Friday, September 9, 2011

Electronic fax? Really. Doing the time-warp with

Efax is yesterday. You know the fax machine you have out there in the cloud where you receive a fax (with a dedicated line) and they send you the fax via email. And you send the fax via email - you post it to your electronic fax provider and they fax it for you.

Electronic fax is so yesterday that when I think about it I also think of dial-up internet, I am working in the Australian Treasury, I had an American girlfriend with a PhD in economics and the Spice Girls are the hottest band in the world. (And my girlfriend objected to the Spice Girls which did not seem sporting...)

Actually electronic fax does not figure very much in my life at all. I have sent only one or two faxes in the last decade and I have received none. The closest I get to faxing is scanning a document and emailing that.

So it was a great surprise to come by (JCOM:NASDAQ). You see J2Global is an electronic fax company - it is the old e-fax. It is the only brand in electronic fax I remembered.

And I could not imagine that it possibly had about 300 million revenue and a market cap of almost $1.5 billion.

Like really? Like who are you kidding?

And note this was a price to sales ratio of five - a number consistent with the most profitable or highest growth companies in the world. I mean there are not many companies in established businesses with a price to sales ratio of five.

And very few businesses that should be in such obvious decline as electronic fax would ever deserve such a ratio.

But there it was, listed, with a big cap, fairly transparent looking accounts and a stock that slowly levitated over the decade in much the way that bricks don't.

This was so unexpected a find that I really did double-take.

There was a bull-story relentlessly promoted of course - which was that J2Global was a "cloud company" in the sense that internet-fax was one of the first applications out there "in the cloud" but it was a cloud company a little less sophisticated than Hotmail or the AOL "walled garden".

I had to find a customer

I was wondering whether my experience with fax machines (they have rapidly become irrelevant) was the universal one. I could not imagine being a subscriber (at $16.95 a month no less) to efax and wondered why anyone else was.

I rang our foundation client - an industrialist about 15 years older than me who is on the road quite a bit. I thought he might be a subscriber - but he can't imagine wanting to join. He wondered where I found companies like this.

So I kept asking people whether they would subscribe to an electronic fax person - and I found one - a fund manager who thought he paid $3 a month and it was before MyFax purchased his supplier (MyFax is owned by Protus - more about them below). My friend now pays $6 a month even though the cheapest price on the MyFax website is $10 and has been for some time.

The price differential was so large ($3 a month versus $16.95 per month) that I wondered why anyone would pay it. It is fairly easy to find suppliers at $8-9 a month. and others have prices for very light users down to $3.50 per month. And lets face it - most people who use faxes are very light users...

The first explanation for the continuation of the business was inertia (which is the same reason why some people still use dial-up internet).

Inertia or ripping off your customers?

YouTube is a valuable resource - you can usually find someone complaining about or proselytizing for a product.

In this case it was complaining. You see it is very hard to cease being a efax client. The video below shows you how when you dial up they tell you that you should cancel your account on the net. On the net it gives you a number to dial up (with a long wait). That gives you another number (with a long wait) and that tells you to cancel on the net.

You get the idea: phone center hell.

I wondered whether this was typical: whether the modus-operandi of this business was to keep ripping people off even when they wanted to cancel.

Alas there is strong evidence that it is. One of my colleagues found that the Better Business Bureau gave the company an F rating (on an A+ to F scale). The Better Business Bureau had closed 419 complaints against the company in the last year. Here is a summary:
Most complainants allege billing disputes or difficulty canceling accounts. Many customers complain of excessive hold times, anywhere between 10 minutes and up to several hours, when attempting to speak to a customer service representative. Other complainants allege the fax service does not work or fax numbers are reassigned without notice or justification. A few complaints allege the company does not disclose the fact that the number of faxes you can send is limited or that there is a per page charge for every outgoing fax. The company responds to some complaints by canceling accounts, issuing refunds, and apologizing for the interruption of service. In a few cases, the company retrieves and reactivates fax numbers or verifies the service is operational. The company further responds by claiming customers have more than one account or that they have no evidence of cancellation.
You see these guys are appear a little sharp (at least in the view of the Better Business Bureau). You cancel an account and they will keep billing you claiming no evidence of cancellation or that you have more than one account (as if anyone would have more than one electronic fax number).

This sort of business behavior is harder to pull off outside the US. The US seems to have a culture that accepts consumer rip-offs (predatory lending for example). Other cultures think of unconscionable conduct. This is unconscionable. In the US I guess it works until they get a class-action lawsuit.

