who would like it) but if the GSEs are put into rapid run off it would have profound (and negative) effects for the price of conventional mortgages and for any housing recovery.  I think it is reasonable to assume that they are not insane – so I think three extra years income is a reasonable assumption.  However again I am just exposing the assumptions.

The fourth issue is simple double counting.

Freddie in the KBW model is assumed to write no new business.  If it writes no new business it can incur no credit losses on that business.

However KBW assumes 5bps of credit losses and 5bps of credit-associated costs. 

They are going to have credit losses on the old business (but they count them below).  Counting additional credit losses is double counting.

It would (of course) be reasonable to assume credit losses would be incurred on new business – but KBW is asserting that there will be no new business.

The extra credit costs in the KBW model add up to another $6 billion over ten years. 

I think on reasonable assumptions – including a rapid run off of the book after three years the pre-tax earnings of Freddie are thus 28 billion higher than in the KBW note.  Nonetheless I am just reporting the assumptions implicit in the argument that Fannie and Freddie are permanently impaired.

Credit losses in the KBW note

There is no real model of credit losses for the existing book in the KBW note.  However they do give a chart with base case, stress case and best case.


Note the cumulative loss in the best case is $33.7 billion at Freddie Mac.  My model was a little worse than the best case - $37.6 billion of losses still to incur (so over 40 billion cumulative losses on the book).  Since I wrote that, there has been a solid bounce in the demand for houses around the $200-300 thousand dollar mark (that is largely GSE foreclosures) in the key bubble states.  This video from Jim the Realtor (who is usually a dour bear) explains just how strongly the San Diego area has bounced. 


It is pretty clear from stories like this (and there are many more) that it is much easier to clear inventory.  My model assumed that it was going to get much harder and that severity on the book would rise from the current 43 percent to about 50 percent. 

Now this bottom-end housing bounce could be “the mother of all head fakes” – but again – like the interest margin – I am only reporting what is happening now.  The housing market could take another big swan dive and then my model will be wrong.  That is the bet I spelt out in the long series. 

Anyway we should look at the current losses – and projected forward losses.  The last Freddie Mac quarterly credit supplement gave the credit losses, provisions and reserves by quarter. 



In the second quarter the cash losses on the Freddie guaranteed mortgage book were $1.907 billion.  Cumulative cash losses have been a bit over 5.8 billion dollars. 

My model assumes that the future losses will be roughly 6.4 times losses booked to date.  That is my estimate is consistent with the housing market getting dramatically worse.  The evidence for that is thin.

Not only is the anecdotal evidence (such as Jim the Realtor) pointing the other way, but foreclosures are up only 5 percent from summer to fall.  Housing bears treated that news as evidence of crisis – and I thought it was a remarkably good number.  The foreclosure moratoriums have expired and we are not swamped by foreclosures.  My estimate that cash losses on the existing book are likely to be about 6.4 times the so-far-recorded losses does not seem low.

That said – the base case in the KBW note have cash losses being 10.1 times already booked losses.  That seems unreasonably high with the strong evidence of a turn in the housing market.  The stress case (which are the Congressional Budget Office numbers) are for end losses to be 24 times the amount of losses already recorded.  If you believe that then depressive illness is probably the best diagnosis.  Surely you should not be doing stock analysis – and it puzzles me why the CBO should be putting out such patently ridiculous estimates. 

That said – KBW uses their base case in their model (59 billion of cumulative losses) and I use my base case (37 billion).  There is a further 22 billion difference between my assumptions and theirs.

I note that they do not justify their 59 billion number – and I went to great lengths to justify mine.  However if the housing market takes another massive turn downwards theirs (not mine) will be right.  If the housing market continues to bounce (as it has) then we will both be wrong – losses will be lower than either of our estimates.

The non-mention of write-backs on the private label securities

The KBW note does not make any mention of the possibility of write-backs on the private label securities.  I went to some lengths to show why – at least in Freddie Mac’s case – those write-backs were likely.  I produced an estimate of about $10 billion.  These write backs are reflected in part in current market prices for these securities. 

This adds another 10 billion difference between my model and the model used by KBW.  The total difference is thus $60 billion. 

You can do a little bit better than that too – because Freddie will earn some return on the 60 billion it does not lose – but lets ignore that.

