Geoff Gannon July 27, 2017

How to Read Between the Lines of a 10-K


“I have read some of your Singular Diligence content and one of the most interesting parts of the reading are the notes that you and Quan used and published in those reports.

So one of the most remarkable things that I have noticed in the notes is that you did not use a lot of information in the annual reports. And this sounds intriguing to me since I am a regular reader of your blog and you have always emphasized how important it is to read (the) 10-K, annual reports, (and) quarterly reports. Having said that, it would be good to understand why did you not use more info from those sources. Moreover, I was actually curious how do you allocate your time when doing research? I am currently also doing research so I was looking to see which sources you prioritize?”



It depends on what you mean by prioritize? If a top priority is the thing you read first, then I make 10-Ks a top priority. If a top priority is the thing you spend the most time on, then SEC filings are a low priority. And then, if a top priority is something you quote a lot – as you mentioned – SEC filing are a very low priority for me.

I’ve said before that I always read the newest 10-K and the oldest 10-K of a stock I’m researching. So, imagine I am researching a U.S. stock that has been public for a long time. The SEC’s database of company filings – which includes 10-Ks (annual reports), 10-Qs (quarterly reports), and “going public” or spin-off documents if the company has any of those. If the company has been public in its current form for a long time, the oldest annual report (10-K) will be something from maybe 1994-1996. That’s around the time companies started being required to file electronic copies of their 10-Ks with the SEC. As a rule, a U.S. company that has long been public will now have about 20 years of annual reports for you to read in full.

Don’t do that. I just read the 2016 (last year’s) annual report and then the 1994 or 1995 or 1996 – or whatever the oldest report you can find is – annual report. I want to get a sense of how the business has changed.

The 10-K – when read on its own – is of limited value when making an investment. All of the financial data that Quan and I used to create the “datasheet” of a Singular Diligence comes from the 10-K. So, the financial statements – especially the income statement, the balance sheet, and the cash flow statement – are useful. The datasheet we presented was usually a 15-25 year history of a company’s finances. It’s by far the most important thing in those reports. It may be possible to make an investment based purely on the financial statements if you have both income statement and balance sheet data for at least 15 years and preferably closer to 30 years. I’m not going to say that qualitative judgments are entirely unimportant – but long-term historical financial data is by far the most important source of information an investor will ever see. A great investor – like Warren Buffett – will often be able to make the decision to invest in a business or not invest in a business based on just 15-30 minutes spent with 15-30 years of past data. Most people don’t believe this. But, it’s true. If you know a lot about how businesses normally look, what ratios are normal, what years were good or bad years in the economy, etc. and you have 15-30 years of really detailed history in front of you – you have 90% or more of the information you need to make an investment. The difference between ordinary and extraordinary businesses jumps out at you in the long-term financial results. Honestly, it doesn’t take even 60 seconds to discard at least 50% of all investment opportunities based purely on their 15-30 year past financials.

But, I will assume you are asking about the written part of the 10-K – the text the company includes. This text is of limited value for a number of reasons.

One, like accounting data, the business summary and executive summary of the past year’s operations and so on in the 10-K are created in a way to increase comparability. Comparability is a double edged sword. For example, under U.S. GAAP (Generally Accepted Accounting Rules) there is a high degree of comparability in the way earnings per share are presented for all companies. We are given an EPS figure for every public company. And those EPS figures are calculated in a similar way for all of them. Certain rules are followed.

How helpful are those EPS figures in making decisions about what a business is worth?

Sometimes, not very. For example, railroads and ad agencies both report EPS. As I’ve said before, a railroad trading at 20 times earnings is much, much more expensive than an ad agency trading at 20 times earnings. The management of both the railroad and the ad agency knows this. Analysts and investors who specialize in those two industries know it too. But, there isn’t a lot of care given in the 10-Ks of either railroads or ad agencies to explain the actual underlying cash economics of their respective businesses and to explain why EPS is not a good guide in either case. I cherry-picked two unfair examples. But, there are many more industries where reported earnings in any one year can be very misleading. Insurance and bank results for any one year aren’t really helpful. Management has wide discretion in reserving for losses in both cases. Both insurers and banks are sensitive to the level of interest rates. They are also sensitive to the “animal spirits” – basically the carefulness or recklessness – of competitors. In interviews that management does with the media (or with you if you try to approach them and they answer your email, take your phone call, etc.) they will be more open about the actual numbers that matter. So, we tend to get more of our notes from articles (that quote management, analysts, etc.), interviews with branch managers, customers, etc., earnings call transcripts, investor presentations, etc. We also get a lot of notes from competitors rather than from the company itself. In fact, it’s possible we tend to cite non-10-K sources from competitors more than we cite the company’s own 10-K.

This has to due with the usefulness of the info we can get – especially the candor. Candid information is the most valuable information. 10-Ks aren’t very candid. You need to have read a lot of 10-Ks to be able to read between the lines.