Anyway it is worse than just relying on inertia. The company raises rates regularly and rely on inertia not to object to or shop higher rates. Here is a blog post which complaints about their rate raising:
eFax today announced that they are raising their already gouging rates from $12.95 a month to $16.95 a month, unless you want to “lock in” the $12.95 a month rate by paying it annually (i.e. pay the $12.95 x 12 up front to the tune of more than $150.00 a year).
I call it “gouging” because eFax originally started out with this model: you could either pay to get a fax number that was local to you, or get a free fax number which could have an area code in any part of the U.S. except local to you. 
At that time it cost a mere $4.95 to have the local number, which was part of a service called “eFaxPlus”. So eFax Plus was only $4.95 a month, and that was as recently as the year 2000. 
However, once they got you hooked (and having distributed your fax number far and wide) they boosted their fees to $12.95 a month - more than double what you’d signed up for - and you were stuck, unless you were willing to lose the fax number you’d given out to everyone. Still, it was month-to-month so you could make a decision each month as to whether it was worth it. 
But now they are doing it again, saying either pay that $12.95 a month up front for a full year (a total of $155.40 a year), or pay the exhorbitant rate of $16.95 a month. And, if this prompts you to decide to cancel the service, let me tell you up front that there is no easy way to cancel your account [the blog post purports later - not quoted - to tell you how to cancel the account]. 
Here is the relevant portion of the email that I myself received today:
“The monthly subscription fee for all eFax Plus numbers on your account will be changing.
Starting on your next billing date, the monthly fee for each of your eFax number(s) will be $16.95. 
You will also receive an enhanced level of eFax service. 
Receive up to 130 fax pages and send up to 30 fax pages free each month. Store faxes up to one year with your eFax Message Center. Get 24/7 live phone support. 
To lock in the old $12.95 rate for the next year, switch to annual billing by clicking here.
Please respond by October 01, 2006.” ...

But here is the nub of it. They raised the rates to $16.95 per month in 2006. They have not raised them since... there is a limit to how much you can gouge - even if you make it so hard to exit that people have to stay in a phone center queue for three hours or take you to the Better Business Bureau just to exercise their rights...

And nobody would rationally sign up - as the same blog post makes clear this service is at least twice as expensive as the competition. If you do a simple internet search you would find that efax is poorly rated, and actively disliked by consumer groups. This is a dumb product to sign up to.

But the revenue is still growing. Why? Why the hell would anyone want to sign up to this?

My first thought was that the accounts were a lie.

Falsifying the fraud theory

I see a lot of stock fraud - and I immediately thought that they were just making the rising revenue up. After all I could not see why revenue from electronic fax should be rising. It just seemed so unlikely.

And they really are (at least in the words of their complaining customers) nasty. Customers just say the company is a scam. Just the sort of people who would fake accounts - or so I incorrectly thought.

Bolstering my theory that they were faking their accounts was their less-than-mainstream auditor. They use Singer Lewak LLP. The SEC database allows us to see all the other companies audited by them. It is not a list of known frauds - but it is hardly inspiring.

So I tried to puzzle how they did it. The company generates lots of cash but pays no dividend which makes it hard to confirm the cash is real. But if the cash is real the profits are real. And if the cash is not real the profits are not real.

So I tried to work out whether the cash was real. Here is the balance sheet:


December 31, 2010 and 2009
(In thousands, except share amounts)

Cash and cash equivalents
Short-term investments
Accounts receivable, net of allowances of $2,588 and $3,077, respectively
Prepaid expenses and other current assets
Deferred income taxes
Total current assets
Long-term investments
Property and equipment, net
Tradenames, net33,3968,760
Patent and patent licenses, net 18,10214,955
Customer relationships, net 36,6747,743
Other purchased intangibles, net
Deferred income taxes
Other assets
Total assets
Accounts payable and accrued expenses
Income taxes payable
Deferred revenue
Liability for uncertain tax positions
Deferred income taxes
Total current liabilities
Liability for uncertain tax positions
Deferred income taxes
Other long-term liabilities
Total liabilities
Commitments and contingencies (Note 8)
Stockholders’ Equity:
Preferred stock, $0.01 par value. Authorized 1,000,000 and none issued
Common stock, $0.01 par value. Authorized 95,000,000 at December 31, 2010 and 2009; total issued 53,700,629 and 52,907,691 shares at December 31, 2010 and 2009, respectively, and total outstanding 45,020,061 and 44,227,123 shares at December 31, 2010 and 2009, respectively
Additional paid-in capital
Treasury stock, at cost (8,680,568 shares at December 31, 2010 and 2009, respectively)
Retained earnings
Accumulated other comprehensive loss
Total stockholders’ equity
Total liabilities and stockholders’ equity

During 2010 cash dropped from 197 to 65 million - the difference been spent on acquisitions. Goodwill went from 81 million to 282 million and the cash flow statement reveals that they spent $249 million on acquisitions - that $249 million being cash generated over the business over time.

