Wednesday, July 9, 2008

The Best Shareholder Letters

I am a big fan of reading fund managers' shareholder letters. Going back through ten years of shareholder letters from the best fund managers gives you an incredibly useful insight into how they think, analyze opportunities, view risk and rewards and conduct valuations.

The best shareholder letters in my opinion are those by Third Avenue and the Olstein Funds as they really show the reader the investing philosophy and style that underpins these funds. Close runner ups are those by Orbis and Southeastern, though not nearly as in-depth and detailed as Third Ave or Olstein, reading the shareholder reports of these two outfits over the years yields some really useful nuggets of information.


Tuesday, July 8, 2008


Japan is notorious for cross-shareholdings. Today cross-shareholdings stand at 12.2% of the Japanese stock market's market cap for last financial year, according to Nomura. What seems to be getting the attention is that this is the first rise in cross-shareholdings by Nomura's measurement in more than a decade and presumably prompted by some management teams embracing these structures to thwart attempts by activist shareholders to force them to sharpen up their acts. But what most people gloss over is that cross-shareholdings have been on a downtrend for some time, having peaked in 1987, again according to Nomura.

Anyway, regardless of this, these set-ups can actually be quite an appealing investment in some cases. Japan has numerous situations where small companies sit atop of substantial equity stakes in much bigger corporate brethren.

(Apparently, this is also the case with Italy, according to a former colleague who used to work in Milan. If anyone knows of a list of interesting cross-shareholding investment opportunities I would love to see a copy).

With these kind of investments headline ROE, PER etc are likely to be misleading metrics as they tend to be asset plays. So how to play these?

One can buy one of the shares in the cross-shareholding arrangement, short out the cross-shareholding to back out the underlying business of the one of the companies, which may be trading at very cheap stand alone valuations.

While I'm not sure whether Orbis did this with Fuji Electric back in 1999 the investment at the time certainly lent itself to this and Orbis shareholder letters show that the fund was interested in how the cross-shareholding discounted the underlying business. Asatsu-DK and its capital tie-up with WPP also looks like a possible candidate for shorting out the WPP holding to gain exposure to a cheap ad business.

One can also use the smaller company as a cheap back-door entry into its larger cousin, particularly, when the larger company is trading at cheap valuations. A good example of this is Third Avenue's investment in Toyota Industries, which has a substantial holding in Toyota Motor and Denso.

In some instances the stock may even be trading marginally above or below the value of its cross-shareholdings giving an even greater margin of safety. Southeastern recognized this in its investment in Nippon Broadcasting and its holding in Fuji TV. Value here was eventually realized with Livedoor's hostile bid for Nippon Broadcasting.

Nippon Broadcasting touches on another factor to consider - corporate restructuring. Many of these cross-shareholdings tend to be legacies of keiretsus (group company structures). A buy-out by the head company of the smaller firm is a quick way to realize value - assuming the head company pays a decent price - and is also EPS accretive to the acquirer as theoretically they can retire the shares held by the target thus boosting returns for existing shareholders.

Anyway, I am going to try and draft up a list over time of interesting cross-shareholding plays. This is no doubt going to take a very long time (and right now is less than embryonic) as I don't have access to a Bloomberg, and my list will definitely not be exhaustive but I will be posting sporadically on these at this Google spreadsheet .

Sunday, May 25, 2008

Jean-Marie Eveillard's Musings On Japanese Equities

Investors who acquire over a 5% stake in a listed Japanese company must file a report with the Japanese financial authorities within a few days of going over the 5% threshold level. So if you are interested, as I am, in global value funds' investments in Japanese equities (think Southeastern, Third Ave, Silchester Investors International etc) it is very easy to be able to track them in near-real time once they breach the 5% level.

The caveat being you need to read Japanese and know how to use the EDINET filing system to unearth what you are looking for. If you are interested, you can click on the 5% filing rule link on the right hand side of this blog although the details are in japanese otherwise you can e-mail me and I can send you a list of some big funds filing codes in Japan.

