Hollysys - a blog post inspired by the railway crash

Disclosure for people who do not get my sense of humor: We are short Hollysys. 
If you get to the end of the post and do not understand why read the note on the blog as to humor - or just read this.


----------------------------------


Hollysys is a Chinese rail technology company listed in the USA (NASDAQ:HOLI).

The stock is having a rough trading day today. As I write the stock is down about 20 percent. My first reaction was the Chinese rail crash but Hollysys has denied its signalling is involved.

That is a relief because Hollysys is an amazing company. And because it did not cause the railway accident it should be able to sweep the board of its inferior competition.

How to approach Hollysys

I approach Hollysys like I approach many amazing Chinese companies - businesses with amazing technology, amazing growth and amazing margins.

Some of the amazing claims made have turned out to be false - but some are yet to be falsified and some of those might actually be true. After all 1.3 billion people should be able to some amazing things.

Hollysys however is possibly the most amazing company I have seen yet.

As this blog is a fan of Fox News I wish to be fair-and-balanced. I report. You can decide.

What Hollysys does?

Hollysys makes control systems. It started in the railways industry and specializes in very fast trains. Here is a description from their website:

HollySys Automation Technologies is a leading provider of automation and control technologies and applications in China that enables its diversified industry and utility customers to improve operating safety, reliability, and efficiency. Founded in 1993, HollySys has approximately 2,400 employees with 9 sales centers and 13 service centers in 21 cities in China and serves over 1700 customers in the industrial, railway, subway & nuclear industries. Its proprietary technologies are applied in product lines including Distributed Control System (DCS) and Programmable Logic Controller (PLC), high-speed railway Train Control Center (TCC) and Automatic Train Protection (ATP), subway supervisory and control platform (SCADA), and nuclear conventional island automation and control products. HollySys is the largest SCADA systems supplier to China's subway automation market, and is the only certified domestic automation control systems provider to the nuclear industry in China. HollySys is also one of only five automation control systems and products providers approved by China's Ministry of Railways in the 200km to 250km high-speed rail segment, and is one of only two automation control systems and products providers approved in the 300km to 350km high-speed rail segment.

HollySys has achieved a wealth load of automation and control experience in the multiple business sectors such as Electric Power, Renewable Energy, Chemical, Petrochemical, Cement Processing, Mining and Metal Processing, Environment Protection, Pulps and Papers, Factory Machinery and Manufacturing, Railway and Subway, and Nuclear Power Instrumentation and Control. For the past 16 years, the HollySys name has always been standing as a symbol of quality and reliability in China. HollySys offers a wide range of automation and control products from PLCs to DCS to help you find the solution for any automation and process control application. Through our sustainable business and continuous development in the area of automation, we are able to provide users with a complete and total automation solution for the industries process sectors.

Our corporate mission is to be one of the major platform providers in Automation and Control with innovative technological products and solution back by the highest quality of services and support. We hope to utilize our expertise in the arena of automation technology for better work, life, and environment. We are now the global supplier of choice for innovative technology backed by the highest level of service and support. When you need products and solutions you can rely on, choose HollySys.

Their website then goes through various businesses that they run and describes their achievements. The achievements are on par with global players in the control-automation business such as Siemens. For instance on nuclear power this is what they say:

Since 1995, HollySys starts providing solutions for the nuclear power plant automation in China and since then, we have been continuously providing digital I&C system for the nuclear power plant. Through hardship of innovation and development, HollySys has now self-developed products and solution for the nuclear power plant control system. With more than 10 years of experience and over 40 successful NPP projects, HollySys has both the ability to become very competitive product suppliers and service providers.

• In 1997, HollySys developed China’s first computer control system for the nuclear power plant. The system was exported to Pakistan Chashma 300MW NPP and has been successfully operating ever since.

• In 1998, Qingshan Phase II NPP adopted HollySys DCS system. It is equip with centralized supervision and safety control of the nuclear power plant with system of conventional island design.

• In 2004, HollySys was the first to enter the field of nuclear safety testing and successfully implemented the project of digital safety testing devices.

• In 2005, HollySys became the localized sub-contractor of Ling’Ao Nuclear Power Plant phase II, and undertaking the project implementation for digital I&C system.

• In 2007, HollySys has successfully developed its first generation totally digital I&C system, the HOLLiAS NMS system.

By far, HollySys has 44 signed and completed contracts within our nuclear power business unit covering almost all of the China’s Nuclear Power Plant either in construction or in operation. These have put us in the top league in competition among other competitors in China. This vast achievement wins recognition from our customers, at the same time, have proven the reliability and safety of our solution, and the professionalism in our engineering services that we provided.

They make similarly large claims about their business in other sectors such as control systems for subway automation or factory automation.

One long investor I know visited the plants and was shown a very impressive company. There headquarters photographed on the website - is very impressive.

But as regular readers of this blog know I am a balance sheet kind of guy. When I see such amazing businesses built out of nothing I want to know what assets are employed and how much cumulative R&D there has been. After all HollySys looks like a serious threat to Siemens on its claimed businesses.