If that $249 million was real then the business is real - and there is no wholesale fakery in the accounts. If they bought real assets from real, unrelated people then I would conclude the accounts were probably more or less OK. If they purchased assets from parties I could not identify then they could be from fake parties or related parties. The only assets you can buy with fake cash are fake assets and you tend to have to buy fake assets from fake parties or related parties. So I went looking for the assets purchased and who they purchased them from. This was my test of fakery. [Lots of people have asked how I do this. This is as good an explanation as any though there are more than a few tricks in the Bronte arsenal...]

So I went looking for what the acquisitions were. Here is the key text from the 10K:
During 2010, j2 Global acquired eight businesses: (1) the voice assets of Reality Telecom Ltd, (2) the fax assets of Comodo Communications, Inc, (3) the unified messaging and communications assets of mBox Pty, Ltd, (4) the assets associated with the email hosting and email marketing businesses of FuseMail, LLC, (5) the assets of Alban Telecom Limited, a UK enhanced voice services provider, (6) Venali, Inc., a Miami-based provider of enterprise Internet fax messaging solutions, (7) keepITsafe Data Solutions Ltd., an Ireland-based provider of online backup services, and (8) Protus IP Solutions, Inc., a Canadian provider of Software-as-a-Service (SaaS) communication services and solutions to the business market.
Now I had a work program. I wanted to find out what I could about the acquisitions. If the acquisitions were real companies from reputable parties and over $200 million was paid I could confirm that the business was substantially real. If they were acquisitions from dodgy-brother-related-parties then I would be ringing the Longtop-type-fraud bell.

I had to work out where they spent that $200 million. Being a methodical type I did the acquisitions in order.

This press release covers (1) Reality Communications, (2) Comodo and (3) and something called Quexion which is not on the list of acquisitons. The total consideration is not material.

Mbox (3 above) is a small Australian supplier. It is hardly the use of $200 million. Nor was it Fusemail (number 4 above) as that had only 6000 subscribers and the price was not disclosed. Alban telecom (number 5) was also specifically described as not material.

Venali inc (number 6 above) was the first with a material purchase price ($17 million). It also had $10 million of revenue and the purchase involved the settlement of some patent disputes.

By this time I was getting excited. Nothing here came close to being a real purchase using over $200 million in cash. For number 7 on the list ( I went to the Irish companies office and pulled the balance sheet. This was a trivial purchase: is a remote-drive business a bit like Amazon Cloud Drive but with much higher and opaque pricing. They do not tell you their pricing on the website but I wrote to them for a quote:

50G - $75 per month
100GB - $100 per month
Additional GB is .80cent per month

This is approximately 12 times the pricing of Amazon leaving me wondering what this business does at all. (If you wanted to protect your data would you sign up for this company or Amazon with the better balance sheet and only one twelfth of the subscription price?)

I further looked up the directors and they are not key contributors to the tech world.

Whatever - it is, the pricing (opaque and outrageously expensive) and its lack of transparency or even an obvious competitive offer made me fairly sure this was a nonsense business. (It may have a product - but whatever - you can get it elsewhere cheaper...)

Now I am getting really excited. I have checked the first seven out of eight acquisitions and with one exception (Venali Inc) they are either not material or nonsense acquisitions.

I think I have my next Longtop Financial Technology. I was actually dancing around the office. (It happens - and you do not want to see it...) We already had a small short on this stock but I was going to make it a huge put-option position and see what I could do to make it pay. That sort of money-making opportunity gives me goosebumps in excitement - not quite like a 15 year old boy going on a date - but it is up there... (It is also a very large proportion of Bronte's cumulative returns.)

But I am a thorough guy and so I checked the last one - Protus IP Solutions. And my "its a Longtop" thesis collapsed. Simply collapsed. You see they purchased Protus for $213 million. If that $213 million were paid to dodgy parties or related parties I could hazard a guess that the $213 million and the purchase was a fiction. But the vendors were highly reputable venture capital funds: Bank of Montreal Capital Corporation, Edgestone Capital Venture Fund, L.P., B.E.S.T. Discoveries Fund Inc. and New Millennium Venture Fund Inc. And the number was reported in the Canadian press.

So we can conclude the $213 million was real.

And that $213 million had to come from somewhere - and the company did not raise it in the market. So we know the company actually generated that money. Protus is billed as a "software as a service" offering but it is in fact another electronic fax company. MyFax - the service my friend subscribes to - is a Protus product.