I have noticed that Jean-Marie Eveillard's First Eagle Funds/Arnhold and S. Bleichroeder Advisers have been very active in acquiring stakes in Japanese companies over the past six months from their filings with Japanese regulators.

Anyway, I was reading the latest (May) First Eagle Funds' Conference Call transcript and Mr. Eveillard had a number of interesting things to say about Japanese equity prices. In particular, his musings on net-nets among the Japanese small cap space were particularly interesting as this is something I have touched on before and am looking at establishing a basket position in too.

Also his remarks on SMC may be worth looking at further, there is a write-up of the company over at Value Investors Club and for non-Japan based investors the company's stock trades on pink sheets in the United States.

Finally, he had interesting things to say on currency risk exposure for exporters and how to think about this, and also Japanese insurers - Controlled Greed has a position in Millea, and other value mutual funds are keen on names like NipponKoa, Millea etc.

Edited highlights below

"Today, the Ben Graham-type companies, in other words, the deep value stocks, are not available, really, in the U.S. or in Europe. They’re only available in Japan.

And then, the great majority of those deep value stocks are small stocks... we have been doing some work and we already own a few, have owned sometimes for years, a few small deep value stocks in Japan and we’re in the process of doing some additional work on more.

I mean there are stocks in Japan where net cash, cash and sometimes portfolio securities, net of all financial liabilities, net cash is in excess of market cap, which means that you pay less than nothing for the business. Now, as Marty Whitman likes to say, “There is always something that can go wrong.” In that case, if the company were to suffer a string of losses for the next five years, then of course five years from now the cash would have disappeared as a result of the losses.

... in the Tokyo stock market, there are a few industrial companies, some of them world-class, where the stocks had declined to levels where valuations, we thought, were very attractive. Now, when I say industrial, it implies, of course, cyclical, speaking very generally.

So, we assumed in those cases that operating profits would go down 30% over the next 12 months, which nobody assumes today, but just in case. And, we found out that even if we assumed that operating profits would go down 30% over the next 12 months for those companies, the valuations on that basis remain quite moderate. So, we own SMC, which is the world leader in pneumatic equipment...

We also own Keyence3, which is world renowned for their automation products. Again, I’m not saying that the Japanese economy is doing well; it’s doing so-so. I’m just saying that Japan is not going to sink into the sea and that we think we found some very specific investment opportunities among some Japanese industrial companies...

As I said before, we are long term investors, we hold securities on average for five years, which means that sometimes we own them for six, seven, eight, nine, ten years or more. So, we don’t pay a great deal of attention to whether the company is export sensitive or not unless we felt that number one, the Euro would be strong forever vis-à-vis the American dollar and number two, that the competitive advantage of the European company would be ruined truly by the strength of the Euro, which often is not the case.

I mean there are many German companies, for instance, that do not complain today about the strength of the Euro because they are selling, say, machinery that’s so complex and so competitively strong that price is a minor, not quite a minor factor, but price is not a considerable factor.

So, unless we thought that the Euro would be strong for the next three to five years, we don’t mind owning the stock of a company that either exports much to the U.S. or does a large amount of business in the U.S...

We own the securities of a few insurance companies outside the U.S., specifically in Japan. Although it’s a very particular case, I mean these companies in Japan are vastly overcapitalized. Most of the excess capital is invested in Japanese equities. So, we look at them and, so does Marty Whitman of Third Avenue, as disguised investment companies. "

Sunday, May 4, 2008

Walter Schloss speech

Combined notes from a speech by Walter Schloss at the Ben Graham center for value investing and a Forbes profile of the man in Feb of this year

If there is an industry having a problem look at it and you will likely find a good buy there.
Do the companies pay dividends if the dividends are under threat then wait till the cut, see how much the price drops as they may become attractive.

Schloss likes to buy stocks hitting new lows.
Look for companies without much debt - this is important when company stumbles (and takes the stock price down with it).
With these companies it pays to focus on assets.
Has the book value been rising over the past few years? (e.g. five)

Stks below book value with long track history and inside ownership are good.
Look at annual reports and proxies how much stock directors own, company's history and who owns stock
Schloss doesn't like talking to management as they can sway your judgement, look at the figures instead.
Pay special attention to footnotes.