Here is the balance sheet from the last annual filing (on form 20F). Numbers are in US dollars and the balance dates are June 30:

	2009				2010
ASSETS
Current Assets
Cash and cash equivalents		$	128,882,666			$	119,501,945
Contract commitment deposit in banks			5,504,375				4,383,684
Accounts receivable, net of allowance for doubtful accounts of $6,276,670 and $8,408,318			56,548,509				64,384,519
Cost and estimated earnings in excess of billings, net of allowance for doubtful accounts of $744,113 and $1,102,016 (note 5)			51,094,660				60,928,056
Other receivables, net of allowance for doubtful accounts of $178,532 and $214,789			4,148,842				4,102,136
Advance to suppliers			7,867,856				10,676,175
Amount due from related parties (note 18)			7,203,058				10,764,828
Inventories, net of provision of $1,114,140 and $2,393,546 (note 4)			18,837,270				23,554,331
Prepaid expenses			1,368,918				1,022,803
Income tax recoverable			-				1,083,640
Deferred tax assets (note 16)			319,737				956,969
Deposit for acquisition of equity interest from minority interest			2,195,582				-
Total current assets			283,971,473				301,359,086
Property, plant and equipment, net (note 6)			47,102,749				65,345,618
Long term investments (note 7)			13,570,578				17,348,159
Long term deferred expenses			91,779				-
Deferred tax assets (note 16)			706,943				677,388
Total assets		$	345,443,522			$	384,730,251
LIABILITIES AND STOCKHOLDERS’ EQUITY
Current liabilities
Short-term bank loans (note 9)		$	5,854,887			$	1,472,559
Current portion of long-term bank loans (note 12)			5,123,026				1,472,559
Current portion of long-term bonds payable (note 11)			-				11,780,471
Accounts payable			37,421,717				41,479,662
Construction cost payable			10,929,116				12,562,565
Deferred revenue			21,072,540				33,552,968
Accrued payroll and related expense			4,162,851				4,386,681
Income tax payable			1,397,706				5,971,136
Warranty liabilities (note 8)			1,631,407				1,916,654
Other tax payables			9,152,197				10,632,611
Accrued liabilities			2,634,107				8,078,783
Amounts due to related parties (note 18)			1,464,683				2,610,599
Deferred tax liabilities (note 16)			277,337				-
Total current liabilities			101,121,574				135,917,248
Long-term b ank loans (note 12)			36,593,041				35,341,413
Long-term bonds payable (note 11)			11,709,773				-
Total liabilities			149,424,388				171,258,661

$

345,443,522

$

384,730,251

Well lets piece this apart. The company claims $65.3 million in property plant and equipment (PP&E) and it directs us to Note 6. Note 6 breaks the PP&E down further:

A summary of property, plant and equipment at cost is as follows:

		June 30,
		2009				2010
Land use right		$	7,238,446			$	7,282,148
Buildings			13,520,841				13,608,227
Machinery			2,873,789				3,118,451
Software			797,327				1,206,719
Vehicles and other equipment			8,008,915				12,005,119
Construction in progress			23,343,595				39,329,699
		$	55,782,913			$	76,550,363
Less: Accumulated depreciation and amortization			(8,680,164	)			(11,204,745	)
		$	47,102,749			$	65,345,618

Construction in progress consists of capital expenditures and capitalized interest charges relating to the construction of facilities and assembly lines projects. Interest of nil, $206,595 and $1,102,772 during the period of construction for the years ended June 30, 2008, 2009 and 2010 respectively have been capitalized.

During the year ended June 30, 2009, the Company commenced the construction of a new facility with building area of about 150,000 square meters.  As of June 30, 2010, the construction in progress of the new facility was $39 million.   The Company completed this new facility and related construction project in August 2010, and the whole construction cost amounted to approximately $44 million including the cost of land use right of $5 million which had been acquired during the year ended June 30, 2008.  

The depreciation and amortization for the years ended June 30, 2008, 2009 and 2010 were $1,817,657, $2,241,344  and $ 2,683,042 respectively.

We learn that the 65.3 million in net PP&E is $76.6 million gross PP&E - the difference being accumulated depreciation. Of that 76.6 million most is buildings and land use rights. There 13.6 million in buildings, 7.3 in land use rights and another 39.3 million under construction.

Only $3.1 million of machinery is in use. That is kind of funny because this company is a manufacturer of some pretty high end kit. Indeed it is pretty hard to imagine how you manufacture the control systems (sensors etc and the computers to go with it) with only $3.1 million in kit - but the company seems to do so. Those machines should also include all the computers on which staff make the software to run all this fancy kit. The company has $3000 of kit - so there is less than a laptop per staff member in machines - and surely all of this manufacturing requires some machines.

Whatever: this is a miracle of machine efficiency - as amazing as any other Chinese company I have written on.

The rest of the balance sheet is perverse. The company had revenue in 2009 and 2010 of $157.5 million and $174.1 million respectively. Costs estimated in excess of billings are over $60 million. In other words they have done a third of year's work and not been paid. Net income last year was $27 million. Over the past three years net income is about $20 million. The company has many years income in work which it has done but has not yet got around to billion.

That is perverse. You work and work and work and it is not cash - it is just unbilled receivables. They better be sure they can bill somebody and that they will collect it in cash.

If you look at the last quarterly the costs estimated in excess of billings have ballooned to $86 million. They continue working. They do not collect in cash.

Another 10 million - or half the profits for the last three years - is advanced to suppliers. That number is rising too.

There is 23 million of inventories too. I guess that makes sense - they are a manufacturer - but there are next to no machines to work on these inventories with...

Admittedly their liabilities are also larger than would appear normal - there is for instance 33 million in deferred revenue. That is also large relative to profit. I guess working capital management is just different in China...

Revenue in the last nine months is skyrocketing. Its up over 60 percent. Property plant and equipment however is almost unchanged. This company is one of the most advanced manufacturers in the world - and as far as I can see it does this almost without machines.

Reasons to be long this company

About six months ago I was talking to a (former) long in this company and he told me that I should be wary shorting it. It was very close to the Railway department and the Railway Minister. (My friend just assumed I would be short - but truly I am looking for companies to go long...)