What you are seeing here is my thesis (that this was a Longtop style fraud) collapsing around me. I danced around the office prematurely - and the huge option position I was considering: well we never put it on.

I went home thinking I had done a decent two days work and achieved not very much. This is - of course - the lot of a fund manager. (Most days we achieve very little...)

But I am still left pondering the business.

So what J2Global really is

Much to my chagrin we now know that J2Global really is a fax company. And it really generates a lot of cash which means that there really are suckers (ahem customers) who pay $16.95 per month for a fax line from a company that it is very hard to unsubscribe from.

But over time the company builds up cash and then buys something - another fax company.

We also know that it puts up the rates. My friend above who had the fax at originally had it at another provider (and he remembers it as $3 per month). That company was acquired. Now $6 a month. Now it is inside J2Global the price will push up and up - and may eventually reach $16.95 - a price where J2 seems to stop.

And this really is a business about yesterday: he sends less and less faxes each year and so does everyone else. But J2 Global will raise his rates bleeding him for as much cash as they can before he gives up (analog) fax as an antiquity like dial-up internet.

Some numbers

Protus says they had revenue at about $72 million per annum. The other significant acquisition (Venali) had $10 million revenue per annum. Acquisitions thus added revenue of about $80 million per annum or $20 million per quarter. The other acquisitions had to be a few million more.

I looked at the first quarter of this year (the first full quarter with those acquisitions which can be compared to a PCP without those acquisitions). Revenue goes from $60 million to $73 million. That is a nice rise - but not quite enough to account for the acquisitions. There is a pretty sharp underlying decline.

The next quarter they seem to have put some prices up (which is what they do) and revenue growth resumes. But you have to imagine that the clients are kicking back - slowly fading away even if it does take 3 hours on hold to cancel your account.

So what we have here is actually pretty straight forward. It is a business in decline - but it is dressed up as a cloud computing company which I guess makes people happy. It can't be much of a cloud computing company because property, plant and equipment is only $13 million and is not growing and cloud computing is capital intensive.

Because it is (at least from an equity-marketing perspective) a cloud computing company it sports a respectable PE multiple and a reasonable stock price. But revenue growth comes from two places (a) buying the competitors and (b) raising prices - and that is offset by the general decline of the electronic fax industry.

They slow that decline by making sure that customers have a really hard time unsubscribing.

Sure it generates cash - but it spends the bulk of that cash over time on acquisitions - necessary to actually get revenue growth.

The management behave as if the stock is expensive. They are selling their shares at a pretty good clip. The management at Salesforce.Com are also selling shares at a pretty good clip and that seems not to hurt them. I have sometimes purchased shares from management and done well.

And I could be wrong in the core thesis. Electronic Fax may really have a future. Maybe they can crank the rates to $30 a month and the customers will continue to love them. It seems unlikely to me - but no more unlikely that some hip 18 year old could do a Spice Girls cover on YouTube and get millions of views.

And that happened. And to complete the homage they repeated the Spice Girls trick of doing the video in one continuous take.


PS. Disclosure: I still have my small short on. It will stay small - technical obsolescence is something we generally short. Hyping old industries with fashionable words (like "cloud" and "software as a service") is also something we generally short. But my excitement: that was misplaced.

Wednesday, September 7, 2011

Repost: My old notes on Northern Rock

This is a post I made fairly early in the history of the blog - and a post I think should have got more attention. (The original post had no comments and nobody much link to it.) I re-read it today (because it came up in conversation). I am kind of proud of it - so allow me the luxury of a repost.


In 2005 I travelled to the UK to study the UK banks. I should have shorted the lot of them. But I didn’t. But for the record here are my notes – written on a slow English train – about Northern Rock – and never finished. I have edited it only to remove references to my actual sources.

I put this up not to gloat (but its nice). Rather I am going to do an expose of another UK bank shortly.

I cannot gloat too much - because whilst these notes are amazingly prescient I did not make a fortune on the stock. I predicted rain - but its making an ark that counts!


Northern Rock – leverage mortgages to the max

Northern Rock is a very simple bank. It has only one strategy and it makes no bones about taking this strategy to its absolute limit. They are completely non-forthcoming about where the limit of this strategy might be – but we will see that later.

The strategy of Northern Rock is to grow the mortgage book. Fast. All decline in margin is to be made up by volume growth. They are absolutely explicit about this – the corporate objective is:

  • Grow the asset base by 25 per cent per annum plus or minus 5 per cent
  • Grow earnings by 15 per cent per annum plus or minus 5 per cent.
It is pretty clear that they have even de-emphasized the old building society funding base which is I think might be actually shrinking before “hot money” high rate deposits and foreign deposits[1]. At the conference they told us how they were still concentrating on the deposit base but it had the tone of protesting too much. Besides its clear that rating agencies and bond markets want some deposit based liquidity.