Having target buy price is good move
As you scale your buys in so you should scale your buys out
If you really confident 20% should be max allocation

Wednesday, April 30, 2008

Mason Hawkins Speech

Below are some notes from Mason Hawkins' speech at the Ben Graham Center for Investing he gave in 2005.

Moats and hidden assets
Quantitative undervaluation and qualitative appeal - competitiveness of business and quality of management

Read trade journals.
Reprice most attractive companies so you have target prices
Read every word of the k/s and read the last five to six.
Read the no. 1 competitors' ks too.

Be careful of combining operating and financial leverage mixed together as you don't know when a recession will occur.
One or the other is reasonably handleable.

Most oil & gas companies have horrible economics.
Coca cola bottlers are good businesses. (Not too sure why?)
Ad companies good because of high FCF generation and ROIC ability
Look at the concentration of business in clients tho.
Saatchi & Saatchi got smacked coz Delta & BA were main clients and then came Gulf War and stock tanked.
Service companies assets are their people - you need LT contracts and no compete agreements in there to ensure you can keep generating high ROICs.

Some screens:
1. ROC over 12% and sell for less than 8x
2. Low price to FCF. i.e. below 10x (average of last 3 years capex and working capital charges is what you take out)
3. All companies below book after intangibles taken out

Look at:
- all companies on new low list
- 10 best investors 13-Fs, e.g. Peter Cundill, have stocks gone down since they bought them?
That warrants attention.

I. Balance Sheet valuations based on today's economics
Inventory adjusted for LIFO + tax liabilities
Receivables etc.
You need to make good adjustments
Intangibles can be incredibly valuable.
What you pay for intangibles should be DCF based.
Liabilities should be cross-checked with footnotes; litigation exposure, pension fund financing etc
What is debt coverage? Is debt manageable?
Buying even a poor business at 1.2 of adjusted book can be a great investment.

The five steps: 
1. Use FCF this is cash flow minus working capital charges and maintenance capex
2. Grow out FCF for next 8 years, say they have been growing at 9%, then you wanna grow it out at 6%.
3. Your terminal multiple will be assigned at GDP figures.
4. Then discount at levels way higher than treasuries.
5. Then buy at 60c on the dollar of your valuation.
You do this five hurdles of conservatism and then chuck in qualitative factors, YOU WILL BE FINE.
You need this conservatism coz there are so many inponderables - what rates will be, what management will do etc etc.
Parsimony is also very profitable
Buying at 50c on $ gives you extra compounding effect as business grows.
(If it takes four years to realize value as Hawkins says some businesses can then the compound growth rate here will be 13.6%, this shows why steeper the MOS not only is it safer but the return potential is better) 
Discount Rates
In March 2005 US 10-yr treasury yielded 4.5%
Treasury and a few points of risk premium.
But be cautious in low rate environments.
Historically yields have been nearer 6% if risk premium of a few % is tagged on this gives Southeastern's 9% DCF hurdle rate back in March 2005.
N.B. If a business looks too risky don't either bother running a DCF and valuing it, can it.

Terminal Multiple 
Terminal multiple should gen be based on LT GDP growth figure and assumes very little/no growth.
So once you hit year nine you multiply projected FCF for year 8 by LT GDP growth rate and then multiply by inverse of your discount rate to get terminal multiple.
E.g. Say company is conservatively expected to grow FCF by 4%pa up to year 8. Year 8 FCF flow is estimated to be $100 and LT GDP growth rate is 2% pa, and you are using a 9% discount rate your terminal multiple is:
(100 x 1.02) x 1/0.09 = $1,133.33 for all cash flows out past year 8 into infinity.
This is then discounted back to present.
You can see more about how Southeastern do their DCF modeling here
(BTW - The various "discount factors" in the table in the linked article are just the discounts that the present value cash flows have compared to their future values)

Problems with DCF Models & Why Good Management Is Important.
The DCF model is essentially a FCF yield model for companies and is thus similar to YTM calculations for bonds.
YTM assumes all coupons occur under steady interest rate environment, this almost always false
You need to grasp what will happen with your coupons in order to gauge your returns.
Unlike a bond where the holder gets the payout and can invest it themselves it is up to company management to invest the FCF coupons on your behalf (this is why competent management is important).