In the U.S., the text portions of 10-Ks tend to be very, very conservative compared to what you will find in the presentation of official annual results in other countries. There a couple reasons for this. One, American business culture is less promotional than what you will find in some emerging markets. It’s not less promotional than what you will find in Japan, the E.U., or the U.K. though – and yet, the annual reports of American companies are often far more conservative in their dire phrasing of business risks, etc. than the annual reports of those companies.

The reason is probably the influence of legal advisors on how a 10-K is written. Lawsuits are common in the U.S. Bad press is common in the U.S. And very few American investors actually read the 10-K. So, putting bad stuff in the 10-K is a good way to avoid lawsuits and muckraking reportage later if things do go bad for your company. You can always point to the 10-K and say those risks were disclosed – often in the most dire yet broad and ambiguous ways – when you are accused of tricking investors into believing your stock was a better investment than it turned out to be.

For this reason, you have to get very good at reading between the lines of a 10-K. For example, there is a section of a 10-K that tells you about competition. The normal, boilerplate statement here basically consists of three parts:

  1. The industry is “highly competitive” or “competition is intense”
  2. The company has competitors who “are larger”, have “greater financial resources”, than the company
  3. The company competes on the basis of “price”

Finding any of those phrases in a U.S. 10-K means absolutely nothing. All U.S. companies tend to say in their 10-K that their industry is highly competitive, some competitors are larger, have established brand names (if that matters in the industry), and have greater financial resources than the company. And they all include a series of factors on which they compete of which “price” is the most common.

So, a lot of beginning investors reading a 10-K with those warnings will take them seriously. They might even highlight those terms, write about them when they blog about the stock (if they’re a blogger), etc. They believe this is part of good due diligence.

It’s really not. That’s just boilerplate. It is only the omission of those phrases that is worth paying attention to.

Anything in a 10-K that is ordinary needs to be ignored. You only need to highlight, take notes, remember, etc. those parts of the 10-K that are extraordinary. All the normal stuff you can forget the second you read it. All the abnormal stuff you need to remember forever.

For example, if a company does not say it competes on the basis of price – that’s a statement of extraordinary importance. It may be the most important thing you “read” in that 10-K. And yet, almost no U.S. company will ever say “we believe price is not a primary basis on which we compete”. Yet, some U.S. companies will omit “price” from the list of factors on which they compete.

I’m going to test your 10-K reading ability with this excerpt from the most recent Landauer (LDR) 10-K:

“In the U.S., the Company competes against a number of dosimetry service providers. One of these providers is a division of Mirion Technologies, Inc., a significant competitor with substantial resources. Other competitors in the U.S. that provide dosimetry services tend to be smaller companies, some of which operate on a regional basis. Most government agencies in the U.S., such as the Department of Energy and Department of Defense, have their own in-house radiation measurement services, as do many large private nuclear power plants. Outside of the U.S., radiation measurement activities are conducted by a combination of private entities and government agencies.


The Company competes on the basis of advanced technologies, competent execution of these technologies, the quality, reliability and price of its services, and its prompt and responsive performance.”


Okay. Go back and read it again if you have to. I’m about to highlight the parts of that paragraph and a half that matter – and matter in such a big way that you’d immediately latch on to Landauer as a possible wide-moat company. Here are my highlights:


“In the U.S., the Company competes against a number of dosimetry service providers. One of these providers is a division of Mirion Technologies, Inc., a significant competitor with substantial resources. Other competitors in the U.S. that provide dosimetry services tend to be smaller companies, some of which operate on a regional basis. Most government agencies in the U.S., such as the Department of Energy and Department of Defense, have their own in-house radiation measurement services, as do many large private nuclear power plants. Outside of the U.S., radiation measurement activities are conducted by a combination of private entities and government agencies.


The Company competes on the basis of advanced technologies, competent execution of these technologies, the quality, reliability and price of its services, and its prompt and responsive performance.”


So, what was hidden between the lines in Landauer’s discussion of competition?