KBW’s solvency model

KBW then presents a solvency model for Fannie and Freddie.  I reproduce it here:


There are some nuanced differences between my model and their.  My model of losses is a model of losses not yet recognised whereas they provide an estimate of end-cumulative losses.  That differs by the losses recognised to date ($5.8 billion though KBW state incorrectly that they are 8 billion).   These add to the difference between the KBW model and my model. 

Also Freddie Mac currently has a 7 billion dollar positive capital position (remember it made a profit last quarter and it had write backs of the private label securities).  KBW has ignored that (something I consider another pure date-input error).  So you could add another 15 billion benefit to Freddie on my assumptions over KBW.   

Nonetheless my model of Frannie is – on fairly easily justifiable assumptions – 60 billion better than the KBW model. 

Add the 60 billion to the net capital position as estimated by KBW (the –39 billion at the bottom of the table) and it is pretty clear that Freddie can repay the shortfall and make the preference shares whole.  Add in the remaining 15 billion (being the chargeoffs to date and the current net worth of Freddie after profits last quarter) and it repays easily.

A plan for Obama

Reform of the GSEs is quite tricky at the moment.  The jury is still out on their end losses.  Moreover the ether is full of self-interested lobbyists who want to take the good bit of their business (mostly interest rate risk management) and leave the bad bit (credit risk management) with the government. 

Winding down the GSEs right now runs the risk of killing the nascent recovery in the housing market.

The sensible course of action is to just wait.  This is policy that can be delayed without any real additional risk to the government.  (The government is already on the hook for the losses.) 

If my math is right – and I think it is – then the GSEs will appear solvent in time for the 2012 election.  The government can demand (and receive) almost 100 billion in capital to be repaid from them (which will make the budget look good and undermine the only viable Republican argument that the Democrats are irresponsible).  It will make the government look like good conservators of key institutions.  It will make Obama look like safe hands for running America. 

The anti-GSE lobby knows this is a possibility and they are determined to capture as much GSE business as possible right now – so they are vociferous in their claims.  Sensible people should ignore them.




Thursday, October 22, 2009

The goldsmith as retail bank

The parable of the evolution of private banking comes about with a story about a goldsmith.  The goldsmith has the strongest safe in town – so people deposit their gold for safe keeping.  People consider the certificates of deposit equivalent to gold.  The goldsmith however lends the gold in the vault for a fee.  That is real gold.  And thus banking acts as a gold multiplier.

These days of course it happens with paper money.  I have never borrowed gold – and nor has anyone in my usual social acquaintance. 

But I am not Indian.

Extracted from the State Bank of India annual report:

Gold Banking

• The Bank has taken several initiatives to undertake bullion business in a big way.
• The number of branches for retail sale of gold coins has increased from 250 in 2008 to 518 in 2009. The Scheme will be extended to cover all important centres of the country in 2009-10 by increasing the number of branches selling gold coins to about 1100. The Bank also undertakes supply of customised gold coins to corporates.
• The Bank has re-launched Gold Deposit Scheme at 50 branches to mobilise gold from domestic market for deployment as metal loans to jewellers.
• The Bank is in the process of setting up a dedicated Bullion branch at Mumbai to undertake bullion business in a focussed manner.

Friday, October 16, 2009

Whatever pleased the Lord, he did, in heaven and on earth, in the seas, and all the depths…

After considerable exchange with Zion's lawyers I have amended this post. The well is "probably" dry. For an explanation see this post.

For a small exploratory oil company with limited funds a dry well is really bad news. Three dusters and it is game over. Two and – well – you probably should be looking elsewhere.

And so I want to report to all that Zion Oil and Gas has had a probably dry well.

Zion is a special company drilling for oil in a special land. An alliance of fundamentalists Jews and fundamentalist Christians are fleecing their flock with a string of rights offerings to fund drilling in essentially non-prospective land in Israel. The company’s promotional dross is simply funny. This you-tube clip is simply a gem…

Zion however has reported the main well they were drilling to be dry. But – even funnier than the video is the press release announcing the bad news.


25 September 2009 – Operations Update # 20

As noted last week, we have successfully drilled this well to a depth of approximately 17,913 feet (5,460 meters).

This past week, we ran a ‘velocity survey’ in order to help increase our understanding of the geology of our license area.