I am also not exaggerating in the slightest about what the corporate strategy actually is. The management must have used these two bullet points five times in my presence (and I was not with them long).

Well it is pretty clear that growing the balance sheet by 25 per cent per annum grows risk by something near 25 per cent per annum (the company will deny this – more on that later). Growing profits by 15 per cent per annum means that capital will wind up growing by 15 per cent per annum (give or take a little).

If you grow risk by 25 per cent and profits and capital by 15 then either

  • You will run out of capital and the regulators or rating agencies or bond markets will not allow you to fund your growth – in which case the growth fizzles out at best, or
  • You will eventually be taking so much risk that the return on capital will not be rational in an ex-ante basis. Some point ex post you will blow up, possibly spectacularly.
If you think I am exaggerating what this strategy is then here are the five year summary numbers from the annual report. Ten year numbers were reported at the conference and they had pretty well the same appearance.

INSERT [sorry I wrote this on the train and had a hard copy of the annual. I never bothered putting the actual table from the soft copy in the report]

Note there is no credit data here. Nowadays credit losses are negligible in UK mortgage banking.

Obviously you should notice the massive expansion of leverage in this book. The asset number to look at is the “total assets under management”. This number includes securitised mortgages where the residual credit risk is at Northern Rock. (The main buyer of this paper are Japanese banks both major and regional.[2]The total leverage of book has moved from 27 times to 42 times. Obviously this can’t go up for ever but I suspect it can go for quite some more time. (You will see that 43 times leverage is not unusual for a UK bank.)[3]

When I was with the company I tried to explore the limits to the strategy and got nowhere useful. It would be nice to know though because when the company reaches the limit of its leverage it would be a safe short (unable to grow and possibly facing further margin erosion). Until then its probably a better long then a short as there seems no impediment to earnings increasing at at least the teens and the PE is only XXX now.

That said – here goes for my discussion about the limits to Northern Rock’s growth. The company told everyone at the conference that mortgages were safer than conventional loans probably deserving a 33 per cent risk weighting. In Australia and the US the standard is 50 per cent risk weighting for mortgages with no insurance less than 80 per cent loan to value (LTV ratio) so by international standards 33 per cent is aggressive.[4] That said Northern Rock suggested that there mortgages were substantially safer than the average (measured delinquency at about half the market rate) and hence they should have half the risk weighting – call it 17 per cent. They even went as far as to say that the regulator agreed with them. [Some comments have been removed here because they report indirect comments from regulators.  I cannot vouch for them on this blog.]

Now if your mortgages require only a 17 per cent risk weighting then you can be 84 times levered with a Tier One ration of 7 per cent. [Figures: 1/(.17*.07)]. If a third of your tier one capital is subordinated debt (not uncommon in banking these days especially in the UK) then your total leverage ratio might be well over 100. I did this calculation for them and they were quite uncomfortable – because they are hardly wanting to telegraph to the rating agency that they will one day be 100 times levered. (It would increase the cost of their funds now and hence further compress their margin.)

I did not get any useful feed on where the limits to growth are. However looking at the other banks (discussion later) I suspect that the limit is roughly 60 times levered. That would suggest (growing capital at 15 and earnings assets at 25) that there are four to five years left. However by that point the bank has almost ₤200 billion in assets – large compared to the UK mortgage market. Its funding would be totally ridiculous. [Comment deleted because a senior executive of another UK bank thought Northern Rock would go bust in 2007. I just do not want to dump him in it.] Something will crack – but in five years earnings could double again and the stock could be an abysmal short.

If you look at the five year summary above you will notice that the mortgage originations in any year the gross lending is substantially larger than the net lending. In 2004 gross lending was GBP23 billion - about 45 per cent of the total managed book at the end of 2003. Asset growth is only about 45 per cent of net lending.

This leads into the way that the lending is done. Its TEASER RATE lending. In the UK new mortgages (especially from this bank) tend to have a “teaser rate” which applies for two to three years (mostly two years). The fashion of late is to have two year fixed rate loans on very low spreads (the yield curve is flat in the UK) and to offset the spread a little bit in up-front fees. The loans revert to old fashioned (and fat margin) standard variable rate (SVR) at the end of the teaser rate period. The profitability of this business is determined by how many of the loans you manage to keep on your books after the teaser rate wears off and on any incidental products you might sell to the mortgage holder. If the loan comes in through a branch rather than an IFA the loan might be more profitable because it does not cause a broker fee.Internet channels are also relatively profitable.

