Closing the Book on Breeze-Eastern

by Geoff Gannon

Quan and I did an issue on Breeze-Eastern last year. The stock has since been acquired by TransDigm (TDG). When we did the issue Breeze-Eastern was priced at $11.38 a share. We appraised the stock at $15.02 a share. TransDigm would later pay $19.61 a share for all of Breeze-Eastern.

What lesson can we learn from our Breeze-Eastern experience?

Here’s what we said about Breeze-Eastern’s stock price at the end of our issue:

“Breeze should – based on the merits of the business alone – trade for between 10 and 15 times EBIT. It is unlikely the stock market will ever put such a high value on Breeze…It is a small company. And 3 long-term shareholders own 70% of the stock. That doesn’t leave a lot of shares for everyone else to trade…Some investors may not like that kind of illiquidity…Breeze is not a fast growing company. And it’s not in an exciting industry. So, it is unlikely to get attention based on anything but its numbers. This might cause investors to underappreciate the qualitative aspects of the company…It is possible that the investment funds that hold most of Breeze’s stock will not hold it for the long-term. They may want to sell the company.”

(Breeze-Eastern Issue – PDF)

Last year, Value and Opportunity did a blog post called “Cheap for a Reason”:

Every ‘cheap’ stock you will find has problems. Some of those problems might be individual (bad management, too much debt etc.), some of those problems might be more sector specific (oil&gas, emerging markets exposure) or a combination of both.

The most important thing is to be really aware what the real problem is. If you don’t find the problem, then the chance is very high that you are missing something.”

So, why was Breeze-Eastern cheap?

Quan and I thought it was that the company had been spending on developing new projects in the recent past that wouldn’t pay off till the future:

“Between 2007 and 2011, Breeze-Eastern’s EBIT margin was depressed by 5 full percentage points as a result of development spending on projects like the Airbus A400M…Breeze-Eastern’s gross margin and operating margin will be higher in the future than they were in the last 10 years.”

The merger document for the acquisition includes a projection by the company’s management to its financial advisors that suggests the reason TransDigm – or any other 100% buyer – would pay more for Breeze than the stock market had often valued the company at was because:

  1. Breeze will have lower costs as it spends less on development projects

  2. AND Breeze will have higher sales as a result of the development projects it spent on in the recent past

The projections show EBIT going from $12.9 million in 2015 to $29.3 million in 2021. This is a 15% annual earnings growth rate. The projected growth is largely due to management’s belief that revenue from platforms under development will go from $0 in 2016 to $28 million in 2021.

This would seem to suggest that we were right about two things:

  1. The stock’s illiquidity made its shares more attractive to a 100% buyer than to individual investors buying just a small, tradeable piece of the company

  2. Breeze was cheap today versus its likely future value, because the company’s reported results included present day expenses as incurred but did not include the expected long-term payoff from supplying new helicopter and airplane projects that won’t be launched for several years

So, maybe the two lessons we should learn from the Breeze takeover being done at a much higher price than the stock traded at or than we valued the company at are:

  1. Always value a stock based on what a control buyer would pay for it as a permanent, illiquid investment – never value a stock based on what traders will pay for small, tradeable pieces of the business (Ben Graham’s Mr. Market rule)

  2. Look for businesses that have to report bad results today even though you know they will report better results in the future

Quan and I weren’t sure if Breeze would have higher revenue from these projects one day. The acquirer here is counting on future projections for revenue. However, we were sure that Breeze would spend less in the future than it did in the past. That was a sure thing.

There is one problem with this analysis of the learning experience we got from Breeze. Mr. Market actually did re-value the company upwards before the acquisition – not after. Breeze went from like $12 a share in the summer of 2015 to $20 a share in the fall of 2015. You didn’t have to hold the stock through the acquisition to make money. We were wrong that Mr. Market would never pay up for such a boring, obscure, and illiquid little stock.

It’s worth mentioning here that Breeze’s enterprise value had been $160 million in 2009. We even mentioned in the issue we wrote that Breeze fell in EV from $160 million in 2009 to $95 million in 2015. Yet, we didn’t speculate that Mr. Market would once again assign Breeze a $160 million enterprise value. It seemed more reasonable to us that someone would buy the whole company. So, again we proved we are really bad at guessing what Mr. Market will do. And maybe it is better to assume we know nothing about how a stock will be valued by anyone other than a control buyer.

Also, we were clearly too conservative in our appraisal of Breeze. At about 7 times what we considered normal EBIT to be, it was a cheap stock when we picked it. And we probably presented it as too much of a value investment and not enough of a quality investment. I think we were biased against Breeze – we ticked off its extraordinary virtues in the text of our issue but still slapped an utterly ordinary EBIT multiple of 10 on the company – due to its small size as a stock and its low growth in recent years. We may have pigeonholed it as “microcap” value. In fact, we knew that based on signs like market structure, relative market share, and the bargaining power Breeze had when dealing with spare parts buyers that its “market power” was among the strongest of any company we’ve covered. Probably John Wiley and Tandy Leather are as strong. Hunter Douglas also has an excellent competitive position.

Since Breeze is a small company in a very small industry, we didn’t have precise data to give on market share the way we often do. All we could say was that Breeze had “a greater than 50% global market share” in a “true duopoly” and that “the only reason customers ever seem to switch from Breeze-Eastern to UTC or vice versa is when they get annoyed that a critical spare part is taking too long to arrive.”

This last sentence is probably the most important sentence in our issue. We rarely come across companies to analyze where the customers tell us flat out that they just aren’t going to switch providers. The two clearest examples of this are Breeze and John Wiley.  

Finally, Breeze is a classic example of a “Hidden Champion”. We’ve only done a few truly dominant companies for the newsletter. Tandy and Hunter Douglas are probably the closest to Breeze in terms of market leadership. They’re also similar in that they have no real peers. We try to present “comparable” peers in each issue. We said flat out in the issue that “Breeze has no good peers.” The same thing is true of Tandy and Hunter Douglas.

There’s a lesson in here. Look for companies with no publicly traded peers. Analysts cover entire industry groups. And investors like to pick from the top down too. If you started from the top down would you ever get to the “helicopter rescue hoist industry”? Where does that fit in a portfolio? What about leathercrafting? Most people don’t even know that’s a real hobby. Or shades and blinds? Is that housing related?

For me, the biggest lessons from Breeze-Eastern are both about timing.

We can time normal earnings. For example, we could see Breeze was under earning now. This isn’t hard. We know Hunter Douglas will make more in the future than it did in the recent past (since U.S. housing was lower than normal). We know Frost will make more in the future than it did in the past (since interest rates are lower than normal). Often, you don’t know “when” this “normal future” is. But, it’s not hard to notice when the present is abnormal in some way.

We can’t time the stock market. Quan and I never would have predicted that Breeze-Eastern’s stock would rise on its own – without a buyout offer – to anything like the level it did. And we never would have expected it’d do it so fast.

I think it’s a lot easier to figure out what an acquirer will eventually pay for a company than what the stock market will eventually pay for a stock.

So, how do we combine the ideas of finding companies that are under earning today compared to a normal future year with the idea of ignoring Mr. Market entirely and simply valuing a stock based on what an acquirer would pay for the entire company?

I guess we could distill that down to a simple investment recipe:

Step One: Fast forward 5 years.

Step Two: How much would an acquirer pay for this company (in 2021 not 2016)?

Last Step: Work backwards to decide how much you should pay for one share of the stock today assuming the whole company is bought 5 years from today.

Read the Breeze-Eastern Issue

Get Your Free Issue of Singular Diligence

Talk to Geoff about Breeze-Eastern

14 Stocks For You To Look At

by Geoff Gannon

Quan and I have picked 19 stocks over the last couple years. One of those stocks, Babcock & Wilcox, split into two different stocks. So, there are 20 stocks that had our blessing. Six of these are not good choices for you to look at. Two have been acquired so you can't buy them (LifeTime Fitness and Breeze-Eastern). Two were stocks we shouldn't have picked in the first place (Town Sports and Weight Watchers). And two are now too expensive to be worth your time (BWX Technologies and HomeServe).

Here is the full list including the six stocks that I'd disqualify.

Once you disqualify those 6 stocks, you're left with 14 stocks that are still worth looking at today:

  • America’s Car-Mart

  • Ark Restaurants

  • Babcock & Wilcox Enterprises

  • BOK Financial

  • Ekornes

  • Fossil

  • Frost

  • Hunter Douglas

  • John Wiley

  • Movado

  • Progressive

  • Swatch

  • Tandy Leather

  • Village Supermarket


America’s Car-Mart

Sells old cars on credit to deep subprime customers mostly in the U.S. deep South.


Ark Restaurants

Runs big single location restaurants (not chains) in high visibility venues (casinos, museums, train stations, parks, etc.).


Babcock & Wilcox Enterprises

Builds and maintains big steam boilers for thermal power (coal, incinerator, etc.) plants.


BOK Financial

A Tulsa, Oklahoma based commercial bank that does a lot of energy lending.



Makes Stressless brand recliners.



Sells watches under the Fossil and Skagen brands it owns and the many fashion brands (Michael Kors, Armani, etc.) it licenses the rights to.



A San Antonio, Texas based commercial bank that also does a lot of energy lending.


Hunter Douglas

Makes the Hunter Douglas and Luxaflex brands of shades and blinds.


John Wiley

Sells academic journal subscriptions to university libraries.



Sells watches under the Movado brand it owns and the many fashion brands (Coach, Tommy Hilfiger, etc.) it licenses the rights to.



Writes car insurance coverage for American drivers who go to the company’s website or get their policy through an agent.



Sells watches under the many brands (Omega, Longines, Tissot, Swatch, etc.) the company owns.


Tandy Leather

Runs a chain of leather goods stores that serves both retail and wholesale customers.


Village Supermarket

Runs a couple dozen big Shop-Rite supermarkets (they average 60,00 square feet of selling space)  in New Jersey.

Talk to Geoff about 14 Stocks for You to Look At

Get Your Free Issue of Singular Diligence

The Two Sides of Total Investment Return

by Quan Hoang

I spend about 10-15% of my time crunching data. That sounds tedious but I actually enjoy this task. It forces me to pay attention to details, checking any irregularity I see in the numbers and trying to tell a story out of the numbers. My recent work on Commerce Bancshares (CBSH) led me to ponder the relationship between ROIC and long-term return.

Over the last 25 years, Commerce Bancshares averaged about a 13-14% after-tax ROE, and grew deposits by about 5.6% annually. Over the period, share count declined by about 1.9% annually, and dividend yield was about 2-2.5%. Assuming no change in multiple, a shareholder who bought and held Commerce throughout the period would receive a total return of about 9.5-10%, which is lower than its ROE. Why is that?

Chuck Akre once talked about this topic:

Mr. Akre: What I’ve concluded is that a good investment is an investment in a company who can grow the real economic value per unit. I looked at (what) the average return on all classes of assets are and then I (discovered) that over 75-100 years that the average return on common stock is around 10%. Of course this is not the case for the past decade but over the past 75-100 years, 10% has been the average return of common stocks. But why is that?

Audience A: Reinvestment of earnings.

Audience B: GDP plus inflation.

Audience C: Growing population.

Audience D: GDP plus inflation plus dividend yield.

Audience E: Wealth creation.

Audience F: Continuity of business.

Akre: …what I concluded many years ago, which I still believe today, is that it correlates to the real return on owner’s capital. The average return on businesses has been around low double digits or high single digits. This is why common stocks have been returning around 10% because it relates to the return on owner’s capital. My conclusion is that (the) return on common stocks will be close to the ROE of the business, absent any distributions and given a constant valuation. Let’s work through an example. Say a company’s stock is selling at $10 per share, book value is $5 per share, ROE is 20%, which means earnings will be a dollar and P/E is 10 and P/B of 2. If we add the $1 earning to book value, the new book value per share is $6, keeping the valuation constant and assuming no distributions, with 20% ROE, new earnings are $1.2 per share, stock at $12, up 20% from $10, which is consistent with the 20% ROE. This calculation is simple and not perfect, but it has been helpful in terms of thinking about returns on investment. So we spend our time trying to identify businesses which have above average returns on owner’s capital.”

The restriction in Akre’s explanation is “absent any distributions.” In general, there are two sides of total return: the management side, and the investor side. Management can affect total return through ROIC, reinvestment, and acquisitions. Investors can affect total return through the price they pay and the return they can achieve on cash distributions.


The Problem of Free Cash Flow

Reinvestment into the business usually has the highest return (this post discusses only high quality businesses that have high ROIC). Problems arise when there’s free cash flow. Management must choose either to return cash to shareholders or to invest the cash themselves. Both options tend to have lower return than ROIC.

Cash distributions don’t seem to give investors a great return. Stocks often trade above 10x earnings so distributions give lower than 10% yield. In my example, Commerce Bancshares wasn’t able to reinvest all of its earnings. It retained about 40% of earnings to support 5.6% growth and returned 60% of earnings in the form of dividends and share buyback. The stock usually trades at about a 15x P/E, which is equivalent to a 6.67% yield. The retained earnings had good return, but the cash distributions had low underlying yield. The average return was just about 10%.

Unfortunately, many times returning cash to shareholders is the best choice. Hoarding cash without a true plan on using it destroys value. Expanding into an unrelated business for the sake of fully reinvesting doesn’t make sense. Similarly, acquisitions often don’t create a good return.

The problem with acquisitions is that they’re usually made at a premium so the underlying yield is likely lower than the yield that would result from share buybacks. The lower underlying yield can be offset by either sales growth or cost synergies. Studies show that assumptions about cost synergies are quite reliable while sales growth usually fails to justify the acquisition premium. To illustrate this point, let’s take a look at 3 of the biggest marketing services providers: WPP, Omnicom, and Publicis.

Omnicom is a cautious acquirer. It spends less and makes smaller acquisitions than peers. Its average acquisition size is about $25 million. Over the last 10 years, Omnicom spent only 16% of its cash flow in acquisitions while WPP and Publicis spent about 44% of their cash flow in acquisitions. Publicis is a stupid acquirer. It makes big acquisitions and usually pays 14-17x EBITDA. WPP is a smart acquirer. Like Omnicom, it prefers small acquisitions. When it did make big acquisitions, it paid a low P/S and took advantage of cost synergies. For example, it paid $1.75 billion or a 1.2x P/S ratio for Grey Global in 2005. That was a fair price as WPP was able to integrate Grey and achieve WPP’s normal EBIT margin of about 14%.

To compare value creation of these companies over the last 15 years, I looked at return on retained earnings, a measure of how much intrinsic value per share growth created by each percent of retained earnings. As these advertising companies have stable margins, sales per share is a good measure of intrinsic value. Retained earnings in this case is cash used for acquisitions and share buyback, but not for dividends.

As expected, Publicis created the least value:

It’s interesting that the smart acquirer WPP didn’t create more value than Omnicom. That’s understandable because acquisitions aren’t always available at good prices. So, it’s very difficult for management to generate a great return on free cash flow. Therefore, the value of a high-ROIC business is limited by the capacity to reinvest organically. Free cash flow tends to drag down total return to low double-digit or single-digit return.


The Investor Side of Total Return

It’s very difficult to make a high-teen return by simply relying on management. The capacity to reinvest will dissipate over time and free cash flow will drag total return down to single digit. However, there are two ways investors can improve total return.

First, investors can shrewdly invest cash distributions. When looking at capital allocation, I usually calculate the weighted average return. For example, if a company invests 1/3 of earnings in organic growth with 20% ROIC and 1/3 in acquisitions with 7% return on investment, and returns 1/3 to shareholders, how much is the total return? It depends on how well shareholders reinvest the money. If we shareholders can reinvest our dividends for a 15% return, the weighted average return is 20% * 1/3 + 7% * 1/3 + 15% * 1/3 = 14%. This number approximates the rate at which we and the management “together” can grow earnings (actually if payout rate is high, combined earnings growth will over time converge to our investment return on cash distributions.)

Second, an investor can buy stocks at a low multiple. The benefit of buying at a low multiple is twofold. It can help improve yield of earnings on the initial purchase price. It also creates chance of capital gains from selling at a higher multiple in the future.

Warren Buffett managed to make 20% annual return for decades because he was able to buy great businesses at great prices and then profitably reinvest cash flow of these businesses.

Small investors can mimic Buffett’s strategy as long as the stock they buy distributes all excess cash. They can reinvest dividends for a great return. In the case of share buybacks, they can take and reinvest the cash distribution by selling their shares proportionately to their ownership. That’s how Artal Group monetizes Weight Watchers (WTW).


Share Repurchase at Whatever Price

This discussion leads us to the topic of share repurchases. I think many investors overestimate the importance of share buyback timing. It’s nice if management buys back shares at 10x P/E instead of 20x P/E. But what if share prices are high for several years? Would investors want management to wait for years – effectively hoarding cash – to buy back stock at a low price?

Good share buyback timing can help build a good record of EPS growth but EPS growth doesn’t tell everything about value creation. It’s just one side of total return. What investors do with cash distributions is as important. So, I think management should focus more on running and making wise investments in the business and care less about how to return excess cash. I would prefer them to repurchase shares at whatever price.

By doing so, management effectively shares with investors some of the responsibilities to maximize total return. Share buyback gives investors more options. Investors must automatically pay tax on dividends but they can delay paying tax by not selling any shares at all. If they want to get some dividends, they can sell some shares and pay tax only on the capital gain from selling these shares instead of on the whole amount of dividends. Or they can simply sell all their shares and put all the proceeds into better investments if they think the stock is expensive.



I do not believe in buying a good business at a fair price. If the management does the right things, holding a good business at a fair price can give us 10% long-term return. But great investment returns require a good job of capital allocation on the investor’s part: buying at good prices and reinvesting cash distributions wisely.

Talk to Quan about The Two Sides of Total Investment Return

Get Your Free Issue of Singular Diligence


A Simple Way to Think about Moat

by Quan Hoang

Moat is really about protecting a company’s profit as Warren Buffett said:

You give me a billion dollars and tell me to go into the chewing gum business and try to make a real dent in Wrigley’s. I can’t do it. That is how I think about businesses. I say to myself, give me a billion dollars and how much can I hurt the guy? Give me $10 billion dollars and how much can I hurt Coca-Cola around the world? I can’t do it. Those are good businesses.

A business may make less profit because of lower sales, lower margin, or both. So, there are two sides of moat: the sales side and the margin side. The sales side can be broken down into customer retention and customer acquisition.


Customer Retention

There are many factors that lead to high retention, including customer behavior, price insensitivity, switching cost, etc.

Customer behavior is subtle. Sometimes customers just don’t think about switching. I used to wonder how small banks can compete with big banks. I realized that a bank’s moat doesn’t come from low funding cost or low operating cost but actually from customer habit. Customers rarely change their primary banking account. So, a small bank may have a low ROE because of its high costs but it can still have stable local market share. Despite their big scale to sell many financial products, big banks can’t steal business from small banks very easily.

Similarly, car owners rarely shop for a new insurer unless there's a bad experience, there's a major event in their life (move or marriage) or there's a spike in the price of their premiums. In fact, insurers like State Farm and Allstate have greater retention rate than Progressive and Geico because they sell bundled products. So, even though Progressive and Geico have a huge cost advantage, they have only low double digit market share after decades of gaining market share. There are simply not many people shopping for new car insurance policies each year.

Price insensitivity helps retain customers in the face of price competition. Customers may pay little attention to price when it’s a tiny part of their total spending. Switching from Coca Cola to a private label cola simply doesn’t save much.

Even better, some customers are willing to pay more for convenience, tailored solutions, product quality, or customer services. I find it interesting that Frost often pays less than one-tenth of the Federal Funds Rate for its interest-checking deposits. For example, Frost paid only 0.47% on its interest-checking deposits in 2007 while many other banks paid about 1.50%. I’m not sure why but perhaps Frost’s customer service is so good that customers simply don’t care about getting interest income on their deposits.


Customer Acquisition

Strong customer acquisition can be achieved through distribution, mindshare, product superiority, or price.

The best example of distribution power is perhaps big food companies. By owning key brands, they have the power to convince retailers to carry new brands. Similarly, journal publishers like John Wiley have distribution power by selling bundles of journals to university libraries.

