Today, I’ll tackle a related subject. What to do when you have two new ideas? Which should you follow up on first? How do you decide which lead to pursue?
For me, the first question is often: Is this stock conspicuously cheap?
Some stocks may actually be cheap if you understand them perfectly – you grasp every nuance of their operations, their competitive position, their product economics, etc. The problem is that you probably need to spend dozens of hours analyzing such a company to realize it is cheap.
How can you avoid spending dozens of hours on a dead end?
Try to find a company that – if it turns out to be pretty typical – would actually still be cheap.
I call that conspicuous cheapness. It’s cheapness that jumps right out at you from the start. And it’s a great way to decide what stock is a promising lead.
To play fairly here, I’ll use examples of stocks I don’t know very well. These are stocks I – like you – haven’t studied yet. The question is which would I start with?
A great place to get a quick snapshot of a stock – really have the idea “pitched” to you – is blog posts. If you aren’t reading value investing blogs, you should be. They are the single best source of investment ideas. Not necessarily the best source of investing advice. Many of their ideas are bad. But there are plenty of ideas in these blogs, so if you can learn to pick and choose the “leads” that work for you – reading value investing blogs will quickly become one of the best possible uses of your time.
Obviously, I have my favorite blogs. And I tend to read these first. Two of my favorite blogs – for new stock ideas – are:
1. Value and Opportunity
2. OTC Adventures
Each recently wrote up a stock that interested me. Again, think of these write-ups not as analysis but as a “pitch” for a stock. They are throwing an idea my way. Maybe I like it, maybe I don’t. The question is whether it catches my attention. Is there something here I’d be interested in.
Value and Opportunity’s pitch was for a French company called Thermador. OTC Adventures’ pitch was for a Canadian company called Logistec. Now, I am already being selective in saying these are the two companies that were pitched. In reality, Value and Opportunity has thrown out a few more ideas – in blog posts – since that Thermador post. Why aren’t I mentioning them here? Because Thermador caught my attention more than they did.
Please read these blog posts. They’ll give you a great idea of what I mean when I say a stock pitch.
In this case, both stocks score okay on conspicuous cheapness. One trades at about 5.5 times EBITDA. The other trades at 6.2 times EBITDA. That is right around a Ben Graham style two-thirds of a normal EBITDA. We’d need to know a bit more about free cash flow generation relative to EBITDA and things like that. Although the Thermador post actually digs into that enough to reassure us.
I think there is enough in each blog post to show you what I mean when I say a “promising lead.” Those are two promising leads. A promising lead is a pitch that catches my attention.
Let’s talk about what I mean when I say a pitch caught my attention. What catches my attention?
Conspicuous cheapness is the number one attention grabber. If you talk to me about a bank that trades at 1.7 times book value – it may be a wonderful bank or a terrible bank – but it isn’t a conspicuously cheap bank. A bank trading at 0.4 times book value is. There’s no doubt the bank trading at 1.7 times book may be the better investment. If we’re talking about a bank that could probably earn a nice return on equity – that has intangible competitive advantages – like Wells Fargo (WFC) or Valley National (VLY) or Bank of Hawaii (BOH) – I might be interested. But it would take some convincing. Why?
Because you need to convince me that the current price – again, let’s say it is 1.7 times book value – is really a discount to intrinsic value. You would, in that case, probably need to prove the bank is really worth something like 2.5 times book value. That would give me a roughly one-third discount to what it should trade at.
Let’s revisit my John Wiley (JW.A) example. I think John Wiley has a strong franchise. If you disagree with me, there is no need to discuss the stock further – because it’s not conspicuously cheap on an EV/EBITDA basis. But let’s say you do agree with me. Let’s say I can convince you that John Wiley should really trade at 10 to 12 times EBITDA like some food and beverage companies, media companies, etc., with good prospects often do.
Okay, if you believe John Wiley has that kind of moat and those kind of product economics – then we can talk price. We can apply the same idea of a margin of safety – like a one-third discount to “intrinsic value” whatever that is – and come up with an acceptable price range for the stock of 6.5 to 7.8 times EBITDA (that’s two-thirds of 10-12 times EBITDA). If the company is “fairly valued” at 10-12 times EBITDA, it becomes cheap at two-thirds of that level (in this case, 6.5 to 7.8 times EBITDA).
For most companies, 6.5 to 7.8 times EBITDA is unremarkable. I’d say it might be cheap. But it’s not conspicuously cheap. Those levels do not jump out at me. You’d have to get to at least around 6.5 times EBITDA and lower – probably closer to 5 to 6 times EBITDA – for me to feel I was looking at an unusually cheap stock.
Remember, when I bought George Risk (RSKIA) it had a slightly negative enterprise value. It had positive EBITDA. So, that’s like saying the EV/EBITDA was less than zero. That’s extraordinarily rare. A more recent – and much more expensive – example would be my purchase of Ark Restaurants (ARKR). That purchase was made at 5 times EBITDA. This is about a 35% discount to a “normal” EV/EBITDA of 8 (totally an arbitrary number – but it tends to roughly work out to no more than a P/E of 15, which many stock investors call normal). It was also a discount of maybe 30% from many recent restaurant company acquisitions (which tended to be done at about 7 times EBITDA) and it was a discount to a recent takeover offer of $22 a share. The discount on this last level was small. You can calculate it yourself. For example, at $20 a share, the stock would be trading at a 10% discount to that past (rejected) takeover offer.