A ‘velocity survey’ is a type of seismic survey where the seismic travel time from the surface to a known depth is measured. Geophones are lowered into the wellbore and a pulse wave sent out from ground level; the resulting signals are then recorded.

The velocity survey data will be used to correlate specific formations to reflections seen on the seismic sections that we used to map the Ma’anit structure.

We have decided, for the present, not to drill any deeper in this well and are now analyzing and establishing the priorities of the seven zones that warrant completion testing. However, the well bore is in excellent condition and it is possible that we will drill this well deeper in the future. Next week, I will comment further, but I’ll mention that this week Zion’s Chairman, John Brown, gave me a note with the reference Psalm 135:6 – ‘Whatever pleased the Lord, he did, in heaven and on earth, in the seas, and all the depths…’

Monday, October 5, 2009

Are the Spanish banks hiding their losses? Looking at the American data

Whether the Spanish banks are hiding their losses is a major debate going on in the blogosphere and has been detailed at length in  the Financial Times.  The stakes are very high – this is a debate about the stability of the Eurozone and possibly of Europe itself.


I have a lot of American readers whose interest in finance stops at the American border.  I need to outline what is going on. 

Spain had a monstrous building boom – a building boom on (at least) Californian standards based very much on coastal development.  The building boom has slowed considerably.    The building boom attracted relatively unskilled labour – as building booms are apt to do – and about 40 percent of all migrants to EU settled in Spain.  Wikipedia (I wish I could read the original Spanish source) state that the foreign population in Spain has gone from about half a percent of the population in 1981 to over 11 percent recently.  This change in racial mix has resulted in only minor tensions (with the possible exception of the large terrorist attack in Madrid).  

The financial crisis has hit Spain hard.  Unemployment is about 20 percent – though this overstates the GDP contraction.  A lot of the new immigrants are now unemployed

Twenty percent unemployment would normally result in large bank losses – indeed you would expect bank insolvencies.  However this has not happened.  The two giant Spanish banks (Santander and BBVA) appear amongst the most profitable in the world and have substantial market capitalisation.  Strangely Spain looks solvent despite its apparent economic catastrophe.  Part of the explanation might be that the economic problems in Spain fall mainly on the newer immigrants and the unskilled end of the labour market – and that these people are not the loan customers of the bank.  In this formulation the Spanish recession is about the same depth as the American recession – and the 20 percent unemployment rate is just an artefact of the migrant economy. 

Either way both big banks are depleting loan reserves (at least compared to delinquency and non-performing loans).  But both banks are reporting low losses and low loan arrears.

The banks however could be lying.

The stakes are enormous.  The bears (led by Spanish resident Economics Professor Ed Hugh and the financial research house Variant Perception) argue that the Spanish regulators and banks are conspiring to hide Spain’s insolvency – and when Spain turns out like Argentina either the European central bank (that is the old German central bank) will bail out Spain at great cost to the Central Europeans or the European monetary experiment – and possibly the whole European political experiment will be challenged as Spain fails economically and socially.  It’s alright to bail out Latvia after its economic disaster.  Latvia is small.  Spain however is large and important in a European context.   Ed Hugh would argue that it is best to deal with the problem now – because delayed it will get much worse.

Do not for a minute think that the stakes here are overstated.  Full blown economic collapses (eg Latvia, Iceland, Argentina) usually lead to riots and governments falling.  Where ethnic tensions run high those riots often have a racial element (rioting crowds find scapegoats).  Europe can paper over the Bronze Solider riots in Estonia (which pre-date the crisis).  They can paper over riots in Iceland and Latvia because the economies are small.  But an economic disaster in Spain would pose major difficulties – difficulties I think European Union would survive – but which would stress the system to its core.

To be this bad though the banks would need to be hiding their losses on a grand scale.  Most banks in crises hide a few losses (and spread them over time).  However the bears are truly apocalyptic.  The Variant Perception report is an absolute classic of hyper-bearishness.  If it really is that bad then either central European taxpayers are going to be stuck with a huge bill or the core political union in Europe is vulnerable

Less worried folk have pointed to inconsistencies in both Ed Hugh and Variant Perception’s data analysis.  An ordinary level bank failure could be dealt with by Central European taxpayers with only minor stress – however if you believe Variant Perception we are not looking at an ordinary level collapse – its way bigger than that.  Ibex Salad – a blog with the unlikely topics of the Spanish Stock Market, Spanish Economy and the olive oil business is the counterpoint to Ed Hugh and Variant Perception.