Management needs to be able invest FCF intelligently.
Is working capital eating business up?
Has management done intelligent things like buying back shares when they were cheap e.g. below book?

The company has five choices in how it will invest these FCF coupons:
1. Put in treasuries or pay off debt - usually the yield on treasuries and the cost of their debt will be below your discount rate.
(I.e. not a fantastic use of cash)
2. Put it in earning asset base - i.e. reinvest in business. This will probably earn more than discount rate if company is good.
3. Buy shares - worth doing if shares are undervalued by more than the discount rate. (Good)
4. Go on acquisitions - this uses coupons and principal, this can be bad news. Have they made intelligent acquisitions?
5. Return as dividends - if YOU can reinvest at higher than YTM/discount then this is good.

3. Comparable Sales
You adjust comp sales for interest rates in period in which they occured.
Arms-length transactions etc
C/ Discipline and patience - specific price to value settings and ability to wait it out.
2% returns aren;t attractive but 50% losses are even worst.
Great investments are made when you have liquidity and others don't
If you understand a business PERFECTLY then there is no need for MOS.
The more uncertain you are the more MOS you require.
The bridge 6" above the ground and bridge 2 miles above.
If you don't know how far you can fall, you want a hell of a lot of MOS - that means the bridge has got to be rock solid.
MOS revolves around spectrums of confidence:
The balance sheet at the top is your most confident so you can absolutely load up at this level.
But say with Coke you have to do a DCF so then you are going to be less confident than a company with a wedge of cash on the BS.
Remember the future requires way more MOS.

If there is doubt you can forgo an opportunity

You can sell when you are fully invested and you find 40c dollar then you sell your 80c dollar to buy into it.
It's the 100% rule you will only swap out if you think upside is 100% on replacement otherwise you stick with it.
Remember you get a ton of frictional costs when you sell.

Sunday, April 27, 2008

Francis Chou Speech notes

These are notes from two speeches given by Francis Chou of Chou associates at the Ben Graham Center for Value Investing of the Richard Ivey School of Business in 2006 and 2007.

I have merged the notes I took from both speeches together.

When you are value investing you want to look at new lows.
Rejection rates are likely to be 90% for most companies you are looking at and that will be done in about a few mins.
Get to know an industry well.
It takes about six months to get to know an industry - high-tech is more time consuming as product cycle obsolescence means it requires a lot of and continuous work, retail is one of the easiest.
You don't have to be everywhere to have great returns, you just gotta know what you're good at. I.e. circle of competence.

Retailer Valuation
For retailing management has to be outstanding as it is a very tough business. (Keep your eye on the jockey)
Focus on:
i) Accts rec in days - 60 days is average for U.S.
ii) Inventory turnover vs. COGS
High inventory turnover is good.

Valuation doesn't have to be very precise - you are gonna take a range instead.
Just have a MOS on the value range
If you get close to 60% of your picks right you are doing extraordinarily well
You should also try and minimize the damage from the 40% that go wrong
How you minimize mistakes is what guarantees your performance
Be diversified when you begin investing.

Balance sheet analysis is very important - you want to look at inventory and receivables control, especially, for retailers.

Is this a good company or a bad company? Valuation approaches for the two categories will be different.
With crap companies if you pick them up at 5-6x earnings they have already priced in so much bad news that they can be quite profitable even when they do not grow.
For good companies the emphasis is on FCF generation, you don't wanna buy above 10x FCF.

The holding period for will be longer for good companies than for cigar butts.
Good companies can grow intrinsic value at 10-15% pa so you can see your investments double or triple in about four or five years, and as long as they are compounding it is worth holding on to.
Achieving value for heavily discounted and downtrodden companies can take 2-4 years anyway.