Mindshare also benefits customer acquisition. Mindshare can be created by advertising, word of mouth or simply positive experience. If a customer decides to buy car insurance directly, he’ll pick either Geico or Progressive. Price comparison is actually much less common in the direct channel than in the agency channel. Direct customers know they’ll get a good price from Geico or Progressive.

Mindshare is particularly strong when a purchase is infrequent. The classic example is See’s Candies. People shop on only one or two occasions a year so there’s no incentive to look for other brands. To a lesser extent, some furniture brands in the U.S. spend a lot in advertising so that people think of Tempur-Pedic when they want to buy a memory foam mattress, think of Select Comfort when they want to buy an air-adjustable mattress, or think of La-Z-Boy when they want to buy a recliner.

Product superiority usually results from investment in R&D or service. It is strongest when there’s a network effect, which makes a popular product more valuable to customers. People join Facebook because many of their friends are there. People go to Amazon or eBay because that’s where they can find the most sellers and thus the best price.

Finally, low price is always a powerful customer acquisition tool.


Margin Protection

Margin can be protected by maintaining the gap between price and costs. To analyze the margin side of moat, the key questions are:

1. Does the company have lower costs than competitors?

2. Does the company have higher asset turnover than competitors?

3. Does the company have higher customer willingness to pay than competitors?

4. Is the power of suppliers and buyers low?

Low cost, high asset turnover, and high customer’s willingness to pay provide a cushion against price competition in the industry. Low power of buyers and suppliers keeps profits from leaking out of the industry. Increasing concentration of buyers or suppliers is a big threat to future profits. Better organized buyers or suppliers may demand better pricing or they threaten to sell or source directly.


Moat Evaluation

To evaluate moat, we should look at 3 things:

1. Barrier to entry

2. Potential damage of new entrants

3. Rivalry among existing firms


Barrier to Entry

In this step, we should try to detect all advantages of a company in customer retention, customer acquisition, and margin protection. Barrier to entry is high when several factors are required at once so that entrants need not only money but also time to compete well. 

Let’s look at Hunter Douglas.

Customer retention isn’t very important for Hunter Douglas because purchases are infrequent. That said, Hunter Douglas’s customers are highly satisfied. Hunter Douglas’s focus on product innovation and services raise a customer’s willingness to pay. Customers rarely downgrade to a lower quality product unless they move to a rental home.

Hunter Douglas’s greatest strength is in customer acquisition, specifically distribution. Independent dealers choose Hunter Douglas because the product quality is higher and they can sell the brand more easily. In fact, it’s usually the brand that draws traffic to dealers rather than dealers getting the traffic themselves. Hunter Douglas advertises more than anyone else in the window coverings industry. It also enjoys great word of mouth.

Hunter Douglas’s margin is sustainable. Thanks to its huge relative market share, Hunter Douglas benefits from economies of scale in advertising, in R&D, and in fixed investments in its integrated and highly automated manufacturing facilities. Customer’s willingness to pay is higher than competitors because of product quality, advertising, and dealer service. Finally, the power of suppliers and buyers is low. Hunter Douglas is an integrated manufacturer so its inputs are mostly commodities. Dealers have little power because they’re individual mom-and-pop operations.

A competitor needs brand recognition and low cost to recruit dealers. But it needs wide distribution to have low cost and to justify advertising investment. So, this is a chicken and egg problem. The best chance to compete with Hunter Douglas is through the online channel or through the big box channel. The only weakness in Hunter Douglas's moat is the possibility of a market share shift between channels.

It’s worth noting that customer power is very strong in the big box channel, which Hunter Douglas avoids. Home Depot or Lowe’s usually demand a low price. Hunter Douglas’s competitors sell through big boxes and make little profit to reinvest in marketing or product innovation. And they tend to cut product quality to meet the low-price demand from big boxes. So, end-customers can be segmented by channels.

A weak example is Weight Watcher (WTW).

WTW’s customer retention is weak because its members come and go. People don’t take responsibility for their failure so they’re always eager to try new weight loss programs.

WTW’s customer acquisition is a bit better. It has no distribution advantage because fads can pick up easily in health and fitness. However, it has strong mindshare. Many people know what it is and know its effectiveness. In fact, its members re-enroll in the program on average three more times during their lives. All WTW needs to do is to give people a reason to join right now instead of putting off joining indefinitely.

WTW’s margin is great. Customer willingness to pay is high thanks to the social value of weight loss. The power of suppliers and customers is low. So, it’ll make a lot of profits as long as it gets enough customers.

So, WTW doesn’t have moat but it has some strength in customer acquisition. Its former customer base, about 20 million in the U.S., grows over time. The overweight population grows over time. So, it just needs the right program news and the right marketing buzz. In the past, Weight Watchers has always made record profits after each cycle.


Potential Damage from New Entrants

Sometimes a business can be good despite a low barrier to entry. I think WTW is a good example. So, it’s useful to make some attempts to judge the potential damage of new entrants.

The best tool in this step is to look at history. I unfortunately underestimated WTW’s risk in the recent cycle. The problem is that I only had WTW’s financial data back to 2001. I would have been more cautious if I had financial data back to 1990 to see how WTW performed in its previous crisis in early 1990s. Without financial data, we only know that WTW has always been able to reinvent its program during each crisis.

I think there’s some similarity between WTW and restaurant chains. The market is huge. Each chain has a different concept and few have a big market share. But barrier to entry is low and customers are willing to try new restaurants.

New entrants actually don’t do as much damage to existing chains as people think. There are not many new successful stories like Chipotle (CMG). And it took Chipotle 20 years to seize just a 2% market share. A rate of two percent is less than the restaurant industry grows in a single year.

Moreover, new restaurants don’t always open in a brand new location. Often, they open in or near a location another restaurant concept had occupied before. So, the total supply of fast food restaurants doesn’t necessarily increase more than population growth despite the industry’s low barrier to entry.

So, the biggest competition in fast food chains is actually among existing chains.

Recently, Value and Opportunity argued that Fossil has no moat because “…wholesale distribution network seems to be quite open for newcomers like Daniel Wellington.” It’s true that department stores always have store space to test new suppliers. Retailers care about gross profit per square foot so they often switch suppliers in that area of their stores. Sometimes a new entrant may succeed.

But how much will the success of a new watch brand actually hurt Fossil?

Customer’s willingness to pay is high in this category so it won’t hurt margin. It may hurt volume, thus reducing gross profit per square foot of Fossil’s licensed watches. But cannibalization might not be as high as most people expect. Revenue of Michael Kors watches went from $100 million to $900 million in 5 years but it didn’t hurt Fossil’s other brands. And Fossil’s other brands didn’t benefit from the recent fall in Michael Kors either. Perhaps not many people wear watches today so Michael Kors helped bring more consumers to the category. Or it simply induced people to buy more often.

Also, can Daniel Wellington build on its early success and keep its brand relevant? It’s a big challenge for any newly successful entrant.

So, I think it’s safe to bet that a (growing) diversified portfolio of strong licensed brands can help Fossil gain market share over time.


Rivalry Among Existing Firms

This step is similar to the Barrier to Entry step. However, we compare a company’s strengths in customer retention, customer acquisition, and margin with its competitors. We want to see some durable competitive advantages that allow a firm to gain market share over time. And we want to make sure that the moat is widened over time.

In our bank example, every American bank has a pretty good retention rate. However, some regional banks such as Commerce Bancshares (CBSH) and BOK Financial (BOKF) have the scale to sell other products like wealth management or payment systems service. Selling more products to a customer creates a closer relationship and improves retention. More products also mean more touch points to acquire new customers. Finally, more products result in more fee income, reducing net operating cost and creating a cost advantage. A cost advantage may allow for more reinvestment in products and services. So, these banks will possibly gain market share from tiny local competitors over time.



There’s nothing innovative in this this framework. But I think it can help us connect different competitive advantages and see a clearer picture of moat. It’s a simple way to think about moat. But the most important question is always the one Warren Buffett asks:

I say to myself, give me a billion dollars and how much can I hurt the guy?”

Get Your Free Issue of Singular Diligence

Talk to Quan About Moat




Boredom Is a Good Friend of Long-term Investors

by Quan Hoang

Geoff said in the last post that: “simply learning to love illiquidity, boredom, and a lack of headlines in your portfolio might be enough to improve your returns.” The key word is boredom. I think 3 main reasons for a stock to be boring are low growth, lack of catalyst, and a so-so price. A stock with these characteristics is not attractive to growth investors, value investors, and momentum investors. But sometimes these characteristics hide qualities that can generate great long-term returns.


Quality of Growth

I once made a bold statement that Frost promises the best growth investors can find. I think that Frost can have 7-8% growth for the next 20-30 years and I don’t normally find a stock with such high growth potential. My friend was surprised at my claim and he said “you can’t say that because 20% growth is a certainty for companies like Valeant!” What he said represents the attitude towards growth of most people. To them, a single-digit growth isn’t stellar. To me, 7-8% growth is a treasure. That doesn’t mean I’m less demanding. I’m just focused more on quality of growth.

Low growth can be valuable if ROIC is high. Let’s compare Bristow, Frost, and Omnicom.

Over the last 10 years, Bristow’s revenue almost tripled from $674 million in 2004 to $1,859 million in 2014, which translates into an 11% annual growth rate. Annual sales growth was always higher than 10% except for the “bad” years between 2009 and 2011. The problem is that pre-tax ROIC is just about 9%. So, Bristow had to use debt and equity to finance growth. Over the period, net debt increased by $650 million. Share count increased by more than 50% from 23 million to 36 million mostly as a result of the issuance of $223 million in convertible preferred stock in 2007.

Frost is a better business. Frost grew deposits from $7,767 million in 2004 to $22,053 million in 2014. That means intrinsic value has compounded annually by 11% over the last 10 years. Unlike Bristow, Frost can make 18-20% ROE. So, Frost was able to return 40% of total earnings over the last 10 years.

Omnicom is even better. Omnicom grew sales by 5% over the last 10 year while returning 110% of earnings to shareholders. Omnicom doesn’t need to retain earnings to grow. It actually received about $1 billion from the decrease in its negative working capital over the period.

Omnicom’s 5% growth can be as valuable as Frost’s 8% growth.  If we pay 20 times after-tax earnings for both stocks, we can get similar returns. Omnicom can give us 5% growth and 5% yield, adding up to 10% total return. Frost can grow 8% while paying out 50% of earnings. So, it can give us 8% growth and 2.5% yield, adding up to 10.5% total return. A similar calculation shows that we can get 11.67% total return from Omnicom and 11.34% return from Frost if we pay 15 times after-tax earnings for both stocks.

So, growth can be more valuable than numbers suggest. Adjusted for quality, Omnicom’s 5% growth is equivalent to Frost’s 8% growth, and is much superior to Bristow’s 11% growth.

Consistency also contributes to quality of growth. Geometric means are always smaller or equal to arithmetic means. Therefore, low-growth years tend to pull compounding growth more than high-growth years. In other words, of two firms with the same average (arithmetic mean) growth, the one with more consistent growth has the higher compound annual growth rate.

To illustrate this point, let’s look at Frost’s deposit growth over the last 10 year and Select Comfort’s sales growth from 2002 to 2012:

Frost’s annual growth: 13% 11% 3% 18% 13% 8% 14% 11% 14%

Select Comfort’s annual growth: 37% 22%    24% 17% -1% -24% -11% 11% 23% 26%

Over the 10-year periods, Frost’s average growth was 11.8%, and compounding growth was 11.0%. Select Comfort’s average growth was 12.3% and compounding growth was 10.8%. Select Comfort’s growth looks fancy. It was higher than 20% except for recession years. But Select Comfort’s average compounding growth was actually lower than Frost.

A serial acquirer like Valeant may have a great platform to create value from acquisitions. But we never know when the party will end. At some point, smaller acquisitions can no longer move the needle. Bigger acquisitions can be more expensive. The company may become too big for adequate management. So, Valeant’s growth is repeatable but not predictable.

I prefer companies with consistent growth drivers. Some drivers are company-specific and some are external.

Store expansion is a reliable internal growth driver. If a company is disciplined in picking new locations that fit its model, and if it successfully replicates its competitive advantage, performance will be consistent. Growth is predictable. We can pick a conservative number of stores the company can open, and pick a conservative number of years for it to fill in the opportunities. For example, America’s Car-Mart (CRMT) will keep opening stores in small town with populations from 20,000 to 50,000 in South-Central states. It has huge competitive advantages over mom-and-pop operators in these areas. Car-Mart now has 141 locations. It may double store count in 10 years. So store openings can lead to 7.2% growth. With 3-5% same-store sales growth, it can have double-digit growth for the next 10 years.

The driver I’m most comfortable with is market share gain. Some companies have durable weapons to consistently gain market share. For example, Progressive (PGR) and Geico have low-cost advantage to gain market share almost every year. Industry growth is usually predictable so betting on market share gainers allows me to sleep well at night.

Industry tailwinds can make market share gainers more attractive. For example, banking is a better business than car insurance. Total deposits tend to follow GDP growth while car insurance policies face deflationary pressure as technology reduces accident frequency. Texas GDP grows about 1% faster than the U.S. GDP so banks in Texas have 1% higher growth than the average U.S. bank. Frost is a great franchise in Texas. It’ll keep growing its relationships with small businesses. It’ll keep attracting consumers for being a Texas bank and for great customer service. So, it’s safe to expect 7-8% annual growth far into the future.

Omnicom also benefits from changes in the industry. Marketing budget is stable as a percentage of sales at most companies. So, total marketing spending tends to follow GDP growth. However, marketing is becoming more and more fragmented due to the rise of new channels like online, mobile, social media, etc. Marketing spending will shift from traditional media to new mediums. That means more and more work for agencies. Omnicom has done a great job at building new capabilities or making small acquisitions to serve client needs. Another trend is that marketers want to deal with fewer vendors, leading to account consolidation. These two trends allow Omnicom to capture more % of clients’ spending over time. So, Omnicom’s organic growth is about 1% or 2% higher than GDP.

Margin expansion can make growth better than it looks. Margin expansion usually happen when gross margin is high, price competition is low, and fixed cost is significant. Watches are a good example. A watch isn’t a timepiece. It’s a fashion piece or a jewelry piece. Watch brands don’t really compete on price. People have a price range for their watches and they choose the brand and style they like most within that range. So, even cheap Chinese made watches that Fossil (FOSL) sells have 50-60% gross margin. Fixed costs of designing, distributing and advertising watches for a brand are quite significant. So, bigger brands have higher margin. Fossil’s 16% EBIT margin is higher than Movado’s 12% because Fossil has higher revenue per brand.

Within Fossil, Michael Kors possibly makes 30% EBIT margin. Revenue from Michael Kors watches was over $900 million last year. But Michael Kors can be a fad. It can rise and fall, and has a big impact on Fossil’s EBIT margin.

Margin expansion is better when combined with consistent growth. It causes EBIT to grow faster than sales. So, mid-single digit sales growth may mean high single-digit earnings growth. Of the candidates I’m looking at, Grainger (GWW) is potentially a company with both margin expansion and consistent growth. If that’s true, it can have low double-digit earnings growth for many years and deserves a high multiple.

So, investors should be skeptical of high growth, and be positive about low growth. High growth can be a problem if it’s unpredictable and leads to high expectations and thus a high multiple. Low, boring growth can be a treasure if it’s consistent and supported by high ROIC and margin expansion.


Lack of Catalyst

A lack of catalyst can cause even value investors to neglect a stock. Two of the least controversial stocks we’ve analyzed are Progressive and Frost. Most people agree on their moat and quality, and most don’t think the stocks are cheap. Geoff and I didn’t realize how cheap they were until looking deeply at them. Higher interest rates can be a catalyst but most responses I get from people are “interest rates may stay low for a while” or “how quickly earnings will increase?” What’s interesting is that no guru owns these stocks while many own Wells Fargo. According to Morningstar, most concentrated shareholders are ETFs or institutions that put a small % of total portfolio into Frost. So, it’s possible that most professional investors just look casually at the stock.

I’m not sure how important catalysts are. It’s easy to imagine catalysts and get excited. People tend to overestimate the chance of catalysts happening and underestimate the chance of unexpected events. In my short experience, I’ve seen stocks we’ve picked like Ark Restaurants (ARKR), PetSmart, and Lifetime Fitness get acquired or buyout offers. Such events happened more often than I expected.

I don’t think there’s any problem with the lack of a catalyst. Time is such a good friend that good businesses don’t need a catalyst. We’ll do fine as long as we buy businesses that can compound intrinsic value.

That said, I do see catalysts work in some special situations. Sometimes catalysts are playing out but the market is so inefficient that the stock price doesn’t reflect the ongoing developments. That’s true for the two best performing stocks we picked this year: Babcock & Wilcox and Breeze-Eastern (BZC). Babcock returned about 30%. Breeze-Eastern returned about 22% since our issue on the stock was published, and about 40% since when I began analyzing the stock in May.

Babcock is a spin-off. I’m really surprised because value investors follow spin-offs closely. I started analyzing the stock in October 2014. There were several presentations about the spin-off and I was worried that the price would move a lot before we published the issue. Yet, it stayed flat for months. And then it worked out exactly as Joel Greenblatt taught us about spin-offs: the good Babcock (BWXT) is expensive at 13.4x EV/EBIT, and the bad Babcock (BW) is cheap at 5.4x EV/EBIT.

Breeze-Eastern was lucky timing. I analyzed the stock when it was finishing some long-term projects. So, R&D expense was going down and revenue was ramping up. Years of under-earnings were coming to an end. Anyone could expect Breeze to make more profit. But few people acted until it released quarterly earnings.

It’s worth noting that some investors (who are now among the company’s biggest shareholders) bought Breeze-Eastern on the same thesis in 2011. But they had to wait until 2015. So, I was lucky to analyze Breeze-Eastern in 2015 when I saw that R&D had been declining.

Right now I think Ekornes is another special situation. Ekornes keeps most production in Norway. A lot of cost is in Norwegian Krone (NOK). But Ekornes gets 95% of revenue outside of Norway. So, most of revenue is in U.S. dollars, Euros, or other currencies. For years, Ekornes was badly impacted by the overvalued NOK. But the recent crash in oil price caused NOK to decline against U.S. dollars and Euros. On the surface, currency is a risk. American investors buying Ekornes today may sell Ekornes and convert into fewer $ if NOK weakens. But weaker NOK is actually a boon for Ekornes because revenue in $ or € will be converted into much more NOK while costs in NOK won’t change. As a result, margin can expand by a huge amount, and the NOK-based stock price appreciation will far outweigh NOK valuation.


So-so Price

Value investors can have different styles but they all buy stocks on the same principle: they minimize downside.

Buy-and-hold investors seek safety in business quality. They ask themselves if they buy a stock at current market price and hold forever, what is the very conservative expected return they can get? If the expected return is adequate – say 10% - the stock is very safe.

How can they make more than 10%?

They can get more than 10% return as long as they buy a good business at a lower than average price. Sometimes they’re luckily to buy a good stock very cheaply. For example, paying 10 times unlevered P/E for a business that can grow 5% while paying out all earnings can result in 15% total return. More realistically, they can only buy at 12-15x unlevered P/E. That’s still good because the stock – if it’s truly a good business – will soon trade at the normal 18-20x P/E. Even if it takes 3 years for the multiple to expand from 15x to 18x, the expansion still generates 6.3% annual return. Adding 5% annual growth and 6% cash return results in over 17% annual return.

So, just like growth, price can be better than it looks if we consider quality.


Low Expectation

One great lesson I learnt from Geoff is to keep expectation low. He says if he does everything right, he might be able to get 10% a year long-term.

I do think that keeping expectation low is the key to learning to love boredom. 10% sounds low to most investors. Investors like things that can double in 2 or 3 years. In pursuit of high return, they embrace dirt-cheap price or fancy growth. My view is that there’s no certain 20% return. Big upside comes with high risk. There are a few certain 15% returns. And there are some certain 10% returns. So, I stay with such certain boredom. If I’m lucky I can get 15-20% return in my career. But without luck, I can still make 10%. I’m really comfortable with luck-independent 10% return.