The data is mostly ambiguous – as the bears would argue – the data is largely faked anyway – finding inconsistencies in the data is to be expected.  They would argue that common sense – and the overbuild visible when you open your eyes – indicates that there is a serious problem here.

I really do not know.  I am not close enough to the ground in Spain to know – and – frankly – analysing (supposedly) faked data in a language I can’t read from a desk in Australia is unusually difficult.  But there seem to be four variants.

(a).  The Spanish banks are telling the truth – and this is a storm in a teacup,

(b).  The Spanish banks are doing a normal amount of bank over-optimism in the face of a crisis – and whilst the banks are really stretched (but not telling us) the banks are ultimately solvent – and the European experiment is fine,

(c).  The Spanish banks are in fact diabolical – and the losses are maybe 15-20 percent of a year of Spanish GDP – in which case a bailout by (effectively) German taxpayers is possible or

(d).  Variant Perception is in fact unreasonably bullish – and Spain will collapse economically and socially and we will be thankful if all we get back is someone like the Generalissimo.  The modern European experiment will be deemed to fail because a single European Union with a single currency can’t hold together in a crisis because Germany won’t or can’t bail out Spain, Italy and Greece in a crisis.

Instinctively I am in camp (b) above.   However I acknowledge all of the above are possibilities.   

Migration, racism and currency union

I am going to do a little further explaining of the stakes here.  The threat to currency union in Europe always was ultimately racism. 

When you have currency union you can no longer have a high interest rate in Spain or Italy (when those economies warrant a high rate) and have a low rate in Germany when Germany is recessed.  You have a single interest rate across the currency zone. 

The underlying state of most of the past 15 years was Spain booming, Germany mildly recessed.  Currency shifts or shifts in value between the Deutsche Mark and the Peseta can – by dint of currency union – no longer happen.   The main mechanism of economic adjustment is removed.

America has always done that.  The same interest rate applies in the rust belt, in the sun belt, in California and in Boston.  And we know how economic adjustment happens.  Americans move.  A vast number of Americans do not live in their home town and the bulk of the world’s busiest airports are American.  Internal American migration is massive.

However migration within Europe has always involved more issues.  The languages are different.  Several countries have histories of nasty endemic racism.  There are large cultural barriers.

Monetary Union – whether by design or just outcome was always going to confront those barriers.  And it was always going to be slow.  There is a reason why the German/Spanish imbalance was so long lasting last decade – which was that it is much harder for a German to move to Spain than it is for say a Hoosier to move to California. 

The changing racial mix of Spain throughout the boom seemed to show that massive shifts in racial mix and massive internal migration could be accommodated without the tensions of Europe’s dark past.  They were the embodiment and a proof of the European political experiment.  If Spain collapses beyond bail-out (as per Argentina) then monetary union is over – and economic union with racial harmony will be challenged. 

As I said – the stability of Spain is a big issue and the crux point is the losses in the Spanish banking system.

The Spanish American data

Sitting at my desk in Bondi Australia I have no real advantage in answering the big questions about the solvency of Spain and whether Spain really is the black hole in Europe’s balance sheet. 

But I can add to the debate.  The Spanish banks have American operations – and using reasonable comparisons we can work out whether the Spanish banks are hiding their American losses.  So far I have not seen any analyst do this – but it is surely worthwhile.  I am going to focus on BBVA because I once had a detailed understanding of their American operation (Compass). 

Several years ago BBVA paid a premium to buy Compass to form BBVA Compass.*  This is how they describe the bank:

BBVA Compass is a leading U.S. banking franchise located in the Sunbelt region. BBVA Compass is among the top 25 largest banks in the U.S. based on deposit market share and ranks as the third largest bank in Alabama and the fourth largest bank in Texas. Headquartered in Birmingham, Alabama, it operates 579 branches throughout Texas, Alabama, Arizona, Florida, Colorado and New Mexico

This bank files US statutory filings (better known as “call reports”).  There is no reason to presume that the Spanish regulator is conspiring with American regulators to fake the accounts of a bank headquartered in Alabama.  Moreover there are other sunbelt banks to use as comparisons – whereas in Spain you can only really compare BBVA to Santander – and the bears would argue that there is no point checking for faked data by comparing it to other faked data.