  1. “against a number” – This is an unusual way of phrasing the number of competitors. It is normal for the 10-K to say something like “a large number”, “many”, “highly fragmented”, etc. when describing the number of competitors. Saying “a number” suggests the company may not have felt it was reasonable to use such terms because they would be an excessive exaggeration. This phrase suggests there may be only a small number of national competitors to Landauer.
  2. “One of these providers” – This confirms the hint in phrase #1. It is rare for a U.S. company to cite a competitor by name. When a company you are reading about names a single competitor outright, you can be confident the industry has only a small number of major players. When the company you are reading about provides the names of maybe 3-5 competitors, we may be talking about more of an oligopoly type situation. Companies in truly competitive – and especially nearly perfectly competitive industries – will not give you the name of specific competitors.
  3. “tend to be smaller companies, some of which operate on a regional basis” – This is their way of telling you – without being promotional – that Landauer is larger and more national in scope than almost all companies in the industry.
  4. “own in-house” – You might have missed this one. It’s actually one of the most important lines in this paragraph and a half. Companies that face little actual competition tend to talk about in-house competition. For example, an ad agency, IT services provider, etc. that doesn’t really have a lot of competition may say that they “compete” with their own customers in the sense that their customers could choose to provide these services for themselves without hiring out the work to an external firm. Think about this. Most companies face this problem. Yet, they don’t mention it. All restaurants compete with cooking at home. All car companies compete with people taking buses and trains. They don’t make a big deal of it. Why not? Because they have plenty of competition among themselves (in the industry). When a company starts insisting they compete with “in-housing”, you know you’re dealing with an industry with limited competition among the players.
  5. “and price” – It’s unlikely Landauer competes primarily or maybe even secondarily on the basis of price. The way they say “and price” here is really buried. The actual phrase is “reliability and price of its service” which is presented between two commas and yet it’s not the final item in the list – “prompt and responsive performance” is the last item in the list. In English, it’s hard to de-emphasize something any more than connecting it with an “and” within a list where some items are stated on their own (see “quality”), and where the item is presented neither as the first nor the last item in the list. No one is going to remember the word “price” was even in that sentence. Important items are listed first or last in a series, are not paired with other items between two commas (“milk and cookies” de-emphasizes both milk and cookies as standalone snacks if you list them together like that).


So, there’s a ton of information I’d get out of that paragraph and a half of the 10-K. But, then I’d just use them as “leads” and go investigating what the company meant by what it said and try to confirm that with customers, sales people, etc. I could talk to. I’d look for what confirms this or challenges it in any earnings call transcripts, etc.


The most important textual parts of the 10-K are the business description and the competition sections of the 10-K. However, the company is unlikely to include much of the most important competitive information in the 10-K. In the 10-K, companies rarely provide detailed estimates of key facts like:


  1. Market share
  2. Relative market share
  3. Customer retention rate


They also often include competitive strengths that rarely matter in most industries – like patents and “people” – while excluding discussion of factors like location that are much more likely to be important.


For example, I’ve researched companies in industries where imports into the U.S. are not a realistic way of competing because of the low value/weight ratio of the product and the desire of customers to have low inventories of the product on site while having jobs done when scheduled. Under those conditions – where prompt delivery of a product with low value to weight is what all customers want – there is no risk of foreign competition. And yet, in the 10-K, the only disclosure of this fact was something like “because imports are usually less than 1% of U.S. consumption and exports are small, industry data is traditionally compiled using domestic production data alone”. I made up that exact quote. But, it’s very close to statements I have seen. Those kinds of things are very, very important to notice.


But, even in those cases, the most useful data wasn’t in the 10-K. Instead, a mention like that of imports being immaterial lead me to the industry report compiled by someone like the U.S. government or a trade association which gave long-term data showing that throughout all of history, neither imports nor exports were the least bit meaningful. Knowing that history, you know that competition has to be very localized.


This is usually how the research works. The 10-K is important in providing an original lead – usually from reading between the lines, nothing explicitly stated – that gives me questions I then write down to use in investigating the earnings call transcripts, discussions with customers or ex-employees or whoever, industry reports, media reports, etc.


If you look at the Singular Diligence reports, I think you’ll notice I am biased toward 3 things being really important in making an investment:


  1. What is the company’s market power – is it going to get stronger or weaker?
  2. How does management allocate capital – is it going to get better or worse?
  3. What is the stock’s price – will investors value it more highly or less highly in the future?


If you can answer those three questions by saying:


  1. Market power is strong and getting stronger
  2. Capital allocation is smart and getting smarter
  3. Stock price is cheap but won’t stay cheap once the stock’s back in favor


You have the perfect stock. It’s hard to do much research into “market power” which is my number one criterion using the 10-K. Even things like capital allocation sometimes require a little context outside of the 10-K.


The 10-K is most important for the financial data it provides. It is also important for its discussion of how the business works and how competition works. All other parts of the 10-K are of limited use. I read the 10-K as a first step. Sometimes, I have found things in a 10-K that made me decide to eliminate the stock from consideration. I’ve never found a 10-K that got me 100% of the way to deciding to invest.


Long-term (15-30 year) historical financials are much more important than the 10-K. But what you can get out of reading those financials or reading the 10-K depends a lot on how much prior investing experience you have. If you have been looking at long-term financial histories (Value Line / GuruFocus type presentations) and 10-Ks for a long time, you can often get more out of them in 60 seconds than the average investor can get out of reading them for 60 minutes.


If you and Warren Buffett both read a 10-K, it can’t really be considered the same 10-K. There is a lot more he can get out of it than you can get out of it, because he brings the context of mastery of finance, business, and economics that comes from decades of experience.