The way that Northern Rock grows so fast is that it is the king of the teaser rate. It has however very poor retention. Bradford and Bingley told me that Northern Rock would boast about their 400 retention staff (they will cut your rate if you ring up because the alternative is for you to go elsewhere and a cut rate loan is more profitable than a new brokered loan). The competition also target Northern Rock customers. Natwest (HBOS) have regular advertisements on TV showing people on a rollercoaster with very low mortgage rates about to swing up wildly (and quite graphically make them sick). They suggest that you are nuts if you take this swing up and offer you GBP100 if you are an Alliance & Leicester or Northern Rock customer (not a B&B customer) and your rates refinance and you do not want to pick a Natwest mortgage. Its clear that Natwest however is trying to get people through the (lower cost) direct channels. (This sort of competition exists in deposit pricing too.)

There is a test as to whether all this teaser rate activity produces long term customers. Just look at the implied fall off in loans versus the originations two years ago. In 2004 it appears that over GBP10 billion repaid. Gross lending two years ago was 12.5. There is clearly some but quite limited success in retaining the customers. When pushed on accurate data on this issue Northern Rock were simply not forthcoming.

There is one more thing that is quite revealing about Northern Rock – and that is the effect of International Accounting Standards (IFRS) on balance sheet and profitability. UK companies are being forced to adopt IFRS and whilst it is an issue with a lot of noise for many companies the differences are small. They are not small at the Rock. In particular IFRS requires that income and expense charged as a fee but which relates to some period gets amortised over the period. Now remember that the shift in the market has been from floating rate teaser products to fixed rate teaser products with high fees. The Rock has been booking those fees up front inflating earnings and book value. IFRS will (under the guidance given at the conference) reduce book value and earnings by about 10 per cent. (Leverage is probably closer to 47 times – and using a sixty times limit the company probably has only three years rather than five.) The company seems to think that IFRS is a bad idea (aren’t the fees cash). But I am never quite sure whether the mix of fees and spread has shifted driven by accounting considerations or whether its driven by the realisation that churn is going to remain incurably bad or get worse. (Obviously enough up front fees are a good idea if you are scared of churn.)

All of this was enough to make me pretty bearish on the stock. But it got worse. They simply stretched numbers to say what they do not. If it were not for the low standards of America I would say they lied – but I suspect just being economical with the truth was closer. I have referred above to the notion that their delinquency is half the industry average and therefore (as they argue) they deserve only half the regulatory capital charges of the competition. The problem is that a delinquency rate simply does not make sense when your growth has been as rapid as the Rock. I tried to tease out of them the notion of a “growth adjusted delinquency rate”. No luck. I tried to work out what the delinquency by age of mortgage was so I could do the numbers – no luck. They simply were not forthcoming and even attempted to mislead me.[5]

The place however they misled most blatantly was on the margins both historic and prospective. The company stressed that I should not just look at interest margin – rather I should look at fees plus margin over assets – especially as they had shifted to fixed rate low margin loans with relatively high fees.Ignoring the IFRS issue (as they did) the average margin on the book is 125bp and it has fallen every year – most notably during 2004. They wanted to tell me that the INCREMENTAL margin was 110-120bp. This is much higher than the competition tell me the margin is (40-80bps) and simply cannot be squared with the margin figures in the above table. The problem is that they will soon hit limit leverage constraints (but they would not tell me what those constraints were) and were aware that their margins (hence earnings and ROE) would continue to drop once they hit those limits.

As for credit risk. The company told me that they had a position in the broker market as offering the cheapest loans to the best credit. I have one reason to disbelieve them. The Rock has a lower rating and hence a higher funding cost than several competitors – and hence would naturally have a relative advantage further up the risk spectrum (its hard to do good credit well with a low rating). Also they told me in another breath that they had industry leading margins (which did not reflect in the accounts).Stuffed if I know. They seem to think that they will be alright with a 20 per cent fall in the property market. They seem to get concerned when you talk about a 30 per cent fall – and they seem to think that a 35 per cent fall is impossible. I heard all the old hoary clichés: “they are not making any more land in the South East” etc. I should get the stock brokers to organise me some chats with mortgage brokers and IFAs. But I am – until that – inclined to believe that the threshold for pain is about a 30 per cent fall in the property market – and that falls beyond this range could lead to a wipe-out because the loan book is new (hence has not had the chance to get much appreciation into it) and is so levered. [Ok – I was wrong here – they went bust on funding.]