If BBVA is telling the truth in America then there is a reasonable chance they have a culture of truth telling.  That would suggest that they are probably telling the truth in Spain.

However if BBVA’s American accounts are riddled with deception (or at least an overly-optimistic prediction as to their losses) then it is likely that BBVA has a culture of understating losses – and that that culture extends home to Spain.

Of course the truth could be (and I would normally expect the truth to be) somewhere in the middle.  A little bit of excessive optimism is normal behaviour for a banker in a crisis.  But a little bit of excessive optimism does not imply bank or national insolvency – just some difficulty.  The European experiment can survive that. 

So what does the BBVA Compass call report say?  Is BBVA hiding its American losses?  If so – how bad loss hiding culture.

I have posted the key Call Report to Scribd

Here is a a comparison – comparing BBVA USA ratios to a sample of similar bank holding companies chosen by the FFIEC.  (You will need to click for the complete picture with all three tables in this post.)






The instant conclusion is that BBVA has higher delinquencies than the competition but has lower loan loss provisions and is charging less off than the competition.  This conclusion is robust almost no matter how you cut the BBVA USA data.  I think we can safely conclude that BBVA is hiding its losses.  If anything it is slightly worse than the above indicates because nobody in their right mind thinks that the comparables (larger American regional bank holding companies) are honestly stating their losses.  But if the comparables are understating losses then BBVA is understating them more

This puts me firmly into camps (b), (c) or (d) above.  The question is no longer whether they are hiding the losses – but whether the scale of problems (in Spain) is sufficient to cause major political ructions or whether it is just an issue for the stock market.  [Disclosure: I am short BBVA and the position is modestly painful as the stocks have appreciated.] 

How are BBVA hiding their losses in America?

There is a time honoured way of hiding losses in banking – a method that Variant Perception suggests is being done on a breathtaking scale in Spain.  The method is rather than call a bad loan bad – to just extend it a bit more credit.  If the borrower can’t pay the interest give them a bigger loan or line of credit.  They will use the loan to become current.  The slogan is that a “rolling loan gathers no loss”.  Even the most diabolical subprime mortgage book in the US showed only small losses until the market stopped rolling the loans. 

We have some evidence that BBVA is rolling bad loans.  Here is the loans outstanding by sector (again you will need to click for detail):



The level of rolling loans is not at the alarming levels that Variant Perception alleges are present in the Spanish economy.  Then again Alabama and the other states in which Compass is large do not have 20 percent unemployment.  The aggression which BBVA grew the book in the past five years however is breathtaking.  You can see where all that Spanish risk comes from and why Spain had such a monstrous property boom

Construction loans – a perspective from the American book…

The main allegation in the Variant Perception report is that the Spanish banks are massively overweight construction loans – and that they are extending those loans rather than allowing default.  The core statistic is given in the following paragraph:

Consider this: the value of outstanding loans to Spanish developers has gone from just €33.5 billion in 2000 to €318 billion in 2008, a rise of 850% in 8 years. If you add in construction sector debts, the overall value of outstanding loans to developers and construction companies rises to €470 billion. That's almost 50% of Spanish GDP.

They add that they think that most of those loans will go bad (which implies a Spanish crisis many times worse than America – and implied bailout requirements that are similarly bad).  

Construction loans at almost 50 percent of GDP is a truly astonishing figure.  The entire US mortgage market is roughly 10.4 trillion dollars – or about 75 percent of GDP (and as the crisis has shown that seems too large).  The idea that construction loans are nearly 50 percent of GDP had me falling off my chair.  I tried to confirm this figure (as it felt like garbage).  Alas I could not.  However Iberian Securities has done some legwork:

Variant picks-up a classic wild-card to spice-up the report. Specifically, they say most of the €470bn in outstanding loans to developers/construction (50% of Spain’s GDP) could go bad. The report forgets to mention- however- that a chunk of the €32 bn in outstanding loans to developers does not necessarily involve residential lending but commercial lending (which is relatively safe in Spain , in our view). It does not say either that a not-low percentage of construction activity in Spain involves public works, so a proportion of the construction-related debt (€141bn) should be attached to that public sector accordingly. Also it is worth considering that residential work-in-progress in Spain — one of the biggest contributors to the €320bn figure — is generally collateralized (with Spanish major developers reporting LTV of 50-65% approx). Factor-in these and the final loss on this portfolio should be a fraction of what Variant claims. In our models, we assume a 15% peak NPL ratio on Developers (7.6% in 1Q09) and a 10% NPL ratio on Construction loans (6.7% in 1 09).