Get Your Free Issue of Singular Diligence

Talk to Quan About Boredom Being a Good Friend of Long-Term Investors

Should We Care Why the Stocks We Buy are Cheap?

by Geoff Gannon

One of my favorite blogs, Value and Opportunity, recently did a post about how the best value stocks are often those that are not cheap by the most obvious numbers (P/E, P/B, etc.).

The post is entitled “Value Investing Strategy: Cheap for a Reason”. The basic argument of the post is that:

“…Especially in a market environment like now, cheap stocks are cheap for a reason. It is very unlikely that ‘you’ are the first and only one who knows how to run a screener and by chance you are the only one who can buy this great company at 3 times earnings which will quadruple within 6 months…The most important thing is to be really aware what the real problem is. If you don’t find the problem, then the chance is very high that you are missing something.”

This is not at all how I look at stocks.

I usually don’t know why a stock I’m buying is cheap. And I’m not sure I spend much time trying to figure out why someone else would or would not like the stock. I tend to just focus on whether I like the business and how much I’d “appraise” that business for.

I can sometimes come up with possible reasons for why a stock I like is cheap. But, I’m never sure those are the real reasons other people aren’t willing to buy the stock.

I don’t think Quan sees himself – and I know I don’t see myself – as a contrarian investor.

So, I assumed looking to see if a stock was “cheap for a reason” is something I simply don’t do.

At least that’s what I thought before looking through the textual record of what I actually said about each stock I picked.

In my last post, I mentioned 6 stocks that Quan and I picked for Singular Diligence which are now trading at a discount of 34% or greater to our original appraisal value. So, these are the 6 cheapest stocks we know of in intrinsic value – rather than traditional value metric – terms.

I decided to go through the record and check for two things.

One, how cheap are these stocks on the traditional value metrics. I will use Morningstar’s measures of P/E, P/B, and Dividend Yield for this.

Two, what reason did I give (in the issue where I picked the stock) for why that stock might be cheap.

Here are the 6 stocks.


Hunter Douglas

Discount to Appraisal Value: 58%


Forward P/E: 9.6x

P/B: 1.4x

Dividend Yield: 3.6%


Why I Said it Might Be Cheap:

“Hunter Douglas is an obscure stock. The Hunter Douglas brand is American. So, the company’s name is American. However, the stock trades in Europe. The company reports its results in U.S. dollars. But, the stock trades in Euros. The stock is 81% owned by the Sonnenberg family.”

(Note: Quan and I appraised this company – which sells shades and blinds mostly into the U.S. and European housing markets – based on its normal cyclical earnings, which we believe to be much higher than the very depressed earnings Hunter Douglas reported from 2009-2014. That could be given as a reason for why the stock is cheap. However, U.S. traded stocks tied to housing are generally priced much more in line with the idea we are at a cyclically depressed point for their earnings. Hunter Douglas isn’t.)



Discount to Appraisal Value: 55%


Forward P/E: 12.0x

P/B: 1.4x

Dividend Yield: 3.4%


Why I Said it Might Be Cheap:

“Frost has created a lot of intrinsic value since the 2008 financial panic. The stock market has not realized this because Frost has made very little on its loans and securities due to the Fed Funds Rate being near zero. In 2008, Frost had $10.5 billion in deposits. Today, Frost has $24 billion in deposits. Frost’s value comes entirely from its non-interest and very low interest bearing deposits. So, the intrinsic value of Frost as a buy and hold forever stock more than doubled from 2008 to today. The stock did not double, because reported EPS barely budged due to the yield on loans and securities being the lowest in history. When the Fed Funds Rate eventually increases from about 0% to 3% or higher – as all members of the Fed expect it will by about 2018 – Frost’s reported earnings will double. When Frost’s reported EPS doubles, its stock price will double. At that time – when the Fed Funds Rate has been 3% or higher for a year or more – investors will think Frost has become twice as valuable. That is false. Frost more than doubled its intrinsic value from 2008 to 2015, when it more than doubled its free and almost free deposits. A Fed Funds Rate near zero disguised this fact for about 7 years. Frost’s value was hidden for the last 7 years. But, Frost’s value will be obvious over the next 7 years. Frost is the clearest and best investment idea we have had since starting Singular Diligence in 2013. That fact is not obvious as I write this in 2015 with a Fed Funds rate near zero. It will be obvious in hindsight (in say 2019) with a Fed Funds rate near 3%.”



Discount to Appraisal Value: 47%


Forward P/E: 9.4x

P/B: 1.3x

Dividend Yield: 0%


Why I Said it Might Be Cheap:

“Over the last 15 years, Car-Mart’s stock has returned 16% a year versus the S&P 500’s 4% a year return. It sounds strange to propose a stock is cheap when it is at ‘normal’ levels for a typical stock and within the range of multiples it has traded for in the past. However, Car-Mart’s past returns are very high compared to most stocks. In other words, the ‘normal’ historical range of multiples that Car-Mart traded at was simply too low… In the past, the company’s enterprise value has ranged from 0.8 times retail sales to 1.5 times retail sales. This constant undervaluation is what has caused the stock to outperform the S&P 500 by more than 10 percentage points per year since it went public.“


Tandy Leather

Discount to Appraisal Value: 47%


Forward P/E: 8.9x

P/B: 1.5x

Dividend Yield: 0%


Why I Said it Might Be Cheap:

I couldn’t find any quote from our issue on Tandy explaining what might cause the stock to be undervalued. I can come up with an argument now – but I don’t find it very convincing.

Here’s the argument. Tandy is an illiquid stock. It trades less than $100,000 worth of shares on a normal trading day. As the only publicly traded leathercrafting retailer it has literally zero peers. Only one analyst covers the stock. Most investors simply haven’t heard of Tandy Leather.

I find that argument unconvincing because this is a U.S. stock. Tandy trades on the NASDAQ. It shows up on all screens that you’d run in the U.S. Institutional ownership accounts for about two-thirds of the shareholder base. That means institutions hold over $50 million of Tandy stock. It’s a visible stock. People aren’t oblivious to its existence.


Swatch (*Valuation metrics are from Stockopedia for this one)

Discount to Appraisal Value: 39%


P/E: 14.8 x

P/B: 1.9x

Dividend Yield: 2.1%


Why I Said it Might Be Cheap:

“The greatest risk of misjudging Swatch is not seeing a prolonged recession or depression coming in China. China is still much, much poorer than many of the markets it trades with. There is a lot of room for GDP per capita to grow over time. That means there is a lot of room for real wages to grow over time. However, China has several features that could be warning signs for a Japan like 'lost decade'. It is not the point of this issue to speculate on that possibility. But it is rarely talked about by investors. And it is a potential problem. At no point in the last 35 years has Chinese economic growth been poor enough to qualify as anything like a recession. So, the biggest risk of misjudgment is assuming that the trend of the last few decades will always be normal. 

Chinese GDP growth has been in the range of 7% to 8% lately. Chinese population growth is only 0.5%. This means that GDP per capita is growing – even now – at 6.5% a year or faster. About 45% of Chinese GDP is investment in fixed capital. Meanwhile, about 15% of U.S. GDP is investment in fixed capital. This makes the risk of an overhang of long-lived assets much higher in China. There is a very high rate of growth in fixed capital per person. This is because Chinese GDP is very fast growing, almost half of Chinese GDP is investment spending, and Chinese population growth is low. As a result, China would be much less able to absorb a glut of long-lived assets. If the country builds too many apartment complexes, factories, airports, power plants, etc. they run the risk of having them be vacant or idle. Industries like construction are important in China. These are long cycle industries. They are susceptible to periods of overbuilding and then periods where they must be idle to absorb the overexpansion of previous years. China has been rapidly expanding since the late 1970s. So, there is always a risk of the sorts of problems Japan had…Quan and I have no predictions about the future growth of China. But it’s important to point out the impact a stagnant Chinese economy would have on Swatch. In the future – Swatch will likely get both the majority of its profits and the majority of its growth from Chinese consumers. If China’s economy has the kind of experience Japan’s did over the last 25 years – Chinese consumers will not increase their watch buying. Swatch’s growth will decline by at least half. If China is stagnant – Swatch may be stagnant.”


Ekornes (*Valuation metrics are from Stockopedia for this one)

Discount to Appraisal Value: 38%


P/E: 12.0x

P/B: 2.6x

Dividend Yield: 6.1%


Why I Said it Might Be Cheap:

“Ekornes is clearly cheap. There are two possible reasons for this. One, the Ekornes name is unknown worldwide because the company’s main brand – Stressless – is different from the name under which the company is listed. Two, Ekornes trades on the Oslo Stock Exchange. Norway is a tiny country of just 6 million people. Very few investors outside Norway buy shares of Norwegian companies in Oslo. For example, half of Ekornes’s shares are held by Norwegians. Only 50% of Ekornes’s shares are in foreign hands. If Ekornes listed in Frankfurt, London, or New York – it would probably get more attention from investors outside Norway. The share price might be higher…The biggest reason why a foreign investor might avoid Ekornes is concern that the stock – which is bought and sold in Krone – will fluctuate in the investor’s home currency along with the exchange rate between that home currency and the Krone. So, for example, an American investor might feel certain that Ekornes’s share price of 100 Norwegian Krone will one day expand beyond 125 Norwegian Krone, but that investor fears a 25% drop in the Krone versus the Dollar – like the drop from 17 cents to 13 cents in the past year – would more than wipe out his gain. This is a valid short‐term concern. Ekornes’s share price in dollars will fluctuate even when the price in Oslo stays the same in Krone. However, this is not a valid long‐term concern. In the long‐run, a stock’s intrinsic value will follow its earning power. Ekornes’s earning power comes from the gap between its sales – made in Euros, Dollars, Pounds, Yen, etc. – and its costs. Ekornes gets 94% of its sales in currencies other than the Krone. Only the company’s labor cost is tied to the Krone. So, it is misleading to think of Ekornes’s intrinsic value as being a primarily Krone based figure.”

We can make a few statements from the above list. One, I usually do give some reason for why a stock might be cheap. I’ve never thought of this as being an important part of my own process when it comes to picking stocks. But, apparently giving a reason “why a stock might be cheap” was important enough for me to include when writing to subscribers in 5 out of 6 cases.

Two, the stocks that look cheapest to us also look cheap based on traditional value metrics. Stockopedia tells me that the median forward P/E of all stocks in the U.S. and Europe is 15.2x.  The forward P/Es of the 6 stocks Quan and I think are cheapest ranged from 8.9x to 14.8x.

There is one other test we can run on the 6 stocks Quan and I think are cheapest. Here is a paragraph from near the end of the Value and Opportunity post:

Actually, I strongly prefer “forgotten” sectors compared to those which just have recently started to decline. Yes, everyone is looking at oil companies these days as they have declined a lot in the last months and look cheap. But I actually find better value in banks or financial companies.”

Look at the 6 stocks that are cheapest based on today’s price as a percentage of our original appraisal value for them. Two of the stocks – Frost and Car-Mart – are U.S. financial companies. And two of the stocks – Ekornes and Hunter Douglas – are furnishings stocks.

Finally, we can take a “top down” look as Value and Opportunity suggests. European stocks are cheaper than U.S. stocks. Most of the stocks we pick for Singular Diligence are U.S. stocks. But, 3 out of 6 of the stocks we think are cheapest trade in Europe. Ekornes is listed in Norway. Hunter Douglas (although more an American company than anything else) is listed in the Netherlands. And Swatch is listed in Switzerland.

In the case of Hunter Douglas, I have to admit I do think the stock would trade at a higher P/E ratio if it were listed in New York instead of Amsterdam.

So, although we consider ourselves bottom up stock pickers – there is a top down pattern at work here. European stocks are cheaper than U.S. stocks. Stocks tied to U.S. housing activity are cheap. And stocks tied to U.S. interest rates – that is, stocks that do better when rates are higher and credit tighter – are cheaper.

If you just take the stocks that trade at a 34% or greater discount to our appraisal value – they’re not very diversified at all. Half of that group is European. One third is furnishings. And one third is U.S. financials.

This means that one continent (Europe) and two industries (furniture and finance) explain most of the cheapness in the group.

The one exception is Tandy. There is no top down explanation for Tandy’s cheapness that I can see. And I don’t really have any explanation for why the market values the business so much lower than I do.

There is something else to consider. A top down explanation for why these 6 stocks are cheap may not be the only explanation. True, Hunter Douglas and Swatch are both European. But, they’re also both family controlled. Neither family has any interest in hyping their stock. And neither family is likely to sell at any price.

Which explanation is the right explanation?

Why is Hunter Douglas cheap?

Is it because the Sonnenberg family controls so much of the stock and doesn’t care about getting Wall Street’s attention? Is it because it’s a European stock instead of a U.S. stock? Or, is it because Hunter Douglas’s earnings are tied to U.S. housing and investors are pricing the stock off cyclically low earnings as if they were cyclically normal earnings?

I don’t know.

There are patterns in the “cheap” stocks we find.

They do seem to come from cheap parts of the world and to be in cheap sectors.

Now, I want to balance the numerical evidence with some purely anecdotal evidence.

I’ve never tallied up the emails. But, I get a pretty good number of them from people telling me they liked a stock I wrote about quite a bit – but they never actually bought it.  

When subscribers tell me why they liked a stock a lot but never bought it – they often give one of 5 reasons:

  1. It’s illiquid

  2. The broker they currently use won’t buy it for them

  3. It trades in a currency different from their own

  4. It’s boring

  5. There’s no catalyst – they plan to wait and maybe buy it later

I have no data supporting these 5 possible explanations for “why a stock is cheap”. I think top down explanations for cheapness are often right. There are simply countries and sectors that are out of favor. But, I also think these 5 explanations may be right too. Stocks that are illiquid, boring, and lack a catalyst may always be cheaper than they deserve to be.

If that’s true – simply learning to love illiquidity, boredom, and a lack of headlines in your portfolio might be enough to improve your returns.

My own opinion is that if we are told a stock is “cheap” to start with, we’ll always find a reason why it should be cheap. We don’t value stocks blind. As value investors, we often know the P/E, P/B, the dividend yield – and maybe even the EV/EBITDA – of a stock before we’ve even read the company’s annual report. We come in expecting to find warts and so we find warts. We then tell ourselves that the cheapness of the stock must be due to these warts.

Once you’ve seen a cheap stock price – you can’t undue that kind of bias. Your mind has been tainted with the market’s view of the stock before you even read the business description.

It’s strange but true: in investing, you know other people’s views before you know your own.

The best situation would be to have no knowledge of a stock’s price when you estimate its value.

The next best situation would be to do our best to forget whatever prices we’ve seen and focus instead on calculating a truly independent appraisal of the stock's value.

Get Your Free Issue of Singular Diligence

Read Value and Opportunity’s “Cheap for a Reason” Post

Talk to Geoff About Why Stocks We Like Are Cheap

Misconceptions about Banks (and a Glance at Wells Fargo)

by Quan Hoang

I came to the U.S. in the fall of 2008. It was a turbulent time when I read about bankruptcies or about “too big to fail” banks almost every morning. As a new student of economics and value investing, I developed an ingrained prejudice against banks in this abnormal period.

Like most casual observers, I thought that banking is a risky business because of high leverage. Business cycle is inevitable and a small mistake can be magnified into big losses. So, I’d better stay away from banks.

Geoff asked me to look at several banks over the years. No bank was good enough to change my preconception. But some did recently. I spent the last 3 months researching banks. I realized that there are good banks and bad banks. Some have more stable funding sources than others. Some banks focus on inherently safe loans more than others. It’s not impossible to understand some banks.

In this post, I will discuss why some banks can be a great long-term investment. I will also discuss what I like and don’t like about Wells Fargo.


Good Banks Can Grow Profitably for Many Years

The banking industry is very durable and predictable. Deposit growth tends to follow GDP growth. Over the last 20 years, total deposits of all FDIC-insured institutions compounded 6.2% annually.

Good banks can do better than that. The long-term trend is that there are fewer and fewer banks. The total number of FDIC-insured institutions declined from 11,970 in 1995 to 6,509 in 2014. That means deposits per bank can grow faster than GDP growth. The group of banks with more than $10 billion in assets grew much faster than groups with fewer assets. That can be explained by acquisitions. But the fact is that in most local markets, the top 5, 10, or 15 banks as a group tend to increase total market share over time. So, a durable bank can have a high chance of growing more than 5-6% organically. There aren’t many businesses that have high single digit growth forever. But that’s very possible for good banks.

The banking industry has lower costs than other types of lenders. Banks get funding from noninterest-bearing deposits, interest-bearing deposits, and other interest-bearing liabilities. The rates banks pay for interest-bearing deposits and other liabilities is very stable as a fraction of the Federal Fund Rates (FFR). For example, the cost of interest-bearing deposits for most banks is between 60% and 90% of FFR. Some deposits are more expensive than others. For example, the cost of money market account can be 70% of FFR, and the cost of time deposits can be 90% of FFR. The total cost of funding for banks is usually lower than FFR. The estimated cost of funding as % of FFR for some banks I looked at is:

Frost: 42%

Commerce Bancshares: 48%

First Financial: 49%

Prosperity Bancshares: 53%

BOK Financial: 59%

Wells Fargo: 64%

U.S. Bancorp: 67%

The industry data isn’t available but these are banks with lower than average funding. We can assume that cost of funding for the industry isn’t far from 80% of FFR. So, if FFR is 3%, the cost of funding is 2.4%.

The average net operating cost of banks with $1 billion to $10 billion in assets is 1.8%, and of banks with more than $10 billion in assets is 1.1%. These two groups are representative of the industry because banks with $1 billion to $10 billion of assets hold 10% of total industry deposit, and banks with more than $10 billion in assets hold 80% of total industry deposit. So, the industry’s average net operating cost is less than 1.5%.

That means the industry’s total cost of money is less than 4% if FFR is 3%. That’s lower than risk-free rates and is surely lower than the required rate of return for most money market funds or bond funds.

So, the banking industry has lower cost than other lenders. But strong rivalry among banks forces yields to go up and down along with the cost of funding. Therefore, the industry’s net interest spread, defined as yield minus cost of funding, was very stable at about 3.5%, as shown in the following graph:


The U.S. banking industry's net interest margin had a fairly stable tendency to be about 3.5% from 1999-2014.

Good banks can make 15-20% after-tax ROE. For example, Wells Fargo’s total cost is about 2.5% of total asset assuming 3% FFR. That means a very modest yield of 4% can result in 1.5% pre-tax return on earning assets. With 10x leverage, ROE will be 15% pre-tax and 10% after-tax. A normal yield of 6% results in 3.5% pre-tax return on earning assets or 23% after-tax ROE. FFR can go up and down, but Wells Fargo tends to maintain at least 1% cost advantage over the industry. That means Wells Fargo’s ROA is always at least 1% more than the industry’s average ROA. That guarantees an above average ROE.

Some Banks Have Low Risk

The main argument against banks is leverage. But not all liabilities are equal. Deposits are not debt. Deposits are more like insurance float. Even better, deposits tend to increase a lot in financial crises because people want to conserve cash. Some good banks are considered “safe havens” and thus gain deposit market share in bad times.

Short-term liabilities are what cause liquidity problems. Many banks rely on commercial paper or repo. According to a McKinsey report, deposits were just 49% of liabilities of U.S. banks in 2012. Commercial paper, repo, senior debt, and other liabilities made up 39% of liabilities. So, these banks can have big trouble in renewing short-term borrowings in a crisis. That’s very different from a bank like Frost whose deposits represent 96% of total liabilities.

So, the inherent risk in banking is liquidity. The key for investors is to avoid banks that rely on borrowings other than deposits. Regulators watch some capital ratios but banks with a lot of short-term borrowing could still face liquidity problems in a crisis even if they have a lot of tangible equity to assets. Quality of funding sources is more important than capital ratios. Banks that rely on deposits for most of their funding are very safe. They won’t have a liquidity problem even if they incur some loan losses.

However, banks have to maintain regulatory capital ratios. So, it’s important to avoid loan losses.