These three discounts don’t prove the stock is attractive. But they provide a reference point. I can start asking – do most companies trade around 8 times EBITDA? And then is Ark better or worse than most public companies?
Were most public restaurant companies bought for 7 times EBITDA? And is Ark better or worse than those companies? (I can also go through the list and compare Ark to each acquisition – since we have records of what price was paid for what company.)
Finally, we can think about the offer from Landry’s and consider whether we thought it was a serious offer, whether they might have raised it, whether there was good reason to believe no deal ever could’ve gone through at that level, what management thought, how they reacted, etc.
This last one is complicated. The controlling shareholder (who is also management) rejected the offer quickly. This could be a sign management believes the stock is worth more (and may or may not be right about that), it could be a sign management has no intention of selling the company at any price, etc.
Whether the buyout offer and its rejection is a good or bad sign for the company as an investment today – a little below that offer price – is a complex issue. But, it is a potential sign of conspicuous cheapness.
I can think of five such conspicuous signs of cheapness:
· Absolute value (example: EV/EBITDA vs. say 8x).
· Relative value (example: EV/EBITDA vs. peers).
· Past transaction value (example: EV/EBITDA vs. 7x EBITDA in this case).
· Historical value (example: EV/EBITDA today vs. EV/EBITDA the stock traded at in each of the last 10 to 15 years, etc.).
· Control value (example: $22 a share offer from Landry’s in this case).
These are good first- glance numbers. They don’t have to be EV/EBITDA. In some cases – like financial stocks – they shouldn’t be. Maybe they should be price to book.
It’s important to look through a pitch for a stock – like the blog posts for Thermador and Logistec I linked to – and think about where you see signs of conspicuous cheapness. These can include price-to-book, P/E, EV/EBITDA, dividend yield, etc.
If I had to use one number on its own I would use EV/EBITDA. But you don’t have to use one number alone.
An example of a stock I thought was conspicuously cheap was Bancinsurance (no longer public). The company traded for less than 70% of book value. The CEO (who controlled the illiquid company) made an offer to buy the company out at about 70% of book value. Around that time, you could buy shares at prices equal to or less than both the buyout offer and 70% of book value.
The questions then became whether you thought these discounts were discounts to good dividing lines of cheapness. Did I feel convinced that Bancinsurance was worth more than the buyout offer and worth more than 70% of book value?
I did. Why?
Not because it was an insurance company. Some of those may – rightly – sell for less than 70% of book value. The reason I thought it was cheap had to do with the combined ratio and returns on equity. Historically, the combined ratio had been under 100 (in other words, it had underwriting profits in almost all years). And the ROE had been 10%.
The stock market tends not to return more than 10% a year. So, a stock that returns 10% on equity year in and year out should be worth about book value. Likewise, a company that has assets – like an investment portfolio – plus a stream of underwriting profits shouldn’t be worth less than the value of that investment portfolio.
To argue otherwise, you would need to show that the separate earnings stream (the underwriting profits) was less valuable than whatever “closed end fund” type discount you were applying to the company’s investment portfolio. Whether you should apply any discount – and how large it should be – depends on issues like whether the portfolio is mismanaged or not and whether capital will be paid out to you. In this case, the company had recently needed capital to rebuild a devastated balance sheet (caused by a one-time, very large relative to the company’s size event). They then paid a large one-time dividend. Here, I mean large relative to the stock price. This was just an anecdote. But it was an anecdote suggesting the company had rebuilt its capital and might be ready to pay out capital if it kept earning at the current rate. Less anecdotal evidence was provided by improving ratings at A.M. Best and so on.
My point is that it was a conspicuously cheap stock. You started with the simple fact it sold for less than book value. You moved from there to ask whether it should sell for less than book value. The last 30 years of underwriting said “no, it shouldn’t.” The last 15 years of ROE said “no, it shouldn’t.” And then you had an offer from the CEO to buy the whole company. You also had the historical record of the company’s stock price which – before being delisted – tended to trade above book value about as often as it traded below book value.
Taking these four factors together, it didn’t seem like it was easy to justify a price below book value. And yet, the price was clearly more than 30% below book value.
So, the price-to-book was the “lead.” And I pursued that lead by gathering evidence that could point to a discount or premium to book value:
· Does underwriting add or subtract value from the investment portfolio?
· Is the ROE greater than or less than the historical return of the stock market?
· Has the stock tended to trade above or below book value in the past?
· Is a buyer offering to pay more or less than the stock is trading for?
By answering these questions, you can quickly establish whether or not the company seems cheap. It could have fatal flaws that ruin it as an investment. But, after running this check, it clearly seems cheap.