Hollysys he said had lots of Guanxi - and that Guanxi might help them make money. He thought it even possible that they had obtained machinery and know how from the Railway Department without having to pay for it.

He really did think that there was a relationship with the Railways Minister that would play well for the company. And not being that familiar with the innards of the Chinese Politburo I was not one to quibble.

The stock however got weaker when the Railway Minister was arrested for corruption.

R&D

This company has done about 25 million cumulative R&D in the past three years. This is a lot by Chinese standards but it is tiny by the standards of Siemens. Given the enormous advances that this company has (claimed to have) made in fast train and nuclear power control systems I am getting worried about nuclear reactors controlled on the cheap. Still I report: you decide.

Audit

This company is a June balance date. The auditor is BDO Hong Kong. This is an auditor stung a little by scandal lately but then so has every auditor in China.

This one is amazing. Utterly amazing. It might trip up on audit (which is not far away). But this might actually be the real deal - the company with world-class technology on a modest cumulative R&D budget. The company with world class manufacturing where the manufacturing requires next to no machines. The company that does work greater than its accumulated profit and neither expenses nor bills that work. The company that breaks all the rules and still succeeds.

The company that proves that this time it is different in China.

And imagine - because they did not cause the railway accident they are going to be getting a lot more work.

John

Competition and managing money

I had a long chat the other day with a long-only global fund manager. He observed (correctly I think) that conditions for private-equity takeouts of companies are as good as they have ever been. Debt is very cheap (low interest rate loans with weak covenants seem readily available) and many private equity funds have plenty of money which they have to spend or give back to their investors.

He was putting together a basket of stocks which he thought private equity would buy.

I tried thinking this way and whatever basket I came up with contained some very unattractive stocks.

Private equity has been doing some strange deals and indeed we are reversing several private equity takeovers in the hope (probably vain hope) that the private equity firm will come to its senses and not close the deal. Private equity firms are buying frauds at an unprecedented rate - which is not much fun if you are short the said fraud. (And it is not just in China.)

But for the last 18 months being long a bunch of things that might get taken private has been a winning trade - those stocks have appreciated irrespective of whether they actually caught a bid.

And being long a bunch of things that have no hope of being taken private (eg large cap techs, family controlled German companies with dual voting structures) has under performed.

And I have been (at least on the long side) with the under-performers. But it feels safer to me...

A fundamental difference in the view of how you make money in markets...

Underlying this is a fundamental difference in view about how you make money in markets. I always had embedded - deep in my psyche - the notion that the best place to look for investments is where nobody else was looking. You wanted to buy what was unfashionable but better than current fashion suggests - straw hats in winter because summer would come - and in that case you would only need to wait six months. The best response to competition was to avoid it rather than compete with it.

My friend's basket of stocks contained only businesses that the new power on the street (private equity) was looking at.

My friend wanted to buy what was going up in price - and what was going up in price was what the private equity firms (who were loaded) were going to purchase next.

My friend's attitude to competition is buy what they want to buy and sell it to them later at a higher price.

And so far he has been right.

And I have been wrong.

The new "value" is in stocks that are too big or too difficult for Private Equity. Microsoft and Cisco both have cash adjusted PEs well below 10. They are too big. Target and Walmart are about 12 times earnings. Same story. The German family concerns could also be purchased at low levels. Coca Cola - the definitive Warren Buffett stock is back at 12-13 times earnings for the first time since Warren Buffett originally purchased it.

And frankly nobody is interested - these stocks under perform.

And if you buy something that is cheap you might find even more problems come out of the woodwork (try Yahoo and Alibaba for a recent high profile example).

Many of the great value orientated investors are having a bad time this year. If it were not for our shorts we would have not had a fabulous time either. I don't think the value guys are wrong here - but it has not been fun - and it sure is getting frustrating.

Maybe out performance will be if the market turns really sour and the value stocks only drop 15 percent whilst the stuff private equity never got around to buying drops 45 percent. But that is hardly the fun form of out performance.

We have had very strong performance and not done great on longs because our shorts have been so accurate. But many of the best investors I know are having a hard time - and I keep imagining and hoping their (and their client) suffering is temporary. Because I don't think I am going to buy my friend's long stock basket. Too hairy for me.

Meanwhile I am becoming totally reliant on the short book for out performance. I wish I had more arrows in the quiver.

For comment.



John

Today a single link about China (and Apple)

This post is a gem:

http://birdabroad.wordpress.com/2011/07/20/are-you-listening-steve-jobs/



John

PS. I have had extensive comments (especially from journalists) about the last post.

I am a left-of-center Australian who has lived with Rupert Murdoch far longer than most Americans. If you could not discern that the last line of the last post (about News Corp being the only remaining ethical newspaper company) was humor then Americans are even less able to detect dry humor than I thought. However there was an important point made which is that most journalists - including those that are beating up on Rupert Murdoch - would have a big problem with a proprietor who inquired into how they obtained certain stories.

Rupert Murdoch was always the most interfering of proprietors. That was (correctly I think) the main criticism most journalists had of him. But now they have a different standard: he should be actively aware of how his journalists are sourcing the stories to ensure that illegal methods are not used. That I think is having it both ways - and they want it both ways because (I think for good reason) they actively despise Rupert Murdoch.

For comment.

J

When you talk to a journalist off the record how safe are you?

There is a long history of ethical journalists going to extreme lengths to protect their sources. Many a good journalist has suffered prison for contempt rather than reveal their sources.