When you 1st start off checklists are very useful.
Good company checklist:
Is the business simple and easy to understand?
Does it have consistent business history over the past 5-10 years?
Will it have favorable long term prospects?
Is management chasing revenue for the sake of revenue?
Is earnings all wrapped up in inventory, receivables etc?
Is it highly leveraged?
Good companies don't need leverage they can fund from equity - if it is highly leveraged then you may want to look at buying the debt.
Capital allocation by the management is very important if the management leaves that is a big concern.
Remember to focus on the jockey not just the horse.
What has the company been doing with shares?
Have they been buying back shares and hiking dividends? This is always a good sign.
Having cash on the bal sheet as a cushion for tough times.
10-20% of shareholder's equity as cash is nothing bad.

Net-net is nice and easy way to start investing -It's a brain-dead approach and it really works.
You must play these as baskets as they tend to be troubled companies
20 basket is ideal. 3/10 will probably explode.

Always start with the balance sheet in v.investing and particularly net-nets.
Make sure you are not buying a lot of inventory and receivables, cash is best
Look at tangible book value to share and EV to sales as well
Check on current and quick ratios and cash ratio
Be careful about accounts receivable and inventory measures too (in days) - these can be marked down, unlike cash
Income statement is impt but not that impt for net-nets
Look at capital structure is there debt coming due?
Perhaps they have really cheap debt worth buying or convertibles? Check out the YTM it may be fantastic.
Don't buy into companies about to liquidate as this entails loads of costs that eat up NWC.

Monday, January 7, 2008

Maintenance & Growth Capex

In my last post about rejigging Greenblatt ROIC someone asked about maintenance capex and growth capex in calculating free cash flow.

Maintenance capex is the spending a firm makes in its plant and equipment to maintain current output while growth capex is the spending in plant and equipment made to ramp up production. Free Cash Flow attempts to differentiate between the two and but as many companies do not issue a division between expansion & maintenance capex figuring it out is often necessary, and this requires making judgement calls.

Often it is easiest to identify maintenance capex.
1. Typically for mature businesses most capex will be maintenance capex; depreciation should generally be about equal to capex.

2. Continuously high or stable capex over a number of years that does not lead to increased revenues is also probably all maintenance capex.

3. For rapidly growing firms in fast-changing industries (i.e. tech) most capex will actually be maintenance as is it is unlikely such companies could safely cut capex and maintain market position and cash flows. (Cash flow though may not be a good way to value these businesses though, especially start ups.)

1. You can read the management discussion and analysis and see if the firm is embarking on new business lines. If capex has been steady for many years, and suddenly spiked up this difference probably indicates new growth capex.

It is important to read what management is doing though as some companies scrimp on capex for a no. of years and then suddenly end up having to rush to make a load of new capex to keep the business humming along. Calculating the average age of equipment and depreciation policies of the firm and benchmark to industry norms is another way to confirm this.

2. If cap ex is significantly higher than depreciation it could indicate new growth cap ex. (Again caveats about management that has been less than generous on capex in the previous years applies to this).

Still, some people - like Whitney Tilson on this post - ignore the difference between the two types of capex completely and deduct the whole lot. The argument being this is generally a conservative way of arriving at FCF.

However, even if you do this there is still the question of how one treats acquisitions. Acquisitions fuel the growth of some companies and should be considered capex.

In particular, if a company is a serial acquirer acquisitions will be part of its strategic business operations... therefore, for these firms acquisitions are capex. How do you figure this out? Read the company literature and make your call.

Deducting Depreciation and Amortization Rather Than Cap Ex
If companies are investing heavily in growth, then the lower D&A figure should be closer maintenance capex costs than the full blown capex numbers. So you may want to go for the D&A number to get normalized FCF, especially, if you foresee the expansion phase ending soon.

D&A can also be useful to approximate FCF when capex is highly lumpy and erratic as the D&A figures tend to be smoother. Alternatively, you can create a normalized capex figure over a period of time to calculate FCF.