I found that loans have different risk profiles. For example, consumer loans tend to have more losses – offset by higher yield – than business loans. That’s perhaps because of government programs that encourage lending to home buyers or students. Among business loans, commercial real estate (CRE) loans tend to have more losses than commercial & industrial (C&I) loans. That’s because there is more speculation in real estate development than in other industry.

So, the types of loans a bank makes are as important as a conservative culture in minimizing losses. For example, it’s widely accepted that Wells Fargo has conservative lending. Yet, its average net charge-offs over the last 20 years was 1.09%. That’s way higher than Frost’s 0.48% or BOK Financial’s 0.27%. Frost and BOK make much fewer consumer loans than Wells Fargo.

Overall, banks that have stable funding sources and that make inherently lower risk loans can be very safe. I propose a checklist for good banks:

1. High deposits/liabilities (80% or more)

2. High noninterest-bearing deposits/total deposits (30% or more)

3. Low operating cost (1% or less)

4. High C&I loans/Total loans (60% or more)

My View on Wells Fargo

When I first looked at Wells Fargo, I thought that it can make $60 billion pre-tax earnings. The idea was simple. The median Net Interest Income/Earning Asset from 1991 to 2014 was 4.45%. Net Operating Cost was 0.57% in 2014. Net Operating Cost has been declining over time so it makes sense to use the latest number instead of the past median number. So, Wells Fargo can make 4.45% - 0.57% = 3.88% pre-tax return on earning assets (ROEA). As of the last quarter, Wells Fargo had $1.55 trillion in earning assets. $1.55 trillion times 3.88% equals to $60 billion.

Another approach gives a similar result. Wells Fargo’s cost of interest-bearing liabilities as a % of FFR was stable at about 88% with a variation of 0.20. Wells Fargo’s net interest spread, defined as yield minus cost of interest-bearing liabilities, was very stable at about 4.52% with a variation of 0.13. So, at a normal 3% FFR, cost of interest-bearing liabilities would be 3% * 88% = 2.64%, and yield would be 2.64% + 4.52% = 7.16%. Free funding sources are 27% of total earning assets so weighted average net interest spread is 5.23% (= 27% * 7.16% + 73% * 4.52%). Subtracting net interest spread by 0.84% average net charge-offs/earning assets and 0.57% net operating cost results in 3.82% pre-tax ROEA. That implies $59 billion pre-tax income.

Wells Fargo’s current market cap is $275 billion. So, Wells Fargo is trading at only 4.6 times normal pre-tax earnings.

That makes Wells Fargo the cheapest bank I’ve ever seen.

Even if it takes 10 years for Wells Fargo to make a normal 3.82% ROEA, investors can still make 13% annual return based on today’s price. Assuming 5% growth, Wells Fargo would have $2,520 billion earning assets in 2025. Applying 3.82% ROEA results in $96 billion pre-tax earnings. That’s 10.6% annual growth from last year’s $35 billion pre-tax earnings. Adding 2.8% dividend yields results in 13.4% annual return. There can be also some multiple expansion and share buyback.

Wells Fargo has characteristics that I like.

It has a low cost of funding. Noninterest-bearing deposits are 32% of total deposits. Noninterest-bearing deposits are only 20-25% of total deposits at most banks.

It has an extremely low net operating cost of 0.57%. That’s the lowest I’ve ever seen. I looked at about 50 publicly traded regional banks and most have between 1.6% and 3.3% net operating cost. Wells Fargo has low net operating costs thanks to cross-selling. For example, its retail bank cross sells on average 6.17 products per household. That generates a lot of fee income.

Wells Fargo has strong consumer brand awareness. All national banks have this advantage but Wells Fargo is better at cross-selling than any other bank. It’s likely that Wells Fargo can gain market share over time. In other words, it can grow deposits by more than 5% annually.

Wells Fargo is conservative. For example, it exited subprime lending in 2004. Most of its problems during the Great Recession were related to the loans acquired from Wachovia. Its lowest pre-provision earnings before tax (2.43% in 2014) covers 1.4 times the highest net charge-offs (1.76% in 2010.)

But there are things I don’t like about Wells Fargo.

Wells Fargo is very big in mortgages. The mortgage origination business is okay. This business requires scale and Wells Fargo funds one out of every three home loans in the U.S. The problem is that it retains about 10% of the loans it makes. Mortgage loans are 21% of earning assets. I’m not comfortable with these loans.

There are some adjustable-rate mortgages (ARM). Can customers afford interest expenses of these loans if interest rates increase? It’s likely that loans made after 2008 are very safe. Mortgage loans have been so restricted after The Great Recession. For example, total mortgage credit availability index was just 125.5 in July 2015 compared to almost 900 in July 2006. Regardless, a higher interest rate is still a concern for ARM.

Fixed mortgage loan can limit upside. Consumers took advantage of the low interest rates to refinance mortgage loans. So, Wells Fargo may have to carry a low yield portfolio for a while. For example, the yield on 1-4 family first mortgages was 4.19% in 2014, which is way lower than 7.27% in 2006.

The securities portfolio is another concern. Securities totaled $322 billion or about 21% of total earning assets. Mortgage-backed securities (MBS) totaled $124 billion. Wells Fargo expects the value of MBS to decline by $8.2 billion if interest rates increase by 2%. It doesn’t give the same expectation for the whole securities portfolio. But it can possibly lose over $35 billion if interest rates increase by 3%.

This is just a paper loss. The weighted-average expected maturity of securities available for sale was just 6.6 years. So, Wells Fargo can hold these securities until maturity and incur no actual loss.

This potential loss is only relevant because of regulatory capital ratios. Basel III requires banks with more than $250 billion assets to include marked-to-market gains or losses in the calculation of capital ratio. Assuming a $40 billion loss, tier I ratio would be 7.6%, which is well above the requirement of 6%. Supplementary leverage ratio, which takes into account off-balance sheet exposures, would be 5.9%, which is higher than the required 5%. So, Wells Fargo passed my stress test.

My last concern is that Wells Fargo is a “too big to fail” bank. Even if it will never have a liquidity issue, it can be forced by regulators to get a cash injection, resulting in share dilution. The book “Too Big to Fail” by Andrew Ross Sorkin tells the story about the protest of Wells Fargo’s CEO Kovacevich against the TARP:

Dick Kovacevich, for one, was obviously not pleased to have been given this ultimatum. He had had to get on a flight—a commercial flight, no less—to Washington, a place he had always found contemptible, only to be told he would have to take money he thought he didn’t need from the government, in some godforsaken effort to save all these other cowboys?

“I’m not one of you New York guys with your fancy products. Why am I in this room, talking about bailing you out?” he asked derisively.

And Henry Paulson responded with a threat of a regulatory crackdown:

You’re going to get a call tomorrow telling you you’re undercapitalized and that you won’t be able to raise money in the private markets.

Well Fargo is a good bank. It’s very cheap. But it’s too big. Its loan portfolio is too complicated. And it gets too much attention from regulators. Many people follow the stock. They may know things that I don’t know about Wells Fargo. So, I prefer smaller banks that have a simpler balance sheet.

Get Your Free Issue of Singular Diligence

Talk to Quan about Misconceptions of Banks

Current Price/Appraisal Ratios for All Our Past Stock Picks (That We Still Believe In)

by Geoff Gannon

I was reading an interview with the mutual fund manager Wally Weitz when I noticed he kept mentioning the “price to value” of his portfolio. He calculates what he thinks the intrinsic value of each holding in his portfolio is. Then he compares the market price to his appraisal price. This gives him an updated valuation of his portfolio in terms of his own appraisal of each stock’s intrinsic value.

For our newsletter, Quan and I pick one stock a month. We end each issue with an “appraisal price” for that stock. So, it’s easy to calculate today’s price as a percentage of our original appraisal price for each of our past picks.

You can see that calculation in the table below.

Stocks with a price/appraisal percentage in green have an adequate margin of safety. Stocks with a price/appraisal percentage in yellow are somewhat undervalued. Stocks with a price/appraisal percentage in red are overvalued.

Here are all our past picks for Singular Diligence that we still believe in:


Stocks with an adequate margin of safety (34% or more) are in green.

We consider two of our past picks to be mistakes: Town Sports (CLUB) and Weight Watchers (WTW). They do not appear in this table as possible stocks for you to consider (as we don’t think you should consider them – they’re simply too risky to recommend).

Note: As of today, Quan and I don’t own any shares of CLUB. But, we do both still own shares of WTW. So, we’re being hypocrites in regards to Weight Watchers. We haven’t sold the stock ourselves. But, we’re not suggesting anyone else should buy it. You can decide for yourself whether you should pay more attention to our money or our mouths on that one.

Also, Life Time Fitness was picked but does not appear in this table. It went private. So, you can’t buy it anymore. Sorry.

Finally, the “original appraisal prices” for the two parts of the Babcock & Wilcox spin-off (BWX Technologies and BW Enterprises) are after the fact re-calculations. The original issue gave one appraisal price for all of the old Babcock & Wilcox. That stock split into two separately traded parts after we wrote the issue. However, it was relatively easy to re-calculate the values we would have assigned each part had they been separate on the day we wrote the issue. So, Quan and I consider the price/appraisal you see for those two parts of the former Babcock & Wilcox to be accurate.

At today’s prices, we really do think BWX Technologies (BWXT) is an overvalued stock and BW Enterprises (BW) is an undervalued stock.

I’ll have a post about “Good Babcock” and “Bad Babcock” for you soon.

Get Your Free Issue of Singular Diligence

Talk to Geoff about Price/Appraisal of Various Stocks

Indicators of a Good Business

by Quan Hoang

I recently had a constructive debate with my friend about the return on invested capital (ROIC). I said that we don’t calculate ROIC for fun; we calculate it to know what return retained earnings can make. High return on retained earnings means good business. He shot back that See’s Candies has little volume growth and it’s still a good business. His point led me to the broader topic of what a good business is.

In a nutshell, a good business can create value. In other words, it can generate more than 10 cents for each $ of earnings it retains – assuming a 10% hurdle rate. But there are special cases in which a company can make more profits by retaining zero or negative earnings.


Exceptional Businesses Have Negative Invested Capital or Pricing Power

One special case is negative invested capital. Omnicom is a good example. It pays for advertising spaces slower than it bills clients. Working capital is about -20% of sales. The negative sign means that Omnicom gets 20 cents pre-funding from clients for each additional $ of sales. If growth is stable, a business with negative invested capital deserves a higher than average multiple of EBIT.

Another special case is exceptional pricing power. See’s Candies has exceptional pricing power. From 1972 to 1998, See’s Candies raised price per pound by about 6.9% annually. Inflation over this period was about 5.4%. So, pricing power generates about 1.5% real growth each year. That leads to margin expansion. This magnitude of pricing power is rare because the product becomes more expensive relative to a customer’s purchasing power overtime. That’s not sustainable in most cases. But See’s Candies has been able to do so for many years.

Another good example of pricing power is luxury Swiss watches. Swiss watchmakers managed to reposition mechanical watches from a utility product to an emotional product. But after that repositioning, it’s difficult to raise price faster than inflation. To do so, a brand must move upmarket and become more exclusive. Omega, on the path to regain its past prestige, has raised price from Longines’s price range closer to Rolex’s price range.

Without exceptional pricing power, value is normally created through volume growth. Volume growth normally requires additional investment in production/service capacity and working capital. Value is created only if return on investment is high.


How to Calculate ROIC

A practice that many analysts use is to say that a business is good if it consistently make a high ROIC. Joel Greenblatt’s formula for ROIC is EBIT/NTA. NTA is Net Tangible Assets, which is the sum of net fixed asset and net working capital. There are several versions of ROIC. But all version use net fixed assets in calculating the denominator. And that creates some controversies.

Joel Greenblatt explained why he uses net fixed asset:

Why are we taking Net Fixed Assets (NFA)? It is not always right.  Say we buy a hotel for $10 and it is going to last 10 years and we write it down over 5 years and now it is at $5. But if this goes down to zero, I might have to invest another $10. This would give me ($5) a skewed return (being too high) because of not considering replacement and reinvestment into the fixed assets.

Say you have 100 hotels and they are all on different cycles, then on average, you will be correct in using NFA. 10% of your hotels will be refurbished each year over a 10 year normal cycle. That is my quick and dirty for an ongoing business.


Denominator is NWC + NFA--why using net and not gross fixed assets?  On average that is the right thing to do.  Because in general what happens to your fixed assets, you buy something and you depreciate the assets so the value of your asset goes down, but to maintain your asset, there has to be on-going capex.  Depreciation and Capex cancel out (assume Deprec = Maint. Capex).  If capex is more than depreciation, then FA will increase accordingly and you will be updated.   If you are in expansion mode, you build new stores and the FA balloon before you earn on those assets, so your ROC will decline--so you must normalize or adjust for that. Fixed Assets minus depreciation plus Maint. Capex is why I use a Net number.

There’s some logic in his argument. But he didn’t examine how accurate EBIT/NTA is as a measure of ROIC for an ongoing business. If we own the 100 hotels in his example, we get cash flow roughly equal to EBITDA each year (assuming no tax). 10 hotels are totally depreciated each year. We can choose not to make any refurbishment at all and let EBITDA decline by 10% next year. We can refurbish 10 hotels and maintain EBITDA. Or we can refurbish and build 10 more hotels to grow EBITDA by 10%. In either case, ROIC of each new build or refurbishment will be based on the $10 gross investment in each of these projects because that’s what we have to spend upfront.

Let’s take another example. The Fresh Market (TFM) spends about $4 million in a new store, which generates about $10 million sales and $1 million EBITDA. TFM remodels its stores every 10 years. The remodel cost is lower than $4 million in real term but let’s assume the remodel cost to be $4 million. So, annual depreciation is $0.4 million and EBIT is $0.6 million.

A very optimistic assumption is that the store requires no remodel. So, the $4 million upfront investment results in $1 million annual cash flow forever. That translates into 25% annual return (25% = ¼). Realistically, there’s remodel cost every 10 years. So, 25% is the ceiling of ROIC. Generally, ROIC is always lower than EBITDA/Gross NTA. (Gross NTA = Gross fixed assets + Net working capital.)

A very conservative assumption is that we set aside “DA” each year. In the TFM example, we set aside $0.4 million each year so that after 10 years we have $4 million to spend on remodeling. That way, we’ll have $0.6 million free earnings each year (the “free” part is borrowed from the term free cash flow). So, the $4 million upfront investment results in 15% annual return (15% = 0.6/4). Realistically, we don’t set aside $0.4 million each year but use that money to fund new store openings. So, 15% is the floor of ROIC. Generally, ROIC is always higher than EBIT/Gross NTA.

If we open Excel and calculate IRR for various scenarios, we can see that IRR tends to be in the upper end of the range between EBIT/Gross NTA and EBITDA/Gross NTA. The midpoint of the range is quite a good estimate of ROIC.

Using EBIT/NTA is dangerous when fixed assets are a big part of NTA. I made that mistake when I first looked at Town Sports International (CLUB). Median EBIT/NTA was 20%, which looks good. But median EBIT/Gross NTA was 9% and median EBITDA/Gross NTA was 19%. So, pre-tax ROIC is around 14% instead of 20%. That’s a mediocre return.

We must be flexible when estimating return. We have to look at composition of NTA. It’s okay to use EBIT/NTA when PPE is a tiny part of NTA because the error is small. If PPE is a big part of NTA, using the midpoint of EBIT/Gross NTA and EBITDA/Gross NTA is preferable. If receivables are a big component, we should make adjustments. For example, America’s Car-Mart (CRMT) has $324 million receivables, $34 million inventories and $34 million PPE. However, Car-Mart doesn’t really lend money. Car-Mart lends cars. So we should adjust receivables to (1-gross margin) * receivables to estimate the total value of the cars it lends and use that number to calculate NTA. A better method to estimate ROIC is to look at the economics of each loan.


Return on Incremental Invested Capital (ROIIC)

What we really want to know is ROIIC rather than ROIC. We can calculate ROIIC by taking incremental EBIT or EBITDA over incremental invested capital over a 1- to 3-year period. That’s not a good approach. Sometimes a company has excess capacity so growth doesn’t require fixed investment for a while. Or sometimes a company has excess working capital and it can take capital out. But these examples are short-term adjustments. In the long run, volume growth requires investment in new production/service capacity and in working capital. So, ROIC is a good starting point to estimate long-term ROIIC.

Reinvestment in the same business tend to achieve returns similar to past ROIC. That’s why many businesses have ROIC within a certain range. However, we need to make some adjustments to ROIC to have a fair expectation of ROIIC.

Margin expansion can make ROIIC higher than ROIC. Margin expansion is usually a result of volume growth that drives down unit cost. For example, when gross margin is high and SG&A is relatively fixed, volume growth will significantly increase EBIT margin. Tom Russo usually uses Brown-Forman to illustrate the concept of the capacity to suffer. Brown-Forman is willing to incur expenses today to build infrastructure for international growth tomorrow. And the next 50,000 bottles it sells will have better margin than the last 50,000 bottle.

Frost (CFR) is another good example. Frost’s branches grow deposits faster than inflation. So, operating expenses per $ of deposit declines overtime. Gross margin in the banking industry is net interest spread. Net interest spread is influenced by interest rates and demand for loans. It’s cyclical but very stable over a long period of time. So, lower operating expenses per $ of deposits improve ROA. Today, Frost makes lower ROA than it did in the past. But without the impact of low interest rates, Frost should be able to make much better ROA.

We must be careful when volume growth is outside of current goodwill. In such case, high ROIC in the past doesn’t guarantee a high return on reinvestment. See’s Candies wasn’t able to grow profitably in other states because it failed to replicate the mindshare it had in California. TFM is a current example. TFM is a gourmet food chain. Consumers shop at traditional grocers most of the time. But in some special occasions, they may go to TFM for very good foods. Consumers on average go to TFM only once a month. TFM wants to be the first choice retailer for “special.” So, unlike other grocers, TFM relies on mindshare instead of habit. TFM is very strong in the Southeast. It got into trouble in recent years when it expanded into new markets. It’s very difficult to create mindshare in a totally new state. But perhaps it’s easier to open the next store in that state because the first store helped build some awareness and word of mouth.



The term “good business” is perhaps too broad. A firm that achieved high growth and great return but have little growth potential in the future isn’t as good as its past success suggests. Firms that barely made profit in the past might now be done with the investment phase and will enjoy great profitability in the future. What investors care about is perhaps more specific: a good business to buy. I propose 3 indicators of a good business to buy. The first is negative invested capital. The second is exceptional pricing power. The third is high ROIC. Past ROIC is a good benchmark for ROIIC. But to have a fair expectation, we need to consider other factors like whether margin of additional units will be higher and whether volume growth is inside current goodwill.

Get Your Free Issue of Singular Diligence

Talk to Quan about the Indicators of a Good Business

Operating Efficiency in the Banking Industry

by Quan Hoang

The idea for this blog post started with my curiosity about economy of scale in the banking industry. I performed a small research to study this subject. But I ended up talking about the broader topic of operating efficiency.

My research is based on data I collected from FDIC Quarterly Banking Profile, financial data of Wells Fargo (WFC), perhaps the best bank in the U.S., and data of all Texas banks that have more than 15 years of data. Banks in Texas tend to be run more conservatively than the industry so this is a good place to look at.

These Texas banks are:

Cullen/Frost Bankers (CFR): $28 billion asset

Prosperity Bancshares (PB): $22 billion asset

Texas Capital Bancshares (TCBI): $17 billion asset

International Bancshares (IBOC): $12 billion asset

First Financial (FFIN): $6 billion asset

Southside Bancshares (SBSI): $5 billion asset


The Right Efficiency Metric Is Based on Earning Assets, Not Revenue

Analysts usually use the efficiency ratio. It’s the ratio of noninterest expense to total revenue. However, I think this metric is irrelevant to a bank’s efficiency. A low-cost bank that makes very safe loans and thus less interest income can have a high efficiency ratio (analysts associate high efficiency ratio with high cost).

To compare cost, I use Noninterest Expense/Earning Assets. Earning Assets include short-term investments like money at the Federal Reserve, securities, and loans. We can also use Noninterest Expense/Deposit but Earning Assets tend to be proportional to Deposit. I use Noninterest Expense/Earning Assets because the data is more widely available. Let’s call it “Operating Cost.”