You do have to give the bank credit for one thing though. They have got their costs quite low – 38bp of assets and probably lower under IFRS. This is one of the lowest cost structures in the world. The management will point this out as almost their crowning achievement. They had to do it (had they kept their old cost structure the squeeze in margins would have wiped them out). It does however look difficult to keep reporting lower costs – this looks a lean operation.

Do I want to short it? I wouldn’t object – but I suspect we can do better with timing. The investment bankers are convinced that if something went wrong it would be purchased at 80 per cent of book on the way down. Maybe that is true now – but it will not always be so. I was staggered by the lack of sophistication of the staff – I met the CFO and he was either dumb or a liar or just assumed I was dumb.This company is totally dependent on the goodwill of financial markets. I put to them that they were dependent on the kindness of strangers – and they bristled. They thought that people invested in UK mortgages because they were good investments. Why – so they thought would you invest in Italy?

For discussion.

[1] The shrinkage does not show in the numbers – but the deposit base includes €2.5 billion in French deposits which are really hot-money commercial paper and some Japanese deposits. The claimed retail deposits in the five year results page I reproduce (17239) does not match the balance sheet (20342) and I am assuming the difference is roughly the above €2.5 billion and other quasi wholesale money. I can’t tell how much “hot money” there is but Northern Rock were pretty keen to advertise a 5.4 per cent rate.The shrinkage is a guess – but the company was not far from admitting the same when pushed on the issue.
[2] Amazingly the CFO was prepared to name the six Japanese banks which purchased the paper. I told him we had an interest in the Japanese banks and it would help me understand their books. He did not remember their names but he had been on a roadshow to Japan. I have virtually never had a company volunteer this sort of information. Its pretty naïve to do so as there is more than one way to interfere with a banks funding. If he had thought about it clearly it was just as likely I was short Northern Rock than long it. It was part of a general attitude I came across in Britain (nobody appears at all concerned about the vulnerability of funding bases). I pretty well floored Michael Oliver (the very experienced IR guy at Lloyds TSB) when I told him this when I took him out to dinner.
[3] The US tends to have a regulatory limit on leverage at 20 times. Any further and the informal rule is that you can expect an intrusive visit from the regulator. [Some comments deleted here.]
[4] [Regulatory footnote removed – partly because it is wrong – and I am embarrased.]
[5] On the train between Leeds and Leicester I chatted to a woman in her fifties about the banks. She had a mortgage on SVR (an old high margin mortgage) with Lloyds TSB. I asked her why she did not refinance it and she told me a story about weakened credit. Her husband no longer worked and her income paid the mortgage and supported the family. She had plenty of equity in the house but she (erroneously) thought that she could not refinance. Refinance would have saved her GBP500 per year. It was an easy decision had she been informed. Lloyds is hardly going to inform her. But there is a lesson here – the back book are going to have higher delinquency than the front book but without necessarily worse credit. It self-selects this way in part. The comparison that Northern Rock had on their delinquency rate was at best grossly misleading. (There is a possibility that they believed it though which would suggest incompetence. In this case they misled so transparently they might well just have been dumb.)

Friday, September 2, 2011

Risk management and sounding crazy

In early June Carson Block and his firm Muddy Waters research published a report which made outrageous sounding allegations against Sino Forest - then a highly respected Canadian listed Chinese forestry company that had borrowed well over $2 billion to develop and expand forestry operations in China.

The base allegation in the report was that most the forests did not exist and by implication the (more than) $2 billion borrowed was stolen. Presumably many more shares have been sold too taking the total theft well above $2 billion.

It was an outrageous jaw-dropping allegation and Carson made no attempt to soften the blow. His language was inflammatory because his message was inflammatory. When $2 billion is stolen by reputable people I can't see how you can say that without appearing inflammatory. Some people, understandably, refused to believe it.

In response I set about reading ten years of Sino Forest accounts and decided Ocham's Razor style that Carson being right was the only explanation that I could think of (and probably the only simple explanation) consistent with the facts. And I shorted some, and then shorted some more. The two main posts that outlined my thinking were here and here.

Still I am obsessed about discovering the ways my positions can be wrong (and my business partner is even more obsessed) and we followed the reports of everyone (and there were many) who doubted Mr Block's research.

Dundee Securities was the most prominent Sino-supporter labeling Muddy Water's research a "pile of crap". Somewhat more considered sounding (but also flat wrong just more reasonable sounding) was Metal Augmentor who found Carson "loose with the facts and somewhat breathless". On the naive-sounding side was Susan Mallin whose complaint was that she had "never seen a research report written in this manner". More prominent people were fooled too. I have heard Richard Chandler is not an idiot (but he seems to like losing money - I keep finding him invested in frauds and he does not answer my emails). He dropped $150 million on Sino Forest.