Even at 6.7 percent of GDP construction loans are too big relative to GDP and the time in the cycle – but they are not big enough to cause problems for Spain.  I still cannot reconstruct the data to get construction loans that small in Spain.  There are over 600 thousand homes under construction in Spain – and most of those are financed.  Add the finance on those loans to other easily identifiable construction loans and you get over 10 percent of Spanish GDP.  I am not confident with the Iberian Securities estimate.

However we do get clean numbers in America for BBVA’s subsidiary – and it is clear that they are into construction loans in a fairly big way – and that their construction loan book is not good and they hiding the losses. 

Here is the composition of BBVA’s American lending book versus its American peer group. 


Its pretty clear that BBVA USA is not afraid of construction loans relative to peers – and – on the evidence presented here – is probably rolling them (and hence deferring losses).  This is nothing like the scale alleged in the Variant Perception report but it suggests that the basic Variant Perception allegation of hiding construction losses is more likely than not to be true

I should note that the construction loans are 10.64 percent non-accrual (which is slightly less than peer).  It seems unlikely you would willingly be expanding lending in this category with those credit statistics.  Its far more likely that the company is hiding losses by rolling non accrual loans.

A note on scale

All these problems of the same type that Variant Perception alleges in Spain – but none are of the scale Variant Perception alleges in Spain.  In other words I can unequivocally support the notion that the Spanish banks are hiding their losses – but support for the notion that these losses are so large that France and Germany will be left “holding the bag” is not to be found in the US data. 

What the Spanish bankers have been telling us about their credit is – at least on the American data – easily shown to be lies.  We just don’t know whether they are big lies.

For the sake of Europe I hope they are not.






Post script:  I have linked to a few blogs here – but the important ones are Ibex Salad, and A Fistful of Euros.  These blogs disagree with each other – but they are of the highest quality.  If you are interested in this stuff then put them on your blog roll. 

*A few disclosures are necessary here.  All the key players in the blogosphere debate (hyper-bears and moderates) are my “Facebook friends”.  In this day-and-age you don’t have real friends – just computer friends.  I do not know what they will think of me after this post.  I expect disagreement and I will post follow-ups.  Some I am sure will disavow my “friendship”. 

Second - at Bronte we have a small position short the Spanish banks – it has not been profitable.  Moreover we are deliberately short them on the US stock exchange – which means we are short the banks long US dollars.  That has been particularly bad of late because the US dollar is weak.  However in a true crisis the Peseta (and yes I mean the Peseta) will be really weak – and we would rather be short them on a US listing where the cash balance is held in US dollars than on a Spanish listing where the cash balance is Euro converted into Peseta at an unfavourable exchange rate.  

Just like with Charter One – BBVA caused me more than a dose of heartache.  I was short Charter One when Sir Fred Goodwin and his RBS idiots came and purchased it at a massive premium.  That was the single worst day of my career.  Similarly I confess to being short shares in Compass when BBVA bid for them (admittedly for a much smaller premium).  Nonetheless, it was – as they say – not a good day at the office.

Finally it is a long weekend public holiday in Australia.  You can tell a nerd when he writes 3350 words on Spanish banking when he is meant to be on holiday.

General disclaimer

The content contained in this blog represents the opinions of Mr. Hempton. You should assume Mr. Hempton and his affiliates have positions in the securities discussed in this blog, and such beneficial ownership can create a conflict of interest regarding the objectivity of this blog. Statements in the blog are not guarantees of future performance and are subject to certain risks, uncertainties and other factors. Certain information in this blog concerning economic trends and performance is based on or derived from information provided by third-party sources. Mr. Hempton does not guarantee the accuracy of such information and has not independently verified the accuracy or completeness of such information or the assumptions on which such information is based. Such information may change after it is posted and Mr. Hempton is not obligated to, and may not, update it. The commentary in this blog in no way constitutes a solicitation of business, an offer of a security or a solicitation to purchase a security, or investment advice. In fact, it should not be relied upon in making investment decisions, ever. It is intended solely for the entertainment of the reader, and the author. In particular this blog is not directed for investment purposes at US Persons.