There’s one problem with Operating Cost. A bank like Wells Fargo sells a lot more financial products than other banks so it has more noninterest expenses. Using Operating Cost is unfair for Wells Fargo. So, I also look at Net Operating Cost.

Net Operating Cost = (Noninterest Expense – Noninterest Income)/Earning Assets.


The Data Tells a Different Story

I found a similar trend when I typed data of Wells Fargo and Frost. Both have lowered Operating Cost over time.

Frost’s Operating Cost from 1991 to 2014 was: 5.77% 5.82% 5.40% 5.00% 4.84% 4.78% 4.83% 4.85% 4.83% 4.94% 4.87% 4.48% 4.07% 4.13% 4.09% 4.02% 4.08% 4.10% 3.86% 3.49% 3.33% 3.02% 2.92% 2.74%.

Wells Fargo’s Operating Cost from 1991 to 2014 was: 5.31% 6.29% 6.59% 6.20% 5.70% 6.12% 6.04% 6.75% 5.70% 5.75% 5.49% 5.16% 5.40% 4.96% 4.96% 4.99% 5.12% 4.33% 4.47% 4.71% 4.48% 4.31% 3.81% 3.43%.

Over the period, Wells Fargo grew its earning assets from $37 billion to $1.4 trillion, and Frost grew its earning assets from $2.7 billion to $24 billion. The data seems to indicate that banks have lower cost when they become bigger.

But it’s not that simple. Data of other banks started making me confused. For example, Prosperity is much smaller than Wells and Frost but has much lower cost. Its Operating Cost from 1996 to 2014 was: 2.79% 2.81% 2.76% 2.70% 2.81% 2.71% 2.53% 2.30% 2.25% 2.34% 2.11% 2.53% 2.47% 2.24% 2.09% 1.97% 1.81% 1.74% 1.85%

In 1996, Prosperity had only $238 million in earning assets, or 1/15 of Frost’s $3.8 billion, but Prosperity had 1.99% lower operating cost. In 2014, Prosperity had $18 billion earning assets, or 2/3 to Frost’s $24 billion, but its operating cost advantage had declined to 0.89%.

Of the 7 banks that I collected data on, 4 banks including Wells Fargo, Frost, Prosperity, and First Financial have consistently reduced their operating cost. Three banks including Texas Capital, International Bancshares, and Southside haven’t really reduced their operating cost.

In summary, operating costs of the 7 banks are now:

Prosperity Bancshares: 1.85%

Texas Capital: 2.26%

International Bancshares: 2.64%

First Financial: 2.71%

Frost: 2.74%

Southside: 2.99%

Wells Fargo: 3.43%


And NET operating costs are:

Wells Fargo: 0.57%

International Bancshares: 0.97%

Prosperity Bancshares: 1.16%

Frost: 1.40%

First Financial: 1.40%

Texas Capital: 1.92%

Southside: 2.26%


So, the questions are:

1. Why did some banks reduce cost while others didn’t?

2. How can smaller banks have lower operating cost if economy of scale is strong?

Regarding #2, it’s true that this isn’t an oranges-to-oranges comparison. Different banks have different business models and different cost cultures. For example, Frost offers greater services than other Texas banks. It makes more Commercial and Industrial loans than other Texas banks. C&I loans have a very long sales cycle. So, Frost’s business model suggests a higher cost base. But I can’t help doubting economy of scale when looking at the data.


Big Banks Are Efficient in Selling Many Financial Products

The industry data gave me some clues. I collected operating cost of all FDIC-insured institutions categorized by asset size. There are 4 categories: banks over $10 billion in assets, banks with between $1 billion and $10 billion in assets, banks with between $100 million and $1 billion in assets, and banks below $100 million in assets. We have comparable data back to 2002.

Operating cost was pretty flat in all categories as shown in the following graph:

Once a bank has $1 billion in assets, getting bigger may not lead to lower costs

The graph shows that banks with more assets tend to have lower operating cost. But banks with more than $1 billion in assets have pretty similar operating cost. Specifically, the median operating cost of each category is:

Below $100 million in assets: 3.98%

$100 million to $1 billion in assets: 3.48%

$1 billion to $10 billion in assets: 3.37%

Over $10 billion in assets: 3.42%

The data suggests some economy of scale but it’s not significant above $1 billion. But net operating cost shows the advantage of big banks more clearly. Median net operating cost of each category is:

Below $100 million: 2.63%

$100 million to $1 billion: 2.23%

$1 billion to $10 billion: 1.83%

Over $10 billion: 0.98%

Big banks make much more noninterest income than smaller banks. Big banks seem more efficient in selling other financial products. But putting this advantage aside, I still doubt that economy of scale is significant on the cost side. Prosperity is a good counterexample. In 2006, it had only $3.7 billion earning assets but its net operating cost was just 1.19%. Its operating cost of 2.11% was incredibly low despite its small size.


3 Factors in Operating Efficiency

I think there are 3 key factors in having low cost

1. Size

2. Unit-level economy of scale

3. Culture

Factor #1 was discussed above. Economy of scale on the cost side is significant when size is below $1 billion. Above $1 billion, its benefit is unclear on the cost side but obvious on the noninterest income side.

Regarding factor #2, unit-level economy of scale means efficiency gained by having high deposit per branch, high deposit per account, or high local market share, etc.

Frost is a great example of reducing costs by growing deposit per branch faster than inflation. Frost’s deposit per branch declined from $110 million in 1994 to $79 million in 1995 because it acquired some branches with low deposits. However, from 1995 to 2014, deposit per branch grew consistently from $79 million to $196 million. That’s a 4.9% annual growth over a 20-year period. And $196 million per branch is very high. Wells Fargo averages just $134 million per branch. Most banks average much less than $100 million in deposits per branch.

The clearest leverage from higher deposit per branch is occupancy cost. For Frost, occupancy cost as a percentage of earning assets has declined from 0.54% in 1995 to 0.23% in 2014. But there’s also leverage of other operating expenses. For example, JP Morgan Chase (JPM) makes about $1 million pre-tax profit per retail office. Frost made $3.7 million pre-tax profit per branch in 2014. Using a total hypothetical to illustrate the point, assume a branch manager costs $100,000 in salary, JPM would have to pay 10% of its branch’s pre-tax profit to its branch manager while Frost has to pay only 2.7%. If a branch manager costs $50,000 a year, Chase would be paying 5% of branch profits to its manager while Frost would be paying 1.4% of its profits.

I actually think high deposit per branch is more important than total size in driving down costs for banks with over $1 billion in assets. Frost offers much greater services than other banks and it usually has lower fees. But high deposit per branch allows it to have a competitive net operating cost.

First Financial matches Frost’s net operating cost of 1.40%. But it isn’t big, and it doesn’t have high deposit per branch. It has only $6 billion in assets and averages $77 million deposit per branch. However, First Financial tends to have 30-40% market share in very small markets. So, I suspect that high local market share is part of the reason why it has low cost. I think high local market share can result in efficiencies in support system.

To be fair, Frost’s account-related fees as a percentage of deposit are about 0.4% lower than First Financial’s. And Frost offers greater services than First Financial. But both have 1.40% net operating cost. So, Frost is actually more efficient than First Financial. Therefore, I think deposit per branch is a stronger force in reducing cost than local market share.

International Bancshares doesn’t really have low costs. Its net operating cost is very low at 0.97%. But its account-related fees are about 1.58% of total deposit. That’s about 1.12% higher than Frost and Prosperity Bancshares. Adjusting for this gap, International Bancshares actually has pretty high cost. That leaves Prosperity Bancshares as the most efficient bank in Texas. Actually, Prosperity Bancshares is well known in the industry for its low cost.

Prosperity averages $72 million in deposits per branch. That’s not high. Prosperity actually grows through acquisitions so growing deposit per branch is perhaps not its top strategy to reduce costs. I think Prosperity has high market share in some markets but I doubt its local market share is anywhere as high as First Financial’s. So, I think the reason for its low cost is culture.

Prosperity is really focused. It isn’t interested in selling other financial products. It just focuses on growing deposit and the loan portfolio, and maintaining a lean operation. American Banker wrote an article about Prosperity’s CFO David Hollaway, who “tracks every penny.” Prosperity’s CEO once joked that “David doesn’t pay for anything.” David Hollaway even cut things like hot chocolate, popcorn, or facial tissue for employees.



To examine a bank’s operating efficiency, I suggest a 3-question checklist:

1. Does it have high noninterest income?

2. Does it have high deposit per branch?

3. Does it have a low-cost culture?

As shown by big banks, noninterest income is a great way to reduce net operating cost. Quality of noninterest income is important. I’m not interested in account-related fees. I would love to see low account-related fees (good for customers) and a lot of income from other financial products such as trust, investment, and insurance brokerage.

Deposit per branch is an important factor in branch economics. High deposit per branch helps reduce cost at the branch level. It’s more important than total size-based economy of scale or local market share.

I use “culture” for the lack of a better word. It may mean a CEO’s action or CFO’s action. In the case of Prosperity, it actually acquired many banks and its CFO cut excess expenses during the integration process.

Culture is hard to judge. In theory, it doesn’t result in a durable low-cost advantage because competitors can try to cut costs too. In practice, it can be difficult for other banks to cut costs to extremes like David Hollaway. Many managers may never question the kind of “unnecessary” expenses that David Hollaway cut. So, culture can be a durable advantage. It’s just harder to logically prove than the advantage from things like high deposit per branch.

Finally, any qualitative study must be backed up with quantitative evidence. Fortunately, industry data is widely available for U.S. banks. So, it’s pretty easy to study a bank’s operating efficiency.

Talk to Quan about Operating Efficiency in the Banking Industry

Subscribe to Singular Diligence

UniFirst (UNF): Maybe Too Expensive; Maybe Just Right

by Geoff Gannon

Here’s a stock trading for 1.4 times sales. We’re sure of that. What we can’t be sure of is whether it’s trading at 10 times normal pre-tax profits or 14 times normal pre-tax profits.

That word “normal” is the problem.

UniFirst provides uniforms and protective clothing to American and Canadian businesses of all sizes. These businesses typically sign a 3 to 5 year contract. UniFirst then personalizes, cleans, and delivers whatever uniforms the business needs. The ongoing task is basically showing up at a customer location once a week to deliver fresh uniforms and collect the dirty ones.

Quan and I have probably talked about most publicly traded uniform and textile rental companies in the U.S., U.K., and E.U. at some point. Sadly, they haven’t been cheap enough for us to buy. We like the industry.

If capital allocation is good and the stock is not clearly selling at a premium price – we’d be willing to consider buying almost any of them.

At the right price.

We’ve decided that “right price” is 10 times pre-tax profits.

Luckily, UniFirst does trade for about 10 times pre-tax profit. However, the price is closer to 14 times pre-tax profit if normalized a certain way. I’ll explain that “certain way” in a second – but first an aside.

When we investigate a business in depth we come up with a unique way of normalizing earnings that is appropriate to that company. For example, Hunter Douglas made $200 million last year but we think it can make $300 million in a normal year and $350 million in a good year for housing. That’s not surprising because its sales are lower in both the U.S. and Europe than they were in 2006 and 2007. Its market share isn’t. The U.S. market for blinds and shades should in a cyclically normal year – assuming the same real prices per window covering and the same demand for window covering per person – be more than 25% higher now than it was 10 years ago. That’s because of population growth and inflation. It’s an easy estimate to calculate. And I’m confident in it. America isn’t going to have a lot fewer windows per person. And blinds and shades aren’t going to cost a lot less in real terms. So, in the case of Hunter Douglas we were aggressive in saying that future earnings will be much, much higher than any year from 2008-2014. That’s a no brainer.

UniFirst’s earnings are not as simple to normalize.

Our standard way of normalizing the earnings of a company we know nothing about is to simply take the most recent year’s sales and multiply that by the median EBIT margin over as many years of history as we have for the company. This is far from perfect. But, it’s also very good at eliminating cyclically overearning stocks from our list. In recent years, UniFirst has had a 13% return on sales. Today, it’s up to a 14% EBIT margin. However, if you study the company’s long-term past (for about the last 20 years) you’ll find that the median return on sales for those years is just 10%. So, current earnings might overstate normal earning power by up to 40%.

That’s a big mistake for an investor to make.

In this case, it’s basically the difference between paying a P/E of 15 or a P/E of almost 22. We like the industry. But, uniform rental isn’t the ad agency business or something. You don’t want to pay 20 times earnings for one of these companies.

To be fair to UniFirst, the S&P 500 shows a pretty similar spike in return on sales. Starting in 2009, UniFirst’s operating margin jumped from around 10% to about 13%. Gross margin moved – but not as decisively or consistently. This is exactly what has happened at a lot of big public companies in the U.S. Taking the S&P 500 as a whole – there has been a reduction in selling, general, and administrative expenses. Gross costs are not lower than in the past. It’s entirely possible that companies got bloated during the 1990s and 2000s. When the crisis hit, these public companies were most concerned with growing EPS and shareholder value and therefore slashed operating expenses – like employees working at corporate – to the bone. That’s a theory. There could be other explanations. But, stopping cap-ex can reduce depreciation a little. Firing people reduces SG&A. And not increasing salaries as quickly as your sales increase, also reduces SG&A. So, lower cap-ex and lower employment and lower salaries than are normal can all reduce SG&A relative to sales. These could all be reactions to low demand.  

Since 2010, UniFirst's operating expenses have grown much slower than sales.

To give you some idea of what I mean, UniFirst’s SG&A as a percent of sales was 24.6% last year. Ten years earlier, it had been 27.1%. So, we have an improvement in return on sales of 2.5% due to a reduction in SG&A. That may sound small. But, consider that UniFirst has $1.44 billion in sales now. So, we are talking about $36 million of cost savings. If the stock normally trades for 10 times pre-tax profits (about a P/E of 15), that is $360 million of value created through getting lean. That’s $18 per share of added value. Again, this isn’t unique to UniFirst. You can find a lot of public companies in America with this same pattern of reducing operating – not gross – expenses faster than sales since the financial crisis hit.

A lot of people email me saying that corporate profits don’t have to “mean revert”. Companies can have higher operating margins than they did in past decades. It’s possible. But, I think we should remember that American workers don’t really make much more money than they used to. And yet they do output quite a bit more than they used to. That obviously benefits employers. In industries that were historically unionized or closed to foreign competition – I don’t disagree with the idea that employers can be in a permanently stronger position when bargaining with employees than they used to be. But a great many companies Quan and I look at were never unionized and don’t really compete much with companies using labor outside the U.S.

There is one other very good reason for why SG&A could be permanently lower. Companies could use information technology to lower the amount of people they need in staff type functions. This invests in capital and economizes on labor. Certainly, a lot of tasks performed by humans in past decades can be performed by computers now.

However, I think it’s difficult to separate cyclical cost savings due to cutting fat from your organization during a crisis from permanent cost savings due to technology. I would caution that in the U.S. you have businesses making unusually good profits while workers are not. Since businesses are the ones who pay workers – I think it’s really important to stress the cake cutting between employers and employees is part of what determines profit margins.

This is why we need to be especially careful when looking at companies that have higher EBIT margins now than they did before 2008. UniFirst is one such company. And whether it is overearning or not is the key to deciding whether or not the stock is worth buying at anywhere near today’s price.

UniFirst is not a stock you want to pay more than 10 times normal pre-tax profits for. The right multiple for a business is determined by the cash profitability of that business. Companies that can both grow and pay out a lot of cash are worth a lot. Companies that can’t are not worth more than the average business.

As a long-term buy and hold, UniFirst has two things going against it. One, it needs to invest in working capital as it grows. Two, it makes acquisitions. As a result, UniFirst does not pay out much in dividends or buy back any stock. The company has a wonderfully stable EBIT margin from year to year. Sales are also stable. So, EBIT is predictable.

UniFirst is a very consistent business – as are most companies in this industry. It has the kind of consistency in profits that you see at John Wiley (JW.A) or Omnicom (OMC). What it doesn’t have is the free cash flow generating ability of those businesses. John Wiley and Omnicom can have P/Es of 20 and yet their annual returns can match an index fund with a P/E of 15 – because they can grow organically while also paying dividends and buying back stock. Since the early 1990s, Omnicom grew sales per share by 11% a year. It didn’t grow its net tangible assets at all. In fact, they shrank from a deficit of $200 million to a deficit of $2 billion.

As an ad agency, Omnicom can get its customers to finance its growth. As a uniform rental company, UniFirst can’t.

And so while UniFirst is a very consistent and adequate performer – it doesn’t have especially desirable cash flow dynamics. It can provide a 10% type return on equity year after year which can lead to 10% type returns in the stock for the long-run. It’s outperformed the S&P 500 over the last 30 years. So, I can’t say it’s not an above average business. But, I am going to say it’s not worth an above average price.

Imagine we can’t settle the question of whether the normal EBIT margin for UniFirst is 10% or 14%. If that’s true, then we can’t be entirely sure if we’re paying something like 10 times pre-tax profits (a P/E of 15) or something like 14 times pre-tax profits (a P/E of 22). And it’s very important in this case not to pay a premium to 10 times pre-tax earnings.

Also, UniFirst does some uniform business in areas with oil drilling. It’s possible drilling activity in North America over the last 5 years has skewed UniFirst’s results in a way we can’t appreciate.

In defense of the stock, it is unleveraged and this is the kind of business you can leverage up. Based on the stability of UniFirst’s EBIT, it can support a lot of debt. Cash flow is not very stable versus reported results. However, peers with less stable earnings have a bit more debt.

UniFirst is definitely a stock worth keeping an eye on. But, I’m not sure it’s possible to have confidence the stock is cheap enough to provide anywhere near the 11% a year annual returns it delivered over the last three decades. With stock prices so high right now, it might make sense to settle for buying a maybe 40% overvalued UniFirst or maybe not overvalued at all UniFirst. Even if you are overpaying by 40%, I would expect long-term holders of UniFirst to do better than the overall stock market.

Quan and I will consider UniFirst if it gets cheaper. We’d love to buy the stock at one times sales. Today, it sells for 1.4 times sales.

At today’s price, we’re not interested.

Talk to Geoff about UniFirst (UNF)

Subscribe to Singular Diligence

Driverless Cars and Progressive's Durability

by Geoff Gannon

Value and Opportunity linked to a Bank of England blog I never would have found on my own. The Bank of England blog did a post on how driverless cars could hurt the future of auto insurers. Last year, we did a Singular Diligence issue on U.S. car insurer Progressive (PGR). A big part of the durability section of that issue was about driverless cars.

So, here is the Bank of England blog post on driverless cars.

And here is Singular Diligence’s discussion of Progressive’s durability…


Originally Published: December 2014

DURABILITY: Progressive’s Focus on a Combined Ratio of 96 or Lower Makes it Durable

Auto insurance is a durable industry. The only risk of obsolescence is driverless cars. Car accidents are caused by human error. If all cars on the road were driven by computers – there would be virtually no car accidents. This would eliminate the need for auto insurance. The technical difficulties of developing driverless cars are not the biggest obstacle to their adoption. Even much simpler safety technologies like front air bags, side air bags, electronic stability control, and forward collision avoidance generally took 10 years from the time they were first introduced on a car sold to the public till the majority of new models sold in a given year included these features. So, the “tipping point” of safety feature adoption by manufacturers is usually around a decade. Complete adoption takes about 15 years. The average car in the U.S. is about 11 years old. This number has increased over time. Cars are more durable now than they were in the past. Based on these figures, it is likely that once the first driverless car is introduced by a major auto maker on a popular model it will take another 15 to 20 years before half of all cars are driverless. 

Auto insurance is required by state law. States will certainly not eliminate this requirement while the majority of cars are still driven by humans. Total adoption of the technology could take up to 30 years. If enough car owners prefer to drive themselves instead of letting a computer drive their car for them, there could be resistance to any laws limiting human drivers. Without such laws, highways would include a mix of human and computer driven cars. Under such conditions, laws might still equally “fault” driverless cars for accidents involving human drivers. These legal complications mean that auto insurance would probably persist into the early stages of a mostly driverless car society.