In the Watergate scandal the editor knew the source of the story but the owner (Katie Graham) did not.

And Katie Graham would never have asked. She may have asked the editor if he was sure of his story but she would never have asked who "Deep Throat" was. As owner she appoints her editors and on really important matters has to trust them.

Ok - now look at News Corp.  The News of the World ran lots of fantastic stories. They were accurate. We now know how they were (criminally) obtained.

The journalist most certainly knew. The editor almost certainly asked and knew. But the proprietor would not normally know and it would be untoward of him to ask.

But now we have the journalist profession - led by Felix Salmon whose judgement I am more and more questioning - absolutely sure that Rupert and James Murdoch knew and are personally responsible for the hacking scandal.

It seems the journalist profession has forgotten all about the ethics of journalists. It would be inappropriate for Rupert and James Murdoch to be told the source if the journalist wanted it protected and it would be inappropriate for Rupert and James Murdoch to ask.

Of course a large slab of the journalist profession has forgotten all about journalist ethics and just assumed that the proprietor should and does know.

Of course they have. These journalists are themselves non-ethical. They have forgotten journalist ethics.

When you talk to a journalist off-the-record you have - in the past - had a reasonable basis for presuming that the off-the-record conversation will be respected. You can't assume that anymore. Journalists have forgotten the basic ethics of the profession.

There is a loss to society in the Rupert and James Murdoch scandal - but the bigger loss is not the News of the World or News Corp. The bigger loss is journalists who have made themselves into scumbags with no more ethics than the average blogger.

There is however one remaining ethical company - one where you can be assured that if you talk off-the-record the proprietor will remain ignorant.

The last ethical newspaper company is News Corp. Do not talk off-the-record to any other company.



John

Why Meridith Whitney's concerns do not matter

Meridith Whitney has made a fuss of late pointing out the (more or less obvious) insolvency of many State and local governments.

She seems to think this is the end of the world.

Maybe there will be a lot of defaults though if something like California defaults it will be by choice rather than by force. (California is plenty rich and has plenty of taxing power if they don't want to give it away in citizens initiated referendums...)

But that is irrelevant to the point I wish to make.

Ultimately I don't care if there is a wave of municipal defaults. I think it will have next to no economic effect. Here is why.

I like to draw a distinction between money lost by unlevered players and money lost by levered players.

Plenty of wealth was destroyed in the dot-com bubble. But it caused no systematic crisis because the wealth was lost by individuals and unlevered players. If you lost money in dot-com you probably spent less (there was a recession) but it did not destroy the system.

However when money is lost by levered players it leads to cascade effects. 50 billion lost by Lehman Brothers (not more than 200 billion anyway) had huge effects because Lehman was a levered entity that had its claws everywhere. It had widespread destabilizing effects - whereas a similar amount lost in dot-com like events has next to no effect (except on the losers).

Well muni-bonds are ultimately held by unlevered individuals usually for their tax advantage. They are not held by people with large mortgages or credit card debt and they are mostly not held by financial institutions.

In other words their default will have economic effects similar to dot-com rather than similar to Lehman.

And in fact they are held by the people with the most excess saving and the lowest marginal propensity to spend incremental income - so their default will have low wealth effects.

It won't matter. Meridith Whitney might be right - or she might be wrong - but unless you hold muni-bonds I can't see why you should care.

Summary

I don't think you are going to find a Goldman strategist who will say that muni-bonds are owned by rich people and you can f--k rich people and they won't spend any less and hence muni-default does not matter.

Nah - nobody from The Squid will say that.

So I will. Somebody has to.


John

China frauds: In (partial) defense of the auditors

The most common question I get asked about Chinese frauds by journalists (and even more by class action lawyers) is where were the auditors?

The class action lawyers are particularly interested in that because it is going to be very hard for instance to collect from China Media Express or Longtop (both probably overwhelmingly fraudulent) but it will be relatively easy to collect from Deloitte who audited both. (That presumes that liability can be attached to the auditor...)

Deloitte audited Longtop for six years giving a clean bill each time before the amazing auditor resignation announcement. Surely - or so the journos and class action lawyers say - they were negligent in their previous (clean) audit statements?

Starr Asia (Hank Greenberg's vehicle) is suing Deloitte for their audit of China Media Express.

The class action lawyers (and presumably Hank Greenberg) will not like my answer. I have told the class action lawyers that they can put me on the stand as a witness if they want - but my testimony will be supportive of Deloitte. That answer surprises them. But here goes:

Imagine you were auditing China Media Express. This is a company that claimed to have video screens showing adverts on twenty thousand buses. Buses by their nature (they move) would be scattered all over China. You can't physically audit the buses.

They sell adverts to maybe 15 advertising agencies. The company gives you - the auditor - contacts at all of those agencies. You ring the contacts. They confirm the contracts, the receivables and the like.

None of this would have spotted the fraud. After all the buses are impossible to audit and the particular advertising agencies are say 15 out of 3000 in China.

The way that you would really conduct this audit is match the business against bank statements. The company claimed roughly 30 million dollars per quarter of profit. An auditor verifies a sample of transactions and verifies the total. Finally they verify the cash balance.

If the bank statements contained verification of all your sampled transactions and in aggregate showed 300 thousand flowing into the accounts per day and 170 million in accumulated cash then you would have verified the business.

To be thorough the auditor would normally get the bank statements from the bank and not from the company. After all the company could provide you fake statements - it is unlikely the bank would.

Audit is a process. If you (a) sample the requisite number of transactions and (b) verify the key totals with creditors and banks you have done your duty. Indeed provided you actually verified the bank statements with a major (and presumably reputable) bank then you have met the audit standard.