The analysis of these people was staggeringly weak and self-referential (I can't speak for Chandler because I never saw his analysis). They judged Sino Forest against data provided by Sino Forest or people associated with Sino Forest. This is an elementary mistake in assessing fraud. To find fraud you need to be able to judge against things you are fairly sure are not fraudulent.

Everything the Carson Block doubters said sounded reasonable. Certainly more reasonable than Carson Block sounded because Carson Block held the radical position. Sounding reasonable however was wrong.

I think what is going on here is a general problem. When someone says something - anything - that is so far from the consensus as to sound outrageous then they will be considered mad, and sometimes they will be considered mad even after they are proven right. This is iconic in the history of science from Galileo (who observed the moons revolving around Jupiter and the crescent shapes of Venus and deduced much from those) to the more minor genius of Émilie du Châtelet (who thought, contrary to Newton, that energy in a moving body was proportional to velocity squared). Galileo was imprisoned. Émilie was excused because she was a woman and from there sprung her eccentricities. Her status was as Voltaire's mistress, not as a great scientist.

The mad genius is iconic in finance too.

I have heard people say - even now - that Carson was sloppy and breathless in his allegations. He was not. He was right. Dismissing him as sloppy and breathless is in itself sloppy and breathless. Indeed siding with conventional sounding language, conventional sounding reason lost you a lot of money in the Sino Forest case. It was better to side with Carson - the seeming madman.

This happens again and again and again. If you have very non-conventional views - even if they are well supported - even if they are right - if you spell them out plainly you will be thought of as a madman because your views sound outrageous.

But a hedge fund manager has it easier than the scientist. The scientist would be ostracized for the views. A hedge fund manager just makes lots of money. If you are right and nobody believes you then there is a chance to make mega-spondulick$. Indeed by far the strongest opportunities for making really good returns come from doing something original. I shudder to think how much Carson and his backers made on Sino Forest but I think we could safely measure it in lifetimes of average earnings.

There is however a problem with this: a portfolio manager who sounds mad may indeed be mad. Even if they are not mad the market may be wrong longer than they can be solvent. Which brings me to the schizoid character of a really good money manager or money manager team. It is highly rational but it appears like controlled insanity. You need the slightly crazy idea-driven maniac who is prepared to think outside the box and you need someone to button them down, to make them appear reasonable and to act reasonable.

Sometimes the two people are in the same body (I know a few of these people and some are not pleasant people because they are so strangely strung). In Bronte's case they are in different bodies with my business partner appearing much more socially, intellectually and even politically conservative than me. We are an odd-couple rather than an odd person.

All this leads me to wondering about Warren Buffett - the greatest fund manager of them all. He is an unusual individual with strange personal predilections. But he sounds so rational. And he is amazingly self-controlled. Indeed his ability to sit on billions of dollars excess cash for years and years at a time waiting for the time to pounce is legendary and extraordinarily hard to duplicate. Buffett says that temperament is more important than brains in investing and I think this is what he means. But it is often Buffett against the world. Buffett is thought of as a has-been by many people. It doesn't matter whether he was avoiding tech stocks in 1998, 1999 and 2000. It does not matter whether he was buying Bank of America now. He is - of course - just mad.

The madness is an important part of how Buffett made all that money. But the self control is - I think - more important.

When we meet clients I do all the talking. Some people have even wondered why I am with Simon (my business partner). We fundamentally have nothing in common. But there is a reason... he is the best control mechanism I have ever found. It took me a long time (and some losses) to realize I even needed one - he has the temperament. But if the career of Warren Buffett is a guide (and I think it is) then to make money over very long periods you need an ability to think right outside the box and a lot of self-control. The self-control is what most people think of as the risk management.

I don't know how you judge the yin-and-yang of this - and I don't know how you pick the individuals or the teams that have it. But when you see two guys who present themselves as partners but one is voluble and one is quiet then make your assessments of the quiet one. Our record at Bronte is fabulous: if it stays fabulous (and I think it will) then it will be control of Simon rather than idea generation of me that keeps it that way.


PS. I often run my blog posts through Simon. This one is in praise of the mad-guy (Carson) and the sane-guy (Simon). I think they should just see it when it pops up.

PPS. The best fund managers hire the misfits who work hard, don't belong in suits and often put them in suits to make them seem more reasonable. (We skip the suits.) This famous letter to a job-applicant by Dan Loeb is a gem.

PPPS. John Paulson is a mad genius. His problems at the moment are as much as anything a function of risk control. His positions were too large and when he was wrong (and everyone is wrong sometimes) his wind-down was horrible. He should hire Simon but I am not going to let him!


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.