Today, there are no commercially available driverless cars. So, the end of car insurance would likely be some point 15 to 30 years after the successful introduction of driverless cars. The vast majority of net present value in a stock comes from returns generated within the first 30 years. Even if driverless cars are successfully introduced in the U.S. soon – and that is a completely speculative assumption – it is very likely that auto insurance will persist as a legal requirement for car owners for at least the next 15 to 30 years. So, even if the eventual adoption of driverless cars is a certainty – the durability of car insurers as a long-term investment is still sufficient to generate good returns for today’s investors. The shift to a driverless society is far enough in the future to justify an investment in Progressive right now.

The greatest risk to Progressive’s durability is underwriting error. Progressive writes more insurance – assumes the risk of more losses – relative to its surplus (the capital buffer available to absorb losses) than other auto insurers. One way of judging the underwriting leverage of an insurer is to look at its premiums relative to its equity plus debt (its capital). Progressive writes 2 times its capital in premiums. First Acceptance writes at 1.7 times. Infinity at 1.4 times, Mercury at 1.3 times, Safety at just 1 times, while other insurers – with large non-auto businesses – like Travelers and Chubb write at well below their capital. Underwriting leverage is only a problem when an insurer’s combined ratio – its losses and expenses divided by its premiums – exceeds 100. Companies with underwriting losses in a normal year must be very careful not to write too much insurance relative to the capital that can absorb those losses. 

To understand the risk in Progressive, it is critical to understand the concept of a combined ratio. Insurers generate a “return on sales” (sales are called premiums in the insurance industry) in two ways. One: the policyholder pays more to the insurer than the insurer pays out in corporate expenses, commissions, advertising, and losses. Two: the insurer makes money by investing the premiums paid upfront by its policyholders in securities like common stocks, preferred stock, corporate bonds, municipal bonds, and federal government debt. Different insurers try to make their money in different ways.

Historically, Progressive has generated more than half of its return on sales from its underwriting. This is unusual. In a normal year, the average insurer loses money on its underwriting. But it more than makes up for that by investing its float. Progressive earns a lot from underwriting relative to other insurers. It earns little from investing. And Progressive takes much less investment risk than other insurers. In 2013, Progressive was 75% in bonds and these bonds were actually short-term government debt due in 2 years or less. In the last 20 years, Progressive’s only major investment loss – when the company had more losses than gains on investments for the year – was during the 2008 financial crisis. Progressive held preferred stock in big banks. The company marked these securities to market. Progressive did not realize actual losses on the preferred stock. After the banks were bailed out, they continued to make payments on their preferred stock and these securities rebounded fully in value in the years since. Given today’s conservative investment policy, the investing side of Progressive’s business does not present any risks to the company’s survival even under crisis conditions worse than 2008.

All of the long-term risk in Progressive comes from the underwriting side. Because Progressive takes in double its capital base in premiums each year, any underwriting loss would lead to a hit double that magnitude relative to capital. For example, in 1991 and 2000 Progressive had a combined ratio of about 105. This means the company had an underwriting loss equal to 5% of its sales. Because sales are twice capital, the company lost about 10% of its capital in each of those years. Obviously, investment gains offset some of this loss. Progressive maintains a ratio of debt to total capital of about 25% to 30%. When debt is 30% of total capital, a 10% destruction of capital causes a 14% destruction of equity. This is because debt only absorbs losses after all of a company’s equity has been impaired.

Shareholders should focus on the amount of underwriting losses relative to equity that Progressive can cause in any one year. Assuming premiums are double capital and equity is 70% of capital, it would take a combined ratio of 112 to destroy a third of Progressive’s equity (12% * 2 = 24%; 24%/0.7 = 34%). Theoretically, it is not difficult to imagine a scenario where Progressive’s underwriting loss forced the company to raise capital by issuing stock and diluting its shareholders. In practice, Progressive’s culture minimizes the risk of large underwriting losses relative to the company’s capital cushion.

Progressive has a 96% combined ratio target. It has been remarkably consistent in averaging a combined ratio below this target. Since 1991, the company’s average combined ratio was 92.6%. In the last 20 years, the average was 92.3%. In the last 15: 92.6%. In the last 10 years: 92.5%. And over the last 5 years: 93%.  Since 1991, Progressive has failed to hit its 96 combined ratio 4 times. The company’s combined ratio was 103.6 in 1991, 96.5 in 1992, 98.3 in 1999, and 104.4 in 2000. Progressive has yet to miss its 96 combined ratio since the turn of the millennium. Some of this consistency in underwriting may be due to pricing data. Progressive updates its prices faster than any other auto insurer. It is usually the first company to raise prices. 

The most important element in Progressive’s combined ratio is not competence. It is culture. The company never changes its stated goal of growing as fast as possible while keeping a combined ratio of 96. It has always said that any growth above a combined ratio of 96 must be avoided. 

Here is what Progressive’s CFO said about the 96 combined ratio target in 2013: “(We) often get asked the question, ‘Would you consider changing your 96 combined ratio target?’ Certainly, in the most recent environment with lower interest rates, would we consider changing the combined ratio target? The simple answer to that is no. We feel that it served us well in a number of cycles, with economic cycles, (and) underwriting cycles. And for us it creates a good balance between attractive margins and competitive rates for customers. It’s important that we meet those profitability targets because we are more leveraged to underwriting results…At the end of last year, our premium to equity was close to 2.7 to 1. A peer set of other…companies…were closer to 1 to 1….This combination of disciplined underwriting, ensuring we meet our profit targets, and leverage the underwriting results is how we create good returns for shareholders.”

Talk to Geoff about Progressive’s Durability and Driverless Cars

Check Out Singular Diligence

You Can Afford to Hold Cash

by Geoff Gannon

In my last post, I said stocks were too expensive. Instead of putting more of your money into diversified groups of stocks, you should just let cash build up in your brokerage account.

A lot of people have a fear that those lost years of making zero percent on their idle cash can never be made up for.

I’ve created a graph to show how much ground you’d have to make up.


Let’s say you have two choices: one is to invest in an overpriced basket of stocks today and hold that basket from 2015 through 2030. This choice will compound your 2015 money at a rate of 6% a year.

The second choice is to do nothing for all of 2015, 2016, 2017, 2018, and 2019. You just hold cash. That cash earns 0% for those 5 years. In 2020, you finally get an opportunity to make an investment that will return 10% a year from 2020 through 2030.

If your investment horizon extends all the way out from today through 2030, the second approach overtakes the first approach about 15 years from now.

Doing nothing for 5 years and then something smart for 10 years is a better 15 plus year strategy than “just doing anything” today.

Here we define something smart as 10% a year and “just doing anything” as 6% a year. You can decide for yourself whether your something smart is 10% a year or not. That's subjective. What the "doing anything" returns is a lot more objective. So, let's talk about that.

Over the last 15 years, the S&P 500 returned about 5% a year. During that time period, the Shiller P/E ratio contracted from 43 to 27. The same percentage contraction – 37% – would be required to get the Shiller P/E down from today’s 27 to a historically “normal” 17.

I see no reason why the S&P 500 should do better from 2015 to 2030 than it did from 2000 to 2015. That means I see no reason why buying the S&P 500 today and holding it through 2030 should be expected to return more than about 5% a year.

(Almost all readers I talk to have a total return expectation for the S&P 500 that is greater than 5% even for periods shorter than 15 years.)

It’s also worth mentioning that while I have no predictions as to when idle cash would earn more than zero percent – the Fed does. And those predictions show cash earning a few percent in 2018 and 2019 instead of zero percent.

For those reasons, the graph in this post is probably an underestimate of how quickly sitting and doing nothing till you can do something smart outperforms continuing to shovel cash into the S&P 500 at today’s prices.

I think the reason people don’t feel secure in waiting for an opportunity to do something smart is that they’re not sure when that opportunity will appear.

Maybe there will be no chance in all of 2015, 2016, 2017, 2018, 2019, 2020, or even 2021 to do something smart. If that’s true – isn’t it possible doing anything now could outperform waiting to do something smart later? If that later is sometime after 2021 – couldn’t it be better to just buy the index today?


I can only point to history.

Pick any year in the past. Then move forward 6 years from that time. In the intervening years, was there an opportunity to do something smart?

The hardest waiting period in history was during much of 1995 through 2007. Although stocks were often cheaper than they are today – the largest and best known American stocks were almost always more expensive than they had been at any time before 1995.

I think this is the real reason why investors I talk to are hesitant to hold cash. Much of their investing lifetime was spent during a time of high stock prices.

There is no advantage in buying something that is unlikely to provide a good long-term return instead of holding cash till something good comes along. If we take 15 years as long-term, we can say that the S&P 500 will not provide good long-term returns if bought today.

You can afford to avoid 5% a year type long-term commitments if you have a real chance at finding 10% a year type long-term commitments sometime in the next 5 years.

You don’t need to know exactly when or where this opportunity will come.

A lot of investors who live outside the U.S. read this blog. They have an advantage. Their home country’s stock market might provide a 10% a year opportunity sometime in the next 5 years. American investors probably won’t notice such an opportunity when it appears.

By buying into an index today, you are really saying you will just take whatever price Mr. Market gives you. You do this because you’re not sure he will ever give you a good price again. Or, if he does, it may come far more than 5 years in the future.

Caving into Mr. Market’s mood is not something value investors think is appropriate when it comes to individual stock purchases. Yet, a lot of the people who read this blog – who are otherwise value investors – feel they have no choice but to continuously add to the actively and passively managed mutual funds in their brokerage account.

The other choice is to hold cash. And the longer “long-term” is for you, the more sense holding cash makes.

It makes a lot of sense right now.

Talk to Geoff about Holding Cash

Check Out Singular Diligence

Stocks Are Too Expensive

by Geoff Gannon

We talk about stock picking on this blog. That means we usually talk about specific stocks. The “market of stocks” not the “stock market”. Today, I’m going to talk about the stock market.

It’s too expensive.

You shouldn’t buy it.

If you have an account where you automatically reinvest your dividends – stop. If you are putting money each month into an index fund, or a stock mutual fund, or a bond mutual fund – stop. Those assets are overpriced. Any basket of stocks or bonds is overpriced. If you are saving money regularly – that newly saved money should now be going into cash instead of stocks or bonds.

The simplest rule in investing is that you never buy an obviously overpriced asset. Stocks generally and bonds generally are obviously overpriced right now. So, you need to stop buying them in a general way.

To put a number on this expensiveness, I think the Shiller P/E ratio is about 27 now. It was about 27 when I wrote my December 2006 post arguing stocks were too expensive. You can read that post later down in this one. Or you can click here to see - via the Wayback Machine - what that post actually looked like on the original site in 2006.

I am writing this post because of 3 separate items I noticed recently.

I came across one while reading an earnings call transcript for Frost (CFR). This is a usually conservatively run bank in Texas. It has a lot more deposits than loans. Deposits have kept growing. So, the company needs to put the money somewhere. And where they’ve put it is “Securities”. Frost now holds more money in securities than loans. These securities are high quality. They aren’t going to default. But they are overpriced. To get a yield near 4% on their securities portfolio – the company had to go pretty far out in terms of the maturities it would buy. In normal economic times – let’s say with a Fed Funds rate of 3% to 4% – these bonds would cost less than what Frost paid for them. At some point, there will be a 3% to 4% Fed Funds rate. I have no idea when that will be. You can look at predictions from the FOMC’s own members and see they thought it would be 3 years down the road or so. Now, if that’s true – you obviously aren’t gaining much by making less than 4% a year for less than 3 years if you will be able to make 4% a year on idle cash at the end of that period. Of course, some events may happen that prevent any increases in the Fed Funds rate for that entire 3 year period. In the 1930s in the U.S. and in the 1990s and 2000s in Japan, investors could have easily overestimated the likelihood that rates would rise within the next 3-5 years to a “normal” level. If something like that happens and you keep all your money at the Fed instead of in long-term municipal bonds and such – you’d have missed out to the point where you now still have $1 when you could have more like $1.12.

You can afford to miss out on those kind of returns. I actually think Frost can too. But, this isn't a post about Frost.

The other two examples don’t involve an actual investor. They are about the “cost of capital”. The car lock maker Strattec uses an Economic Value Added (EVA) approach. They are funding the company with equity right now instead of debt. So, their cost of capital is the cost of their equity capital. They use 10% as the cost of equity capital. Equity investors aren’t going to get 10% a year from this moment forward. Returns will be closer to 5% a year. So, Strattec is really overcharging itself for capital when it presents EVA in the annual report.

And then the last example is a Morningstar analysis I read. The analyst adjusted the value of the company up a little based on lowering the cost of capital for the company – which also uses only equity capital – from 10% a year to 9% a year. This is a concession to the reality that investors are bidding up stocks. But, the cost of equity capital is not 9%. The S&P 500 is not priced to return anywhere near 9% a year. If companies want to issue dilutive stock or borrow long-term – none of that will actually cost them 9% a year.

I understand why Frost, Strattec, and Morningstar don’t spend a lot of time saying today’s stock and bond prices are much higher than stock and bond prices have been through most of history. But not harping on that can make people forget how abnormal today's stock and bond prices are.

I think there is a big danger of complacency here. Investors seem to be pretending today’s prices are comparable enough to past prices that it’s not worth focusing on. At many points in the past, you could make close to 10% a year in stocks. Today, you can’t. And at many points in the past, it was safe to buy bonds at the market price and not expect a very large drop in their market price. Today, it’s not.

Both the likelihood of 10% returns in stocks and the unlikelihood of large paper losses in bonds was due to their prices. They were lower. As a group, stocks and bonds are the same assets they always were. Increases in the price of those groups simultaneously lowers long-term future returns and increases the risk of short-term negative returns.

A little bit later in this post I’m going to give you the entirety of something I wrote back on December 29th, 2006. That was about 8 and a half years ago. If you had stayed completely in the Dow from that moment till now rather than staying completely in cash from that moment till now the difference would be like 5% a year. Of course, you shouldn't have stayed in cash for 8 years. You should have stayed in cash till prices were "normal" again in late 2008-2010. Listening to Shiller or Grantham or this blog or any value investor would've told you prices were okay again once the crash happened.

I thought stocks were too expensive in December of 2006. The Fed Funds rate went to 0% and stayed there. Stocks are – by the Shiller P/E and other such normalized measures – a lot more expensive than they’ve ever been except for years like 2007, 1999, and 1929. So, all of those factors have helped stock returns from the end of 2006 to midway through 2015. And yet returns were no better than the about 5% or 6% a year I warned was likely back in 2006. They were not the often hoped for 9% or 10% a year that people cite as the “cost of equity” and the return investors in stocks expect long-term.

When you might earn 10% a year in the stock market – the cost of not participating is high. When the best you can hope for is 5% or 6% a year – as the period from high stock prices in 2006 back up to high stock prices again in 2015 shows – you aren’t missing much by sitting out till you get an acceptable price.

I am not saying you shouldn’t pick stocks. If you find a business you like at a price of less than 10 times normal EBIT – you can buy that business. That’s a good price in all environments.

So, you can pick absolute bargain stocks. That means a business you like at less than 10 times EBIT. The danger is settling for relative bargains. If the market trades for 15 or 20 times normal pre-tax earnings - then I can pay 13 times for this business and it's a steal. That's dangerous thinking. What you're really saying is that you can never hold cash. The best you can do is to buy something a bit cheaper than the very high price everything else happens to be priced at right now.

Obviously, you should stop contributing more cash to stock funds, bond funds, etc. Stop reinvesting your dividends. Build up cash till you find a bargain for all times – not just for these very expensive times.

I’m not going to spend the rest of this post arguing about today’s stock market level. It’s clearly too high. And future returns will be much worse than past returns. But, I’ve found it is hardest to argue about the present. It is easier to use an illustration from the past which can serve as an analog for today.

So, I am re-posting my December 29th, 2006 piece “In Defense of Extraordinary Claims”. In that post, I argued that:

“Normalized P/E ratios can fall in several ways. However, there are only two ways that seem reasonable given current conditions. Stock prices can either fall over the short-term or they can grow slowly (at less than 5-6% a year) over the long-term.”


“Stocks are not inherently attractive; they have often been attractive, because they have often been cheap. The great returns of the 20th century occurred under special circumstances – namely, low normalized P/E ratios. Today's normalized P/E ratios are much, much higher. In other words, the special circumstances that allowed for great returns in equities during the 20th century no longer exist.

So, don't use historical returns as a frame of reference when thinking about future returns – and do lower your expectations!”

Just about 8 and a half years later, I want to reiterate those same two points. They are as true now as they were at the end of 2006. We are in the same place. Stocks are too expensive again. They can either drop a lot in the short-term. Or they can rise at less than 5% or 6% a year for the long-term. So, when you ask “Should I hold cash?” instead of adding to my mutual funds, reinvesting my dividends, etc. you should think of 3 possible outcomes: 1) I buy stocks and the market crashes in the short-term 2) I buy stocks and they return less than 5% to 6% for the long-term 3) I hold cash instead of buying stocks.

Because those are the only 3 reasonable outcomes. I’m not suggesting you time the market by selling stocks you already own. Nor am I suggesting you stop buying stocks that are good purchases in any market. I don’t think knowing that stocks are too expensive means you have to sell what you already own. Nor do I think it means you have to give up on buying businesses you like at less than 10 times EBIT. That’s always a good decision.

But, knowing stocks are too expensive right now should lead to you cutting off all additional contributions to your mutual funds and index funds till prices return to their normal historical range.

What is that range?

That’s the question I tried to answer in my 2006 post. Here is that post – completely unedited – and just as relevant in 2015 as it was in 2006.

I have one added note. In what you're about to read you'll see I used my own measure of the "normalized P/E ratio" for the Dow rather than the Shiller P/E for the S&P 500. It doesn't matter which you use. I'd just go with the Shiller P/E myself - and I'm the one who made up the measure you're about to see. Both normalized P/E ratios will usually tell you about the same thing at about the same time.

Shiller uses an inflation adjusted 10-year average. I use a 15-year average with a 6% nominal annual escalator in EPS.

For those who care about this stuff: My method was to apply a 6% growth rate to each of the last 15 years of earnings. So, to predict "normal" earnings in 2015 you simply project actual EPS for every individual year from 2000 through 2014 forward at a rate of 6% a year. You assume the average of all these "past projections" is more normal than what is actually reported as for 2015.


(“In Defense of Extraordinary Claims” – Originally Posted: December 29th, 2006)


About two weeks ago in a post entitled "We Have Some Bearish Bloggers Out There", Bill Rempel wrote, "Personally, I’m in the 'extraordinary claims require extraordinary proof' camp." I'd like to think I am too, because Bill is right – extraordinary claims do require extraordinary proof.

So, before making any extraordinary claims about future long-term market returns (i.e., predicting future returns that differ substantially from historical returns), I'd like to spend this post laying out the case for why current circumstances are extraordinary. After all, extraordinary times call for extraordinary claims.

Essentially, this is a post about why the present is unlike the past and what that means for the future.

In a previous post, I wrote:

Stocks are not inherently attractive; they have often been attractive, because they have often been cheap.

Unless they internalize this fact, investors risk assuming that historical returns that existed under special circumstances can continue to serve as a useful frame of reference, even when these special circumstances no longer exist.

Later in this post, I will discuss the possibility of a "paradigm shift" (i.e., a change in basic assumptions within the theory of investment) that began in 1995. The only other period in the 20th century which saw similar upheaval in investment thinking was the 1920s.

Common Stocks as Long Term Investments

That theoretical crisis (and the higher valuations that followed it) has often been partly attributed to a thin volume published in 1924 by Edgar Lawrence Smith. The book was called "Common Stocks as Long Term Investments" and it was based on a study of 56 years of market data (1866 – 1922).

Smith found that stocks had consistently outperformed bonds over the long run. Neither the data in support of this conclusion nor the logical explanation for this outperformance (public companies retain earnings and these retained earnings lead to compound growth) was wrong.

However, a few years after Smith's book was published, the special circumstances of the past disappeared as stocks (which had historically had higher yields than bonds) saw their prices surge and their yields plunge. Soon, stocks had lower yields than bonds – part of the reason for their past outperformance (the initial yield advantage) was gone and the margin of safety which a diversified group of stocks had offered over bonds narrowed considerably.