The shock of Longtop and China Media Express is that the banks appear to have been in on what appears to be fraud. In both cases Deloitte went to the bank head office (not the local branch) and double-checked the bank statements. And in both cases this caused the apparent fraud to come crashing down.

The critical revelation is that the bank was in on the scam. If the bank is in on the scam it is possible for the auditor to conduct all the standard tests and do all their duty and sign-off completely dodgy accounts in good faith.

And when a fraud is exposed it is possible that no liability accrues to the auditor.

I can point to (several) examples where I think auditors are culpable. I can think of several audit firms that should cease to exist when the China frauds are all exposed. But those individual cases are not the rule.




John

PS. If this level of corruption is pervasive in Chinese banks then we are all looking in the wrong place for the next crisis. The next crisis comes from China.

Helping Charles Li find the weeds in his Hong Kong garden

Charles Li, the CEO of the Hong Kong stock exchange, has been on CNBC and again at home telling the world that the scandals happening with Chinese stocks on the American stock exchanges won't happen in Hong Kong.

Compared to the Reverse Mergers on the American Stock exchanges the Hong Kong exchange is indeed a beautiful garden with many flowers and some fruit trees.

But there are some weeds there - indeed many weeds there. Some have grown so high they are strangling the light.

And a diligent gardener does not ignore the weeds lest they starve the flowers and fruit-bearing plants.

So I hope to help Charles Li a little. He is blinded by the bright and beautiful flowers of China that litter his garden (ahem exchange). He can't see the weeds no matter how tall, how prominent.

Somebody has to help him so he can continue to tend his garden and so that it might continue to flower into the future.

So I am going to focus a little on Hong Kong in the future: a service to Mr Charles Li and thus a service to Hong Kong.

But hey - I figure there are far more weeds in Hong Kong than the ones I have found (though these weeds are very tall - they almost grow to the sky).

If readers want to help me find the weeds in Mr Li's garden please contact me through the blog.



John

The Paulson Sino Forest loss

Much has been made of the John Paulson's Sino Forest loss. Many snide remarks have been made parsing John Paulson's letter.

I am more sympathetic than the non-practitioners though I think Paulson's excuses sound lame.

Here is my sympathetic view:

There are two approaches to running a funds management firm. One approach is to have analyst teams covering every sector and to have depth of coverage. The other approach is a small and more flexible team.  Each of these approaches has its problems.

I got a good idea of the problems of the broad coverage approach when PMI (the mortgage insurance company) double-booked an investor relations meeting. I sat in on a meeting between PMI and the Fidelity analyst who covered the sector and watched the incredibly nuanced discussion the Fidelity guy had about mortgage insurance. The Fidelity guy could make detailed distinctions between mortgage insurers and could tell you which ones he would prefer owning.

This was 2005.

Of course the right investment decision for mortgage insurers in 2005 was to own none of them. A year later the right answer was to short all of them.

But because Fidelity had a guy who knew lots about mortgage insurers they owned mortgage insurers (probably at near market weighting) and they duly lost money.

A small team would either decide hey - they want to be part of this high growth mortgage market and own them - or decide the did not want to be part and not own them. If they were aggressive they might short them. There was a reasonable chance that the small team was going to do better than the big team because it was not the nuances between mortgage insurers that determined the outcome.

Generally I think the small-team works better. I came from a shop with a great record and eighteen investment staff. Most the good ideas came from nine of those staff. They were better than large teams - and every time we increased the size of the team we diluted those staff. The main reason for increasing the size of the team was that the new staff - whilst they did not contribute much - were being trained. And after a four or five years one in say five of them might join the anointed ones.  And as we had a good name the others might find worthwhile jobs elsewhere...

But there are problems with the small teams. First small teams have blind spots. The world is a big place (too big to paint anyway) and there is no way that the senior fund manager can cover everything. There are large specialized stock areas about which I know nothing (eg most German industrial companies, US Health Care, South American resource companies). Things just pass me by. A really good nine person team has less blind spots than Bronte (three people) but blind spots are a part of life.

Secondly we use shortcuts. In the pre-internet days I used to "know" for instance that monopoly town newspapers earned 35 percent EBIT margins and trade at 3.5-4 times sales. The dominant competitive newspaper in a duopoly town earned 20 percent EBIT margins and traded at 2-2.5 times sales. When someone came to me with a newspaper stock (a sector about which I knew plenty) I used to ask about the competitive position of the papers (which towns were monopolies, which duopolies etc) and the margins, do the sums quickly in my head and decide whether it was good value. Beyond aggregate numbers I probably did not open the accounts. A portfolio manager with a diversified portfolio makes a lot of decisions and they take shortcuts. The shortcuts are a necessary part of life - and they are how you apportion and guard your time.

I am an expert on fraud - but it is entirely possible that I would miss a well executed fraud at a newspaper company because my starting assumption was that I could take the numbers as gospel.

The shortcuts we use as portfolio managers (and these are often sophisticated shortcuts born from some deep understanding of the industry) can make even the best portfolio manager susceptible to fraud.

Moreover a portfolio manager needs to work out how to ration their time. They can't do in-depth analysis of every stock in a 20 stock portfolio. (Well they can - but most the in-depth analysis you do are of stocks not in the portfolio as you need to decide what to invest in rather than becoming more expert on the things you own.) So you tend to spend your time on things that you perceive pose the most risk or the best returns.