Let’s compare this to a stock I once owned – I bought it around $28 a share and sold it around $45 a share – called Omnicom (OMC). Book value is meaningless here. Price-to-sales and price-to-free cash flow matter most. Price-to-sales is probably most important because of the cyclicality of margins in the advertising business. If you go by price to free cash flow alone, you may tend to consider selling the stock in a bad ad market and consider buying it in a good ad market. We want to avoid that. So use either price-to-sales or something like the Shiller P/E.
The stock – which now trades at $65 a share – is not conspicuously cheap. Omnicom wouldn’t interest me today. That doesn’t mean I think it is overpriced. I don’t. In fact, when I analyzed the stock – and bought it around $28 a share – I figured it would be worth about $65 a share in “normal” times. Normal times came quicker than I expected – I was analyzing the stock in early 2009. So, I might have been thinking the stock would reach $65 a share sometime in 2015 or something like that. I didn’t know if a crash as big as we saw in the economy would lead to a profit recovery in as soon as three years or so.
It did. The earning power of the company isn’t all that different. It bought back stock and raised its dividend. And the ad market improved. Interest rates also fell, pushing up all stock prices. These factors explain the stock price increase of about 130% over the last four years.
I would be willing to analyze Omnicom at a fairly reasonable price. Because I owned it and that’s the best way to get to know a stock. If it traded at about 1 times sales, I’d analyze it. At that level – and below – it would be conspicuously cheap.
Above one times sales, it is not conspicuously cheap. It is, rather, what you might call “contingently” cheap. The company’s cheapness is contingent on a series of beliefs – things you’d have to study – about the competitive position, capital allocation, and the normalness of the current ad market. If you thought the ad market was perfectly normal, the competitive position was as good as 5 years ago, and you knew exactly what the company would do – and approved – in terms of dividends and share buybacks, Omnicom would be a good lead to follow.
But you’d need to know those things first. Which makes it a lot of work. I don’t like to do a lot of work. Instead I like to focus on a level where a company could be doing a few things wrong and the stock could still work out right for me.
For Omnicom, that’s 1 times sales. Above 1 times sales, you are paying up for quality. Below one times sales, you aren’t paying up for anything. So, if you can prove there is quality and you can buy at 90% of sales – then you have not just a cheap stock but a conspicuously cheap stock. Which makes it a promising lead to follow.
I should point out, that by many value investing measures, Omnicom wouldn’t be cheap at 0.9 times sales. For example, the price-to-book would not be low. To me, that’s irrelevant with a company like Omnicom. If you don’t believe its intangibles have tremendous value – it’s clearly a bad deal. The only way for it to be a good investment is if those intangibles have value.
Now let’s return to the two stocks I mentioned: Thermador and Logistec.
How would I decide which one to follow up on first? How would you?
The first question is conspicuous cheapness. Does some measure – or hopefully many measures – jump out at you? Does it look cheap relative to a typical company?
Good. That’s step one. They both pass. Step two is familiarity. In other words, understanding in the sense of how much knowledge from elsewhere can I transfer to this topic? Basically, do I know things about these companies? Do I know things about companies like these? How can I learn? What can apply that I already know from other countries, industries, companies, etc., to this situation?
This is a huge issue. It’s one of the biggest reasons I throw a new stock idea out. It either:
A. Isn’t conspicuously cheap.
B. Doesn’t seem immediately understandable.
Very quickly in your investment analysis of any company, you need to establish that there is some measure you can use to prove it’s cheap and then you need to establish there’s a way to understand the company.
Any company that’s too hard to value or too hard to understand is not a good stock idea. It’s not a promising lead.
Value doesn’t mean you need to appraise the company. All you need to prove is that you are paying less than it’s worth.
I couldn’t value Bancinsurance precisely. Nor could I value George Risk. I felt I could prove Bancinsurance was worth more than 70% of book value and George Risk was worth more than net cash.
Once I realized that, I knew they were promising leads.
The next step is grappling with understanding the stock. It may be cheap? But what if it only seems cheap because I don’t understand it?
That’s tomorrow’s topic.
But we have one last task for today. Logistec or Thermador?
Which lead would I follow first?
Which would you?
They’re both conspicuously cheap. They aren’t so cheap that they are worth buying if they turn out to be lousy businesses. But they are trading at below average prices. So as long as we can prove they are above average businesses, they’re good candidates.
Both companies are conspicuously cheap. So cheapness won’t decided which I follow up on first.
Knowledge will. Familiarity will.
I know more about companies like Thermador. My knowledge would transfer easier. So I’d investigate Thermador first. I know very little about Canadian trade and how normal it is right now compared to the past and likely future. That’s a hurdle. It’s a hurdle I can clear. Both companies are understandable.
But Thermador is the company where the knowledge I already have will transfer easier and faster. It will get the ball of analysis rolling faster. It’s an easier company to dig into first. So, I’d start there.
That’s a personal choice. If you know more about companies like Logistec – or if you live in Canada – you’d obviously start there.
I’d look at them both. In fact, I will look at them both. But I’ll look at Thermador first?
How about you?
Ask Geoff a Question