Simply put, circumstances changed. John Maynard Keynes saw this possibility when he reviewed Smith's book in 1925:

"It is dangerous…to apply to the future inductive arguments based on past experience, unless one can distinguish the broad reasons why past experience was what it was."

That has been the objective of this little study from the outset. In this post, I will focus on how the circumstances of the present differ from the circumstances of the past.

I will also endeavor to demonstrate that historical returns were the result of special circumstances, which (logically) need not apply now or in the future. The historical data suggests these circumstances may yet return – and for the sake of net buyers of stocks, I hope the data is right and one day (soon) historical returns can once again serve as a useful frame of reference for the future.

Today, however, historical returns have about as much utility to the investor as the success rate of a procedure performed exclusively on 25 year-old men has for the surgeon who is preparing to operate on a 92 year-old woman.

There is nothing wrong with the data itself. But, there is something wrong with the assumption that data collected from one special case has predictive power when applied to another special case.

Cheap Stocks and Great Returns

Historical returns in equities have been great. However, it's worth noting that throughout the period we're referring to, stocks have often been cheap. How cheap?

Once again, here's a graph of the Dow's 15-year normalized P/E ratio for each year from 1935-2006:


From 1935-2006, the Dow's normalized P/E ratio ranged from 6.88 – 30.84. The Dow's average (mean) normalized P/E ratio for these years was 14.18. The median was 13.91.

Those figures include the 1995-2006 period, which I will discuss in greater detail later. For now, let's start by taking a look at the period from 1935-1994.

Until 1995, the Dow's normalized P/E ratio had ranged from 6.88 – 17.40. The average (mean) normalized P/E ratio from 1935-1994 was 12.31. The median normalized P/E ratio was 12.41. In other words, the Dow's average 15-year normalized earnings yield was just over 8%.

I would estimate that in a little under 45% of all years, the Dow was priced such that long-term investors were effectively paying little or nothing for future earnings growth. Most market authorities would disagree with me on this point, because they would require an equity-risk premium.

Equity-Risk Premium – An Aside

This isn't the place to have a long argument about the concept of an equity-risk premium. For now, I will simply say that you can not arrive at the conclusion that there is an equity-risk premium via deductive (a priori) reasoning. If you were locked in a room alone, you would never come up with the idea of an equity-risk premium. It is only in seeing the effect that you would seek out a cause.

You can only come to the conclusion that an equity-risk premium should exist by first knowing that it has existed. You have to work backwards from the effect to the cause. That's troubling, because history consists of a series of special circumstances. It is non-repeatable.

So, the existence of a measurable aversion to stocks over some historical period does not necessarily lead to the conclusion that such an aversion is the result of a general principle (i.e., an inherent equity-risk premium). In fact, such a conclusion could merely be a contrived attempt to explain away an observable effect that has existed under certain circumstances – but needn't always exist.

The equity-risk premium isn't a general theory. It's really little more than the acknowledgement that during the historical period being studied, market participants made choices that reflect an aversion to stocks compared to the choices an optimal return seeking automaton would have made.

It's an interesting observation – but, it's not a theory.

How Common Are Cheap Markets?

Returning to the question of how often the stock market has been cheap, I would estimate that during the period from 1935-2006, the Dow was priced to offer double-digit returns somewhere between 75% and 85% of the time.

Here, I don't mean that the Dow did provide double-digit returns 75% to 85% of the time; nor, do I mean that past performance suggested it should provide such returns. Rather, I simply mean that valuing the Dow as an asset to be held until Judgment Day, would lead a clear-headed observer to conclude that double-digit returns were likely in about 75% to 85% of the years being considered.

I know this 75% to 85% number is a bit hard to swallow. So, if you don't believe me, consider what Warren Buffett wrote on the same topic in his 2002 annual letter to shareholders:

"Despite three years of falling prices, which have significantly improved the attractiveness of common stocks, we still find very few that even mildly interest us. That dismal fact is testimony to the insanity of valuations reached during The Great Bubble. Unfortunately, the hangover may prove to be proportional to the binge."

"The aversion to equities that Charlie and I exhibit today is far from congenital. We love owning common stocks – if they can be purchased at attractive prices. In my 61 years of investing, 50 or so years have offered that kind of opportunity. There will be years like that again. Unless, however, we see a very high probability of at least 10% pre-tax returns…we will sit on the sidelines."

Buffett's "50 or so years" of his 61 would translate into just under 82% of the time. He wrote that letter in early 2003. The four years since haven't offered the kind of opportunity he looks for, while the seven years included in the study from before Buffett started investing did offer that kind of opportunity.

So, according to my math, that would work out to be a roughly 80% estimate from Buffett over the full 1935-2006 period. That estimate falls within the 75% - 85% range I cited based on the data.

I think this 75%-85% range is the best estimate you'll find for how often the market has been so cheap as to offer double-digit returns when valued as an asset with a holding period of forever.

Unfortunately, I'm afraid a lot of investors don't realize (or haven't internalized) just how often the stock market has been really cheap. During the 1935-2006 period, stocks were priced as clear bargains in about 8 out of every 10 years. Buffett supports this conclusion with his assertion that stocks could be "purchased at attractive prices" about 80% of the time (50 out of 61 years).

If investors don't start with an understanding of the fact that stocks have been so cheap so often, they won't be able to put the historical data in its proper context. If you have a population that consists of 80% x and 20% y, is it reasonable to assume that data based on the entire population is a good reference point for your subject, if you know your subject is a y rather than an x?

In terms of valuation, 2006 (and thus 2007) is undoubtedly a minority year. Unfortunately, data based on a full population sometimes has little or no relevance when applied to a member of a minority group.

For instance, Turkey's population is 80% Turkish and 20% Kurdish. My guess is that data based solely on the full population of the country (which would consist of 80% ethnic Turks) would tell you very little about any particular Kurd. Now, if you broke the data you had collected down into a Turkish group and a Kurdish group and used the Kurdish group to predict something about an individual Kurd – then, you might be on to something.

A More Detailed Look

From 1935-2006, the Dow's normalized P/E ratio ranged from 6.88 to 30.84. The Dow's average (mean) normalized P/E ratio for these years was 14.18. The median was 13.91. In half of all years, the Dow's normalized P/E ratio fell between 10.53 and 16.43.

Here's a breakdown of how common various normalized P/E ratios were from 1935-2006.

Normalized P/E of 5-10: 18 of 72 years or 25.00% of the time

Normalized P/E of 10-15: 28 of 72 years or 38.89% of the time

Normalized P/E of 15-20: 17 of 72 years or 23.61% of the time

Normalized P/E of 20-25: 5 of 72 years or 6.94% of the time

Normalized P/E of 25-30: 3 of 72 years or 4.17% of the time

Normalized P/E of 30-35: 1 of 72 years or 1.39% of the time

Fifteen Years Later…

For the years with a normalized P/E ratio between 5 and 10, compound point growth in the Dow over the subsequent fifteen years ranged from 4.01% to 15.69%. The average (mean) growth rate was 10.17%. The median growth rate was 10.03%.

For the years with a normalized P/E ratio between 10 and 15, compound point growth in the Dow over the subsequent fifteen years ranged from 0.92% to 12.28%. The average (mean) growth rate was 7.01%. The median growth rate was 8.17%.

For the years with a normalized P/E ratio between 15 and 20, compound point growth in the Dow over the subsequent fifteen years ranged from (0.14%) to 8.93%. The average (mean) growth rate was 2.19%. The median growth rate was 1.76%.

I'd love to show you the same data for the three highest normalized P/E groups. But, I can't.

There is no fifteen year point growth data for years with a normalized P/E over 20, because the Dow didn't record a year with a normalized P/E ratio above 20 until 1996. In fact, until 1995, the highest normalized P/E ratio on record was 17.40 – that high-water mark was reached in 1965. With the benefit of hindsight we now know 1965 was not an ideal year to buy stocks for the long-run.

Rising Multiples?

Today's normalized P/E ratio is extremely high. So what? Hasn't the normalized P/E ratio been rising over time, as investors have come to realize a diversified group of stocks held for the long-run is actually a low-risk, high-reward bet?

I'll let you judge for yourself. I won't even connect the dots for fear of biasing you.

Here's a chart showing the Dow's 15-year normalized P/E ratio for each year from 1935-1994:


Do you see a trend towards higher normalized P/E ratios over time?

I cut the graph off at 1995 for a reason. That's the year everything changed. You'll remember I said the Dow's highest normalized P/E ratio had been 17.40 reached in 1965.

Although I didn't include the data necessary to compute 15-year normalized P/E ratios for years before 1935, I do have enough data to know that the three "peak" normalized P/E ratio years during the 20th century were 1929, 1965, and 1999.

By "peak" years, I simply mean the three highest years that aren't part of a chain of continuously higher normalized P/E years – unless they're the highest year in that chain. Without this qualifier, the highest normalized P/E list would be monopolized by the years from 1995 – 2006. Each year in that group had a higher normalized P/E ratio than every year prior to 1995.

In other words, since 1995, the Dow's normalized P/E ratio hasn't just been above the mean, it's been above the entire normalized P/E ratio range from 1935-1994. You can see that clearly in this graph, which shows the Dow's normalized P/E ratio for each year from 1935 – 2006:


This graph is essentially just a continuation of the earlier graph. In fact, if you cover the points from 1995 – 2006, you can see the familiar outline of that graph with its long undulations and its frothy crest at 17.40. That bound was reached in 1965. In 1995, the Dow broke out of this upper bound and hasn't returned since.

Terra Incognita

In this graph, it certainly does look like there's a trend toward higher normalized P/E ratios. However, that trend only emerged over the last decade – not the last century.

In other words, the Dow's normalized P/E ratio hasn't been rising over time. It simply surged in the 1990s. That surge may be justified. However, it's certainly a departure from the historical data. As a result, there's no reason to believe historical returns from 1935-1994 have any utility whatsoever in predicting market returns in the new era that has emerged since 1995.

All the historical return data from before 1995 was based on lower normalized P/E ratios. Once again, I don't mean the pre-1995 period had lower average normalized P/E ratios – I mean that no year from before 1995 had a normalized P/E ratio equal to or greater than any year from 1995 through today. Simply put, since 1995, market valuations have been in completely uncharted territory.

The only years with normalized valuations comparable to today's occurred during the 1995-2006 period. So, referring to historical return data requires a choice between using data from recent years or using data from dissimilar years.

Paradigm Shift?

Is it possible that the surge in normalized P/E ratios beginning in 1995 was simply the culmination of a crisis within the investment discipline? Maybe normalized P/E ratios have reached "a permanently high plateau" now that a new paradigm has taken hold.

I won't dismiss this argument entirely. There is some logic to it. After all, stocks have been an unbelievable bargain for most of the 20th century. Why should that continue to be the case? Eventually, won't enough investors wise up to this fact and cause the so-called "equity-risk premium" to disappear.

If the normalized P/E ratio remains extremely high, there will be no need for stock prices to fall. Of course, these higher valuations must necessarily cause future returns to fall short of historical returns. But, there's no logical reason why normalized P/E ratios must revert to the mean – future returns can be adjusted down, allowing current prices to remain high.

That's true. In fact, the Dow could theoretically trade around a normalized P/E ratio as high as 40-50 without making stocks so unattractive as to completely eliminate them as a possible long-term investment (all of this assumes the equity-risk premium can disappear).

At around 50 times normalized earnings, the math gets terribly unforgiving. As a result, it's hard to imagine any likely circumstances under which a market trading at close to 50 times normalized earnings could be a viable investment option – though it's theoretically possible if long-term interest rates are very, very close to zero.

But, at lower normalized P/E ratios, such as 30 (and certainly 20) stocks could still compete with other investment opportunities. Stocks might lose most (or all) of their edge over other asset classes; but, stock prices wouldn't necessarily have to fall – they could simply offer much lower returns than they had in the past. This could continue indefinitely – in theory.

I say "in theory", because that seems a rather unlikely scenario. There is absolutely no evidence for it in the data.

Before 1995, the Dow's normalized P/E ratio had ranged from 6.88 – 17.40. The average (mean) normalized P/E ratio from 1935-1994 was 12.31. The median normalized P/E ratio was 12.41.

So, a permanent jump to normalized P/E ratios above 20 would be quite a departure from the past. Could the leap be permanent? Could these new, higher normalized P/E ratios become the new norm?

Maybe. If we really are in a new era, the old historical return data isn't relevant – it applies only to an era of low normalized P/E ratios. New, higher valuations must necessarily lead to new, lower returns. On the other hand, if we aren't in a new era, the old historical return data is relevant – and normalized P/E ratios must fall.

Adjusting to the Norm

Normalized P/E ratios can fall in several ways. However, there are only two ways that seem reasonable given current conditions. Stock prices can either fall over the short-term or they can grow slowly (at less than 5-6% a year) over the long-term.

The data from 1935-2006 doesn't provide much support to one route over the other. In the past, extraordinarily high normalized P/E ratios have been brought down to more normal levels through crashes and through stagnant markets.

The market can reach a more "normal" normalized P/E ratio by going down fast or going sideways for a very long time. During the 20th century, we saw normalized P/E ratios fall both ways.

To return to the 1935-1994 normalized P/E range, the Dow would need to trade around 10,135. That would simply bring it down to a valuation comparable to 1965.

To return to the average normalized P/E ratio for 1935-2006, the Dow would need to trade around 8,260. If the Dow were to trade at the average normalized P/E ratio for the 1935-1994 period, it would need to trade around 7,230.

Are any of these numbers likely destinations? The truth is stocks have probably been too cheap in the past and they're probably too expensive today. Regardless, the Dow has been above 1965's old normalized P/E high since 1995. So, for a little over a decade now, the market has been in uncharted territory. A normalized P/E ratio of 20-25 (today's is about 21.50) is quite compatible with decent long-term returns for stocks relative to other asset classes.

However, such high normalized P/E ratios are not compatible with the kind of long-term returns seen during much of the 20th century.


Stocks are not inherently attractive; they have often been attractive, because they have often been cheap. The great returns of the 20th century occurred under special circumstances – namely, low normalized P/E ratios. Today's normalized P/E ratios are much, much higher. In other words, the special circumstances that allowed for great returns in equities during the 20th century no longer exist.

So, don't use historical returns as a frame of reference when thinking about future returns – and do lower your expectations!

(End of December 29th, 2006 repost)

How to Judge a Business’s Durability

by Geoff Gannon

My last post listed examples of threats to a company’s durability. This post will be about how we assess those threats. You can always imagine a threat. Is it a realistic threat? How do you judge that?

There are some industries where durability is pretty much perfect. The business doesn’t change much. Barriers to entry are high. The future development of substitutes is unlikely. Location advantages are big.

A good example is lime. Lime is reactive and has a short shelf life. You don’t store it speculatively. You don’t import it and export it. Customers need to get their lime from a deposit being worked somewhere within a few hundred miles of them. Over the last 100 or so years, the real price of lime hasn’t changed that much (real price volatility compared to other commodities is quite low). The price right now is perfectly in line with the real average price per ton since 1900. Lime consumption in the U.S. was no higher last year than it was in 1998. The industry is more consolidated and perhaps less competitive than it was in 1998. I don’t think capacity is being fully utilized now. And I do think inflation will always be passed on to customers (as it was over the last 100 years). So, if Quan and I were to research a company like United States Lime & Minerals (USLM), we could probably start by assuming that last year’s EBIT would – in real terms – represent that company’s durable earning power. That could be our starting point for a buy and hold analysis.   

That’s usually not the case. Even when we find a company that has a long history of being the leader in its field – say Strattec in car locks and keys, H&R Block in assisted tax preparation, etc. – there is a risk of change. In these two cases, we know there will be change in the product. For example, more people will prepare and file their taxes online in the future than they do now. And more drivers will enter and start their cars with the use of electronics instead of physical locks and keys. What we don’t know is how that will affect the companies.

Take H&R Block. The company competes in assisted tax preparation. In the 1990s and 2000s, many people switched to using software and then online products to prepare their taxes. But who were these people?

Most were people who had always prepared their taxes themselves. I use TurboTax and know a lot of people who use TurboTax as well. But, I actually don’t know anyone who used H&R Block even once in their lifetime and now uses TurboTax. Everyone I know who uses TurboTax used to – decades ago – prepare their taxes themselves using a pen and paper and a calculator. They didn’t use a CPA. And they didn’t use H&R Block.

Now, this is anecdotal. But, if I hadn’t asked the question “who are these people” I might have assumed that if tens of millions of people are switching to online tax preparation they are coming from companies that include H&R Block. Maybe they are. But, maybe they are a different segment of the market. If you ask that question: “Who are these people?” you can then start an investigation into how customers are segmented.

But, this still presents two problems. One, even if the migration to TurboTax and its competitors was all from do-it-yourself tax filers – that doesn’t mean that is the only group that will switch.

The other problem is that I just mentioned I use TurboTax – I don’t use assisted tax preparation. Most people I know don’t either. This means I will have a poor understanding of H&R Block’s core customer. I will make the mistake of assuming that the tax preparation market is made up of customers like me and people I know – when that’s probably half the market or less.

So, I won’t understand why some people get assistance preparing their taxes. And I won’t understand why these people don’t just switch to something like TurboTax.

This is a very common problem. We run into it all the time.

I’ll give you an example of a stock Quan and I wanted to pick for Singular Diligence – but the price got away from us before we could. It’s a U.K. company called “Greggs”. I live in the U.S. Quan lives in Vietnam. The last time I was in the U.K. was more than 15 years ago. The fast food industry in the U.K. has changed since then. So, we started from a position of real difficulty in understanding the customer. Quan and I are foreigners as far as the analysis of Greggs is concerned. And we’re not even there on the ground to do scuttlebutt. So, this was a tough stock to analyze. Now, we benefited from being in contact – via email – with more than one person in the U.K. So, we could have people visit multiple Greggs locations in the U.K. and report back to us. This was helpful.

But, there was another problem. A cultural one. The folks who write about stocks in the U.K. are not – it turns out – the folks who go to Greggs. The people in the U.K. who write about stocks and who work in that industry skew heavily toward being higher income and London based. London is part of the U.K. But, most of the U.K. isn’t London. And quite a lot of Greggs isn’t London.

So, there is a serious danger here. It’s particularly serious because Quan and I don’t live in the U.K. So, if analysts at an investment bank or a New York Times reporter or someone like that says something about America’s Car-Mart I can say to Quan: “That’s irrelevant. People in New York City know no more about Arkansas than you do. And none of these people know anything about not being able to buy a 9-year old used car without credit and not having a credit card to charge it to.”

I know that we need to get in touch with people who know more about the places where America’s Car-Mart operates and the people who buy cars there. I know not to trust a New York based source’s opinion about an Arkansas based business. And I know not to trust a source with a six-figure income about the need for sub-prime borrowing.  

That’s obvious to me when dealing with America’s Car-Mart because I live in the U.S. It’s a lot less obvious when dealing with a stock like Greggs.

This is the number one most important part of scuttlebutt: talking to the customer. To understand durability, we need to understand customer behavior.

Now, sometimes the customer lies or can’t articulate how exactly they make their choices. But, even then – I think they usually give away a lot. For example, when looking at why Weight Watchers members quit – a lot of customers will cite cost.

However, most customers will admit there are other reasons besides saving money. And they will often berate themselves in a way that makes it clear there is a gap between their words and their actions. Cutting out a monthly expense is a good excuse. But, really they want to quit because it is hard or they have already reached their weight goal or something like that. So, in the case of Weight Watchers some customers say they quit to save money – but we don’t believe them. Even though we don’t believe them, we still need to hear from them. Because we can still gain information both about why they claim to be quitting and why we think they are actually quitting from their own words.