Timberland is famously low risk. There is a German charitable trust that has been around since the 16th Century living off a 500 year old bequest. It owns timberland. Timberland - along with unlevered land below large buildings in major cities - is one of the few assets that seems to hold its value and earn returns for centuries. (Gold may hold value but earns no returns.)

I can understand how a deeply bearish investor (and Paulson was a deeply bearish investor) got suckered into Sino Forest. They saw it as timberland. They did not do the due diligence that was required to spot the fraud. It had a respectable board and reputable auditors and a long history. It did not feel like a fraud.

Chinese stocks present massive fraud risk but trees are low risk.

Paulson didn't see China for the trees.

And guess what - it happens to all of us. Fraud is pervasive in financial markets - its just that nobody talks about it. When they are had they are more-than-likely going to blame it on bad investment decisions than fraud. Moreover it is embarrasing - even business threatening - for professional investors to admit they were victims of fraud - so they don't talk about it even when they know they were had.

Fraudsters themselves are rarely prosecuted because it is hard to distinguish business misadventure from fraud. Sure the business failed. Sure the CEO sold $20 million worth of stock before it failed - but the business failed for honest reasons - or so he tells the jury. It doesn't matter that the business involved mass fraud - unless the fraud is huge the CEO is going to get away with it. And even if it is huge he will probably get away with it. (In the crisis it is hard to tell the deluded promoter of bad paper from the fraudulent promoter of bad paper. In my view most were deluded, not fraudulent, but fraud was still common enough.)

Paulson's mistake on Sino Forest then wasn't "amateur hour". Rather it was a mistake a competent fund manager might make because he is blindsided. Because he uses his shortcuts (trees are safe for instance) rather than does detailed analysis of every stock.

Every fund manager makes mistakes. Its part of the game. I try to minimize them - but I have made some real beauties. I purchased Washington Mutual preference shares for instance - and I blogged about it.

Paulson's excuses

Whilst I am sympathetic to Paulson's mistake I am more concerned about his excuses.

The first Pauslon excuse is the one I have most sympathy for - which is that he estimates the loss based on the cost base of the position rather than its valuation before Muddy Waters comes along. We have a few positions like that.

Take our well-publicized position long the preference shares of Freddie Mac. I blogged about them in a ten part series (see here for Part 1). The preference shares are highly speculative - the money owed to the government needs to be repaid before the preference shares will realize anything. But they were trading at below two cents in the dollar and I thought there was a reasonable chance they would pay and hence were worth a bet.

I think there is an improved chance they will repay now (note falling delinquency for example). They are trading about 10c in the dollar. They are still however a wildly speculative bet which I think is worth making.

If they were to go to zero how much have we lost? Is it the original amount or the high water mark? Should I trim the position just because it it has appreciated? Does it matter that the position is being diluted by new clients anyway? Do the losses feel worse for a new client (who is implicitly buying in at 10c in the dollar rather than 2c in the dollar)?

I do not have a consistent answer to these questions. I am somewhat sympathetic to Paulson's answer because I do not think there is a consistently good answer anyway.

Where I am less sympathetic is to Paulson's statements that staff went to see the operations (and hence they judged they were real) and also to the line that they did a thorough review of the financial statements.

If you go see Sino Forest's operations you will see what Sino Forest wants to show you. They will show you trees. You can't tell whether that is 5 thousand hectares or 500 thousand hectares. Seeing trees does not answer the question. There is no point looking at things that are not going to tell you anything anyway - and so Paulson's staff member wasted his time looking. That is an amateur-hour mistake.

If you are going to look at the operations (and it is often worthwhile) then do the work properly and look through the eyes of a competitor or a customer or a supplier. And find them yourself rather than talk to sympathetic ones supplied by the management.

When management say good things about themselves that provides no actionable investment information. When management say good things about a competitor that is golden. When suppliers you have found yourself say good things about a company that is useful. When management say bad things about their business that is useful.

Speaking to management and hearing good things about them said by them does not help in investment and hence does not constitute actionable analysis. Using that sort of analysis as an excuse is pathetic and Paulson (a man who made his fortune betting against the conventional wisdom) knows that perfectly well.

As for analysis of the accounts - the Sino Forest accounts contain enough red-flags to make any eagle-eyed observer cautious. I am sympathetic to making an investment without looking at the accounts at all because limited time and shortcuts often make that an efficient way of behaving. What for instance would I learn about Microsoft by looking at the accounts that I do not already know? Microsoft will be a fabulous investment if it maintains its strong position for the next decade. It won't be otherwise. The answer is not in the accounts (but it might be in understanding the technology and the way people interact with the technology).

I would be sympathetic to a statement that Paulson did not even look at the accounts for Sino Forest beyond a cursory look because he may have had detailed short-cuts in his head for a forestry company. But if some analyst really did a detailed look at the accounts and did not spot the red-flags then they are incompetent. For that I have no sympathy at all.

For comment





John

PS. I will add one thing though. Sino-Forest accounts have become more plausible over time so the job of detecting problems is harder with (say) the 2009 accounts than the 1997 accounts.

In 1997 accounts the company was outright strange. It claims to have shipped 1.165 million bone dry metric tonnes of wood chips including exporting to Japan. (It did not claim to sell standing forest until many years later.)

However timber sales were only $23 million. This is not very much per tonne. You can find real prices at this link.

Sino Forest claims 603 thousand hectares of forest which it was phasing in (of which quite large numbers were in operation and quite a lot more was being planted).

But the balance sheet was thin. There was $9.3 million in machinery and equipment that was not being depreciated because it was "under construction". The only equipment being depreciated was $211 thousand in vehicles.

You can't move a million tonnes of woodchips with $211 thousand in vehicles.

Indeed Sino Forest had no equipment of any kind being depreciated except vehicles. This was a logging company with no chainsaws!

Those accounts were silly. The later ones are full of red-flags - but they are not so obviously silly.

I am more sympathetic to Paulson buying the stock in 2009 than I am to anyone who purchased the stock in 1999.

The person that purchased in 1999 probably sold some and made money. Indeed if they sold today they are probably still up fairly well.

Which reminds me of the saying among professional investors: "I would rather be lucky than smart". Someone who purchased Sino Forest in 1999 was not smart. But they sure were lucky.


J

Sino Forest: some ancient history

Sino Forest uses intermediaries to market all its timber (or so it says). It will not reveal the identity of those intermediaries for commercial reasons but says that it has revealed the identity to its auditors and to Price Waterhouse Coopers who are doing an independent review of the accounts.

By using intermediaries Sino Forest do not need to own all the equipment to harvest and transport the timber which explains away at least some of the problems Muddy Waters details. Intermediaries also explain why nobody can find the pulp and paper mills they sell to. (They do not see to pulp and paper mills, just to intermediaries.)

They also buy standing timber from "authorized intermediaries". Again they do not tell us who these intermediaries are.

This makes the company opaque. Sino Forest's response to both Muddy Waters and the Globe and Mail articles comes down to:

(a). This process is required in China.
(b). It is opaque, we need it to be opaque for commercial reasons, but
(c). It is legitimate and we can document it as required to our auditors and PWC.

The process is so opaque from the outside that the Globe and Mail (not a shortseller - just a newspaper) could - in Sino Forest's view - be honestly mistaken when they try to verify the facts on the ground in China. Sino Forest is too complicated and too opaque for investigative journalists.

The bear case against Sino Forest argues the intermediaries are fiction and the transactions are round-robins involving related parties. There is little timber. The sales are "fake" and generate fake receiveables. Over time the fake revenue (which would otherwise show as fake cash) is used to buy more fake forests (the only forests you can buy with fake cash are fake forests).

In the bear case the size of the fake forest estate grows over time just because the fake profits need to be deployed somewhere. Ultimately - in the bear case - Sino Forest becomes a fake giant and someone (in this case Muddy Waters) asks if you own so much timber how come we can't see it? It becomes "See No Forest".

In the extreme the only way to solve this problem is to somehow forget you own all those fake profits and start again with a lower land-ownership claim.

Bluntly the bear case is that if the intermediaries are fiction its likely the profits are fake and the forests are fake too. You can audit Sino Forest one of two ways: you can prove or disprove the story about the intermediaries or you can prove or disprove the rights Sino Forest have to harvest trees. They are the flip sides of the same coin. Fake intermediaries (or intermediaries who are undisclosed related parties) almost certainly means fake profits. And the only forests you can purchase with fake profits are fake forests.

Of course if the intermediaries are real non-related parties then the profits are real and the forests are probably real too.

If you decide the intermediaries are real non-related parties with real transactions then Sino Forest is a screaming buy. If not then Muddy Waters is right: the stock is going to zero as there are few forests and much debt.

There is a direct analogy in other Chinese frauds. Longtop claimed it had businesses that generated huge profits. Those profits became huge cash balances. Deloittes resigned when they thought the company (with their banks) were faking the cash balances. Whilst Deloittes never said so the conclusion is that if the cash was not there the profits that generated that cash were not there either.

Sino Forest argues the whole system is opaque for good Chinese commercial reasons. I have my doubts - but if you accept that there are good reasons for the opacity then you would need to check Sino's claims on the ground in China.

The first simple check

I suggested that if there really is 17.9 million cubic meters per annum of timber sales then someone needs to be pulping it and the first due diligence test is to find the chip mills - all of them - and sit there with a clicker counting trucks.  You need a lot of truck loads.

Its not a perfect test because you will not know whether the timber in the mill is actually coming from Sino Forest (directly or indirectly) but at least you will find out the aggregate harvest in the area. If the aggregates are below 17.9 million cubic meters per annum then "Toronto - we have a problem".

Short of going to China and having access to the intermediaries there are a few things we can do. One is just compare Sino's statements over time to the response to Muddy Waters allegations. That is not conclusive but a changing story detracts from Sino's credibility and increases Muddy Waters credibility. Conversely a story that is consistent about the nature of the business detracts from Muddy Waters credibility and increases Sino Forest's credibility.

For maximum contrast I have done this with ancient history - press releases from the very early days of Sino Forest.

History

Sino Forest has been listed a long time - and in the Muddy Waters account it was just a reverse-merger fraud that was a fraud from inception and just went on a lot longer and got a lot bigger than the perpetrators expected. In the process it purchased some forest for show - but the main purpose of the listing was not to raise money for forestry activities in China but to fleece Western investors.

If Muddy Waters is right then we might be able to tell by going back to the early press releases (say 1995-1999) and seeing what the company said then. The small China frauds are not very sophisticated (they do not have sophisticated stories) and a careful reading of the promotional material often gives the game away. I have successfully identified fraud in many without doing anything more than reading the promotional material. When Sino Forest was a small fraud (if it was a small fraud) it would not have required the complex intermediary structure - after all nobody much checks small companies. As it got bigger the story would have had to get more complex.

For this reason I want to repeat old press releases from Sino Forest. I keep my comments to a minimum.

I just want to return to the day of a simpler Sino Forest.

Hong Kong & Toronto: September 28, 1995 

Sino-Forest Corporation ("Sino-Forest" or the "Company" today announces the completion of a Private Placement of 2 million Class A Subordinate-Voting Shares at a price of Cdn$0.60 per share. The proceeds will be used for general working capital requirements. 