For some businesses, Quan and I are able to get a very good explanation of customer behavior. For example, I think we have a pretty perfect 3 part model for how Americans choose which supermarket to go to:

  1. Convenience

  2. Selection

  3. Price

We’re pretty confident that you can explain the vast majority of customer defections in the supermarket industry in terms of a store’s failure of either convenience, selection, or price relative to another local option. In the case of U.S. supermarket shoppers, we can also say that “local” usually means about a 3 mile radius. This last claim has been tested in an academic paper. It is used by a major U.S. supermarket in its 10-K to explain the range in which they think competition occurs. And it is supported anecdotally.

These 4 assumptions were very important for us in deciding whether or not Village Supermarket was a durable business. Village Supermarket operates – mostly quite large – Shop-Rite supermarkets in Northern New Jersey, parts of Southern New Jersey, and now a couple stores in Maryland. New Jersey and Maryland are very densely populated places. New Jersey doesn’t grow much. Barriers to entry in the local markets where Village competes seem to be very high.

I’ll use an anecdote to illustrate. I grew up in a town that is within a Village store’s “circle of convenience”. I lived in that town for about 25 years. For most of those 25 years, you had 3 supermarkets to choose from. For part of those 25 years, you had 4 supermarkets to choose from. All the locations that were supermarkets when my parents moved to that town are still supermarkets today. Some are operated by different companies – but only because they bought the parent company. One location was added. The addition in capacity was much smaller than the increase in local population. The one new store was in a completely newly built shopping center that had previously been undeveloped land (which is quite rare in that part of the country). The existing stores in town invested in expanding square footage, parking, etc. to the extent this was possible.

So, this is an oligopoly where you don’t close the existing sites and you rarely add new capacity. You reinvest in the existing sites as much as possible. But, the number of suitable locations for a brand new supermarket of the ideal size is low. Village leases stores for 20-40 years. It owns some others. So, it’s not like suitable sites for a 60,000 square foot supermarket come up every day in Village’s region.

This information allowed us to ask questions about competition from Wal-Mart, online, etc. We could dismiss Wal-Mart right away. It’s harder to build a Wal-Mart than a traditional supermarket in Northern New Jersey. This market is tougher than the ones Wal-Mart normally competes in. And we knew that Wal-Mart draws from a much tighter “circle of convenience” for its grocery shopping than for its other product categories. Wal-Mart draws from exactly the same circle of convenience as traditional supermarkets. If a Wal-Mart opens 10 miles from a supermarket, it has no impact on the profitability of that supermarket. Remember, 10 miles is a 20 minute drive. You don’t actually average speeds better than 30 miles per hour in your local area. So, Wal-Mart is one threat to durability we did not take seriously.

The next is online groceries. Shop-Rites have competed with Peapod for like a decade now. If you compare the online groceries to in-store groceries in terms of convenience, price, and selection – online has no advantages. You need a scheduled time for delivery. So, it’s no more convenient. You need to tip. Given the size of the average grocery order and the tip people give, you are adding 10% or so to the price of your shopping trip. And, to date, online grocery selection has been narrower than in-store even for companies that use their stores as the distribution point for online.

Finally, there is The Fresh Market. This is a real threat to Village’s durability. The Fresh Market has the best business model for entry into the New Jersey grocery market. Its stores are smaller. The up-front capital costs are lower. The payback period is quicker. And it can siphon off high gross profit sales even if a customer uses The Fresh Market for perishables and Village for non-perishables. Margins are good in perishables. So, The Fresh Market is a big threat. The barrier to entry is lower for The Fresh Market than it is for Wal-Mart, Village, etc. You can put a Fresh Market where you would be unable to put a Village or a Wal-Mart. Generally speaking, a Fresh Market can be as small as half the size of a Village store while a Village store could be half the size of a Wal-Mart.

The Fresh Market has no advantages in price. And it has narrow selection in non-perishables. But, it can be quite convenient and have strong selection for the perishables shopping for a household. A lot of grocery shoppers make more than one trip a week of unequal size. It’s a real danger that a household will split its shopping between a traditional supermarket like Village and a perishables focused format like The Fresh Market.

So, we highlighted The Fresh Market as the biggest risk to Village in our Singular Diligence issue on the stock. And that’s really from thinking about customer behavior and barriers to entry. We disagreed with the arguments in favor of online groceries and Wal-Mart because those options don’t perform especially well in terms of convenience, selection, and price. And because it’s hard to put a Wal-Mart close enough to one of Village’s Shop-Rites to make a difference.

It’s also worth mentioning that opening a Wal-Mart in a local market wouldn’t necessarily have the long-term impact you might expect. If there are 3 supermarkets and you add a Wal-Mart, it’s entirely possible that the now 4th place store will eventually close. In most cases, Village wouldn’t be the operator of the marginal store. We just didn’t feel that the ratio of households to supermarkets in a local area was likely to change except if you added a Fresh Market. That was very likely to increase competition permanently in the town. Because we could easily see how The Fresh Market could enter a town and yet no one else would exit that town. And that would leave the incumbents worse off.

You can also see here that one reason why it’s easier for us to analyze Wal-Mart and online groceries is because they aren’t asymmetric with Village. Village can sell online groceries too. Wal-Mart doesn’t have an easier time adding a location in Village’s markets than Village itself does. And we know Village has a hard time adding locations.

Situations like Greggs and Weight Watchers are different. The competition people were suggesting would be a problem for those companies was positioned quite differently. With Greggs, we would get comments that people wouldn’t want cheap and unhealthy good. They would be willing to pay up for food that is healthier, fresher, etc. And I’m sure some people will. There was just a danger that we were hearing more from that segment of the total customer pool than from the segment that appreciated cheap and filling food.

So, we make a special effort to talk not only with customers of the industry – but some core customers of the company itself.

Western Union is another case where there was tough. Most information out there about Western Union – in the media, on blogs, among analysts, etc. – is written by people who are really, really far from Western Union’s core customers. They don’t have any use for the service. They don’t know people who do have a use for the service. And so they can have a lot of misconceptions.

Now, this one was easier for us because Quan lives in Vietnam and spent several years in the United States. A lot of people from Vietnam take up residence in the U.S. and elsewhere around the world and send money back to Vietnam. So, Quan knew lots of people who use Western Union and competing services.

We were able to talk to these customers. And they were able to explain their behavior. Sometimes, we wouldn’t have correctly imagined customer decision making without talking to them. For example, we would have underestimated the importance of the receiver in deciding which service to use. We knew this was important. But, until we spoke to customers – I don’t think we realized that for most senders they go with the service that the person receiving the money asks them to use. So, if you are sending money back to your mom – you use Western Union because she says there is a location she likes right around the corner. Also, until talking to customers – I underestimated the importance of convenience like the exact hours of the location and whether they will deliver the money to your door and things like that.

I think talking to customers is always the most important part of assessing durability. I also think it’s the most important part of scuttlebutt. People ask all the time if we talk to management. The answer is that we do when we can – but we’ve never found it that useful. I might be overstating that. But, right now, I can’t think of a single time where something a CFO – for example – said was more useful to us than information we got from somewhere else. We’ve quoted CFOs in Singular Diligence a couple times before – but really only because it was a nice, clean, concise quote to use. I can’t think of a time when they gave us information we found particularly useful.

That is not true of customers and the people at the company we’re researching who deal directly with those customers. We got really good information from store managers at America’s Car-Mart. We got good information from customers of Breeze-Eastern and their competitor UTC in helicopter rescue hoists. We get good information from dealers. Actually, independent dealers are probably the best source of information because they deal directly with both the company we’re interested in and with the end users and yet they aren’t employees of the company. Tandy was a very interesting case because Tandy’s biggest customers and biggest competitors are the same people. So, when they told you they bought something from Tandy they were also really telling you why Tandy could sell that particular thing economically and they couldn’t.

Until Majestic Wine made its change in direction by firing its CEO and acquiring an online wine seller – we were definitely going to write about that stock. And that’s another U.K. company. So, despite the difficulties of researching a foreign stock, it’s something we’re still willing to do. I should say researching a company with customers in another country. Because it’s not like we have any difficulty analyzing Ekornes – a Norwegian company – when it comes to sales in the U.S. And Western Union is a U.S. company. But, the receive side is almost always not in the U.S. So, at least half of every transaction is decided by a customer in another country. And the person making decisions in this country was often born in another country – so, Western Union can also be considered a case of difficulty understanding “foreign” customer behavior.

Now, I am going to contradict everything I’ve been saying up to this point. So far, I’ve said the most important part of judging durability for us has been talking to customers. I said we like to get information from the two sides of a deal. If we can find the person inside the company who makes the actual sale – we’d be happy to talk to them. And if we can find the person on the buyer’s side who sits across the table from them – that’s our best source of information on durability.

But, there are two cases that prove we sometimes ignore this source of information. One, is Q-Logic. We got information from folks who operate storage area networks. The information was very good as far as proving Q-Logic’s durability. But, there was a problem. The information was good in explaining why companies who are direct customers of Q-Logic and companies that are the end users would want to stick with their existing solution. The information was not so good in explaining the durability of storage area networks themselves. See, the people we were talking to made their living off storage area networks. They obviously believed they were indispensable. We’ve yet to have a source tell us “This thing I spend every day working on is a total buggy whip. It’ll be gone in 10 years and my job along with it.” No one’s ever said that to us. They might think other parts of their company are doing dumb things. They might think their competitors will soon be extinct. They might even think their suppliers will soon be extinct. But, they never think their own job will ever be in jeopardy. So, that’s a problem. And I just wasn’t sure that we were getting good information on the wider durability issues at Q-Logic. However, the information we did get suggested a lot of “stickiness” in terms of people in IT being reluctant to change their behaviors.

So, that’s an example of where the scuttlebutt on durability was all excellent and yet I wasn’t so sure.

Now, let’s take a look at an example where we had zero scuttlebutt supporting the durability of the business – and yet I was completely sold on the idea the company would last.

We’re talking about Babcock & Wilcox. I’m going to simplify here. I’m breaking down the company into 2 parts instead of the 6 or so it really had. And I’m pretending the only power plants it served burned coal – when really some burned other stuff. So, when we analyzed this business it had two key parts. It made boilers and related equipment for coal power plants in the U.S. and elsewhere in the world. And it made nuclear (fusion) components for use onboard U.S. Navy ships.

The U.S. Navy only uses nuclear power on 3 types of ships: 1) Aircraft carriers 2) Ballistic missile submarines 3) Attack submarines. So, you have to be sure of the durability of aircraft carriers and submarines. You also have to be sure they’ll be nuclear powered. There are huge advantages to using nuclear power on ships you want roaming the globe without the need to refuel. So, let’s put that aside as a given. We’re still left with the a military and political question: “Will the U.S. Navy keep wanting aircraft carriers, ballistic missile subs, and attack subs?” And how can I possibly know they will? I don’t know more about global military strategy than the average person reading this blog post. I don’t know more about the politics of the U.S. Navy’s budget. So, how can I be sure these programs are durable?

And this is where we have to admit it’s all speculative. I read what the programs are and what they are used for. I thought they had some of the greatest strategic importance of any defense programs I could think of. And I asked: “Would you cut these programs or some other programs instead?” And my feeling was that if the U.S. Navy had these 3 programs and little else – it’d still have a lot of weight in the world.

I also thought that the Navy doesn’t have much incentive to reduce its budget. It has less than a for profit buyer.

Can we say Babcock & Wilcox is perfectly durable – either in regard to boilers or nuclear power on ships?

No. But, I think we can compare it to all the other stocks we might buy and compared to almost all of our other choices say it’s more durable. The preferences of the U.S. Navy should change less over the next 30 years than the preferences of most customers in most industries.

But there’s an example of pure speculation. I have absolutely no scuttlebutt to go on when it comes to Babcock & Wilcox. I only have the same reports on coal power plants, the U.S. Navy’s plans, etc. that everyone else can read. We did that issue with no information gained through our own interviews.

Assessing durability is ultimately speculative. I was not sure enough about Q-Logic’s durability even though I had customer testimony in support of that product’s durability. And I was sure enough about Babcock’s durability even though I had no customer testimony in support of that product’s durability.

I still think getting testimony from customers and dealers is important. I think the two people on either side of the actual buyer-seller negotiation are who you want to talk to judge durability. But, I also think that assessing durability is maybe 50% testimony and 50% pure speculation.

In some cases, it’s 100% pure speculation. I speculated on Babcock’s durability with no quotes in support of it being durable. I just assumed based on what I – and everyone – knows about the buyer and the projects that they were durable.

Sometimes I’m not willing to make that speculation. Quan recently pointed me to a good short post on Strattec over at Value Investors Club. I don’t know if Strattec is durable or not. But, I don’t think I can speculate on its durability without customer testimony in support of that durability. So, I’m willing to speculate based only on widely available sources that Babcock is durable. But, I’m not willing to speculate that Strattec is durable.

What’s the lesson from that?

I don’t know. I don’t think Warren Buffett does much scuttlebutt anymore. But, I don’t think he’d be able to do as little scuttlebutt now unless he had done a lot decades ago. Still, he sometimes does. For example, he mentioned that before buying IBM stock he talked to the IT departments at some of Berkshire’s subsidiaries to see how sticky their relationship with IBM was. Quan did the same thing with Q-Logic. But, I wasn’t sure of Q-Logic’s durability.

Some of this may just be bias. Nuclear power is very old and really in a way abandoned tech. Most people have given up on it. Babcock gave up on civilian nuclear in the U.S. after Three Mile Island. If the tech hadn’t been abandoned that way – I don’t think I’d be as interested in Babcock. In both nuclear reactors and boilers, what they do is really engineering rather than technology. But, some people might say IBM is as much a client based professional service firm as it is a tech company. I don’t understand IBM well enough to say one way or the other. My fear is that a lot of people are interested in the ecosystems that IBM and Q-Logic and companies like that compete in. No one is really interested in doing what Babcock does unless they’ve been doing it forever. Nuclear and steam aren’t very sexy.

That’s an explanation Quan and I often fall back on. This industry is safe because no one ever seems to enter it and no one ever seems to want to enter it. It’s really just an appeal to history. If the history of the industry has been that competition is limited – then the future of the industry will be that competition is limited.

Is that a valid way of thinking?

Using history instead of just spitballing possible things that could put the company out of business makes sense. Spitballing doomsday scenarios may seem prudent. But, it’s really just an exercise in paranoia. Companies that don’t change a lot in industries that don’t change a lot are probably safer bets than companies that do change a lot in industries that do change a lot. I’m not sure how prescient you can be unless your prescience consists entirely of just saying “The future will look a lot like the past.” I definitely believe in that kind of prescience. Quan and I always try to come up with a reason or two for why the future won’t look like the past. But, that’s really speculative.

So, there are three approaches you can use to judge durability. One, you could gather testimony about customer behavior. Two, you can go by the history of the industry. Three, you can use a purely theoretical – that is, purely speculative – approach by trying to work out the adoption of future technologies and trends and how that can shift the economics of the industry. I did that for you with Village Supermarket. That discussion was mostly just speculation. It was like something out of a microeconomics textbook. I think that approach has an inherent appeal to most people reading this. It feels like it should be right. And it feels like the kind of work you should be doing. I obviously think it has a place – or I wouldn’t have analyzed Village that way.

But, I don’t think that you should give more weight to theory than you do to scuttlebutt and history. Industry history is a record of the economic interactions that really – not just theoretically – happened. And scuttlebutt can provide an insight into customer behavior which is really what product economics is all about. If you talk to customers, they will tell you about their willingness to pay. And they will especially tell you how “sticky” they are and why they stick with their current choice instead of going and searching for an alternative.

So my advice for how to judge durability is: talk to customers, study as much of the industry’s history going as far back as you can, and try to sketch out the economics of entering the market.

The biggest caveat is not to have too much faith in your calculations on industry economics. You can probably determine who has relatively high costs, who uses relatively high amounts of assets, etc. But don’t put too much faith in the quantities involved. The relationships between players are what matters – the numbers are less important.

Talk to Geoff about How to Judge a Business’s Durability

Check Out Singular Diligence




You Can Always Come Up With a Reason For Why the Stock You Are Researching is Actually About to Go Out of Business

by Geoff Gannon

Someone who reads the blog sent me an email asking how Quan and I judge qualitative factors like a company’s durability.

For most stocks, you can easily imagine a future condition that would obsolete the entire business model.

I’ve decided to make this post nothing but a series of examples.


John Wiley

Open access journal articles.

There is a whole Wikipedia page about this one. The idea here is that someone else will pay the cost of publishing journals in place of the subscriber.


Weight Watchers


Dieters will use free apps like MyFitnessPal to count calories instead of going to meetings or using websites like Weight Watchers.



Illegal marketing.

Without aggressive marketing aimed at old people – would this product even exist? You can read about the FCA (a U.K. regulator) fine imposed on HomeServe and the reasons for it here.


Ark Restaurants

Leases expire.

Ark may not renew its leases because the casino or other landlord would want to charge a lot more rent now that the location and the restaurant is a proven success. So, Ark as a corporation has a finite lifespan except insofar as management reallocates capital to new sites.


Village Supermarket

Online groceries.

Traditional supermarkets have 3 durability risks people raise: 1) Online groceries 2) Wal-Mart 3) Organic and fresh competitors: The Fresh Market, Whole Foods, etc.


America’s Car-Mart


America’s Car-Mart sells used cars so it can collect interest on high risk auto loans. The difficult parts of the business are underwriting and collecting loans. If this could be centralized – as it is in lower risk subprime auto loans – then the loans would become commodities.



Online dog food.

The two concerns here are that places like Wal-Mart can sell more dog food and websites like Petflow can sell more dog food.


Atlantic Tele-Network

Guyana can take away their monopoly.



British shoppers will stop frequenting high streets. Or, they will eat healthier food instead.



Self-driving cars will eliminate accidents and therefore the need for auto-insurance.


Babcock & Wilcox

U.S. utilities will shift away from coal power plants – which use boilers – toward natural gas, wind, and solar power plants which don’t use boilers.

The U.S. Navy could stop using: nuclear powered aircraft carriers, nuclear powered ballistic missile submarines, and nuclear powered attack submarines.  



People will wear products like the Apple Watch instead.






Same. Plus, Michael Kors may be a fad.


Western Union

Online competitors like Xoom can replace agent location based money transfers.


Hunter Douglas

Big box retailers like Home Depot and Lowe’s can sell blinds in their stores. Blinds can be sold online. As a result, people will stop going to the independent dealers that Hunter Douglas gets all its sales through.



Smart keys and push to start ignitions can eliminate the need for locks and keys used in car doors and the steering column.



The cloud will eliminate the need for storage area networks.

Talk to Geoff about Durability

Check Out Singular Diligence

Babcock & Wilcox Sets Spin-Off Dates

by Geoff Gannon

Babcock & Wilcox (BWC) has set the dates for its spin-off. Those who own the stock on June 18th will get their spin-off shares on June 30th:

"As a result of the spin-off, Company stockholders can expect to receive as a dividend one share of New B&W common stock for every two shares of the Company’s common stock held as of 5:00 p.m. EST on June 18, 2015, the record date. The distribution of New B&W shares is expected to occur on June 30, 2015 and is expected to be tax-free. "

Shareholders will then own two separately traded stocks. The stock with the “BWXT” ticker will be the government business. The stock with the “BW” ticker will be the power plant business.

The press release gives an accurate description of what “BWXT” will be:

“BWXT is the sole manufacturer of naval nuclear reactors for submarines and aircraft carriers; provides nuclear fuel to the U.S. government; provides technical, management and site services to aid governments in the operation of complex facilities and environmental remediation activities; and supplies precision manufactured components and services for the commercial nuclear power industry.”

It gives a poor description of what “BW” will be:

“New B&W will continue to be a leader in clean energy and environmental technologies for the power and industrial sectors. New B&W also will provide one of the most comprehensive platforms of aftermarket services to a large global installed base of power generation facilities.”

BW is really the boiler business. They build boilers and related equipment for power plants. Some of those plants are clean energy plants – but a great many are actually coal power plants.

Babcock & Wilcox was a Singular Diligence stock pick. I own the stock personally. Quan does not. I plan to keep both my “BWXT” shares and “BW” shares indefinitely.

I’ll let you know if that changes.

Read the Babcock & Wilcox (BWC) Press Release

Check out Singular Diligence

Talk to Geoff about Babcock & Wilcox (BWC)