Sino-Forest's forestry plantation business has been expanded to cover over 600,000 hectares in three provinces of southern China during 1995. Through its majority control of joint ventures with local provincial and municipal forestry bureaus, Sino-Forest processes and exports hardwood chips to pulp producers in Japan, Taiwan and Korea. Chip shipments are expected to total about 250,000 metric tones in 1995, compared to 156,000 tones in 1994. 

In recent months, the Company has had discussions with several North American forestry companies relating to the establishment of further joint ventures in China for the manufacture of "value added" products such as OSB and MDF. The fiber feedstock for the plants would be supplied from Sino-Forest's plantations, which have the potential to increase the Company's annual fiber supply to 9 million tones over the next 10 years. This volume represents approximately 14% of the Province of British Columbia's current annual cut. 

Sino-Forest has recently negotiated trade finance arrangements with three major banks in Hong Kong, which will support the projected sales expansion over the next two years. The Toronto Stock Exchange has conditionally approved the listing of the Company's Class A Subordinate-Voting Shares.

Interestingly back then Sino Forest claimed to export to pulp producers in Japan, Taiwan and Korea. They used bank-provided trade finance to facilitate this. There was no mention of intermediaries. They also claimed 600 thousand hectares. This compares to this graph in the last annual report:

In 2007 Sino Forest only claimed 300 thousand hectares with almost $1.5 billion in acquisition cost. Sino's (trivially small) 1995 balance sheet allowed them to claim 600 thousand hectares at almost no cost.

They did however think that their 600 thousand hectares might yield 9 million tonnes ten years later (in 2005). That is consistent with - but lower than - the yields they are claiming now.

On 22 October 1998 Sino Forest released a long press release with this opening paragraph:

SINO-FOREST CORPORATION ENTERS INTO AN AGENCY AGREEMENT WITH SHANGHAI JIN XIANG TIMBER LTD.

Sino-Forest Corporation announces that it has entered into an Agency Agreement with Shanghai Jin  Xiang Timber Ltd. ("SJXT"), a company in which Sino-Forest holds a 20% equity interest, to supply various wood products to the National Timber Sub-Markets in the PRC. Under the agreement, Sino-Forest will act as agent on behalf of SJXT to purchase 130,000m3 of various wood products over an 18 month period. Based upon current market prices, it is expected that this volume of wood products will have a sales value of approximately U.S.$40 million. Sino-Forest will earn commission income in respect of the purchases for SJXT.

What is interesting is that they were quite happy to name their business partners then. (This is in contrast to their current stated position.) Secondly Sino Forest were buying timber for Shanghai Jin Xiang (and earning commissions), not the reverse. This is a reversal of the current intermediary business model.

Obviously this made me want to look for Shanghai Jin Xiang Timber Ltd. This was also revealing.

This was a reference in the 1999 annual report:

There are also promising growth opportunities as Sino-Forest’s investment in Shanghai Jin Xiang Timber Ltd. (SJXT or the Shanghai Timber Market), develops. The Company also continues to explore opportunities to establish and reinforce ties with other international forestry companies and to bring our e-commerce technology into operation. 

Sino-Forest’s investment in the Shanghai Timber Market — the first national forest products submarket in eastern China — has provided a strong foundation for the Company’s lumber and wood products trading business.

They were developing a business in the Shanghai Timber Market but they were using SJXT as an intermediary. They also had e-commerce technology (though it is not explained what that does).

By the 2000 annual report the investment in Shangahi Jin Xiang Timber had expanded to a 34.4 percent equity interest.

The Company has a 34.4% equity interest in Shanghai Jin Xiang Timber Ltd. (“SJXT”), an equity joint venture (“EJV”) that was formed by the Ministry of Forestry in China. The purpose of the investment is to establish strategic partnerships with key local wood product suppliers and to build a strong distribution network for the lumber and wood products trading and wood-based panel businesses. The total capital investment of SJXT was $1,509,000 [Chinese renminbi 12.5 million] of which the Company’s required capital contribution was $519,000. As at December 31, 2000, the Company’s required capital contribution of $519,000 was fully made. 

The operation of SJXT is to organize and manage the first and only national sub-market for timber and log trading in eastern China. The investment in SJXT will provide the Company good accessibility to a large base of potential customers and companies in the timber and log businesses in eastern China.

Moreover the purpose had changed. Originally Sino Forest purchased timber for SJXT. Now SXJT gives Sino Forest access to a large base of potential customers and companies in the timber and log business in eastern China.

Finally there is one new detail about SJXT. It was an equity joint venture formed by the Ministry of Forestry in China. Strangely the government involvement did not warrant a mention in the original press release. It is also strange because it seems to me unlikely that the Ministry of Forestry needed Sino Forest to buy timber for them.

Given that Shanghai Jin Xiang Timber was majority owned by the Ministry of Forestry I wondered where it might be now. The only modern reference to it on the internet is here (click link). It gives two phone numbers. I have rung them in business hours and both ring out. The link also gives an internet address. No surprises for guessing that it is www.sinoforest.com. I guess it must be a majority owned subsidiary now (in which case they purchased it from the Ministry of Forestry and they no longer have a marketing business in Shanghai.)

I could go on and on. The old press releases of Sino Forest are intriguing. They are not particularly consistent with the current story. However time has elapsed and business conditions change over time just as surely as the waist-lines of us gray-haired stock pickers. I will let you (dear readers) decide how much weight you put on those inconsistencies.



John