Someone who reads my articles sent me this email:
Hi Geoff,
First of all, thanks for blogging your thoughts on investing. I really enjoy the blog and while driving over the holidays, I listened to most of the podcasts. They were great.
In this post, you described your research process. I was just looking for a little clarification. Is the only 10-K that you read, cover to cover, the most recent? Aside from that are you focusing on the MD&A (Management Discussion and Analysis)?
Thanks,
Reid
Yes. I'd say the only 10-K I read from cover to cover is the most recent 10-K. I always print out and mark up a hard copy of the latest 10-K, 10-Q, and 14A. How much I read of the other SEC reports depends on the specific company and its situation. Normally, I will check all of the 8-Ks and 13Gs from the last year or so. If I know anything specific about the company's past – like there was a scandal, a proxy battle, legal case, unconsummated merger, etc. – I will read the 8-Ks and other documents surrounding that time period even if it was many years ago.
Also, I usually read the oldest 10-K as well as the newest. For example, if a company last filed its 10-K for the 2011 year, but it has been filing with EDGAR since 1996, I will read the 2011 10-K and the 1996 10-K. But unless I have a reason to I will not read all the 10-Ks in between. I will however read any shareholder letters ever written that are still available. So, if the 10-Ks included part of an annual report sent to shareholders like some do, I would read that part of the 10-K for all 15 or 16 years or however long the company has been sending them to EDGAR.
A few companies keep old annual reports, shareholder letters, etc. on their company website. I always check both EGDAR and the company website. In probably 19 out of 20 cases, there is nothing more in terms of corporate communication on the website than you can find on EDGAR. So I end up only using EDGAR.
Of course, companies that sell things have very useful information on their websites about their products, prices, sales methods, etc. which are often not reported to the SEC. Most companies are vague about these things. Although I've seen 10-Ks (especially for smaller companies) where they honestly list the exact names of competitors, the exact names of products, and the actual sales prices. This is rare for larger companies. But in all cases internet searches will provide that kind of information. So you end up doing some Googling along with reading EDGAR.
Otherwise, I do not have a one size fits all approach to how much of the past filings I read. I read the financial statements themselves for all years. And I enter balance sheet, cash flow, and income statement data into Excel sheets I have for every year the company reported to EDGAR. If the company has long been public this will be something like 14-17 years of data (it took a few years for EDGAR to be phased in. Different companies started at slightly different dates). I always have at least 10-year data. And this is the data I use for direct comparisons between companies. But in addition to this I fill out summary balance sheets, cash flow statements, and income statements for every year the company has reported using EDGAR.
Usually, you will end up with more than 10 years but less than 20 years of historical data. Like I said, some companies go further back. Obviously, they will mention data from before this period – especially if you read the oldest 10-K – but complete financial information is rarely available before the 1995-1997 period.
No. I wouldn't say I'm focused on the Management Discussion and Analysis (MD&A). Sometimes it is useful. But it depends very heavily on how honest and complete the company chooses to be. Many times, the MD&A is formulaic. It is very cut and paste. There is no logic to this. Probably the best disclosures I have seen in that area come from very tiny companies. At times it seems small companies don't bother to truly separate SEC reporting from the rest of their business. And so you get some descriptions thrown in there that were probably created for other purposes. For example, I've seen descriptions of new products and things like that which were almost certainly prepared originally for internal use. This is most common when the largest shareholder of the company is also the chief executive. Of course, there are also situations with small companies where almost no disclosures are made. Large companies tend to be the most formulaic in their reporting. Often none of the prose feels like it was written by a human being.
The things I look at most are actually the notes to the financial statements. Together, the general business description, the historical financial data (I take from the financial statements and enter in Excel) and the notes to the financial statements provide the vast majority of my understanding of a public company.
This information will sometimes be supplemented by data sources like competitors reports, trade publications, information presented to customers, interviews, news reports, and things like that.
The two most important bits of information that are usually not provided in the 10-K (or any SEC report) are the management perspective and the customer perspective. What is customer behavior in the industry? Why do customers choose what they choose? And then what does management really think? What are they really like?
Interviews, old news reports, etc. can help with the management perspective. So can (the more candid) shareholder letters. Customer behavior is usually best found in data rather than descriptions. Both customers and companies are badly biased in describing why they think customers behave as they do. Usually, data gathered by the industry, government, etc. is useful. Especially useful is the frequency and size of orders, who makes the purchasing decision, etc.
Sometimes studying the history of the industry and other companies in similar situations helps. For smaller companies, it is often possible to learn about the same industry through many different regional examples.
But, overall, I'd say learning about the customer is very, very important. And it is very hard to do just from the SEC reports. Most are badly lacking in this respect. And my lack of comfort with my understanding of a company's customers is often a reason I pass on buying a stock. I tend to prefer investing in companies where I feel I understand customer behavior best.
How you get this information varies. And different investors will choose to get it differently. We are verging on the Phil Fisher scuttlebutt realm of stock research here. It’s important. But it’s often harder to lay out the process as rigidly as when you are doing Ben Graham type research.
One thing I recommend to all investors is reading the notes to the financial statements. These are very important. They tell you a lot about the company.
And some investors don’t read the notes at all.
Talk to Geoff About Reading 10-Ks [email protected]
Hi Geoff,
First of all, thanks for blogging your thoughts on investing. I really enjoy the blog and while driving over the holidays, I listened to most of the podcasts. They were great.
In this post, you described your research process. I was just looking for a little clarification. Is the only 10-K that you read, cover to cover, the most recent? Aside from that are you focusing on the MD&A (Management Discussion and Analysis)?
Thanks,
Reid
Yes. I'd say the only 10-K I read from cover to cover is the most recent 10-K. I always print out and mark up a hard copy of the latest 10-K, 10-Q, and 14A. How much I read of the other SEC reports depends on the specific company and its situation. Normally, I will check all of the 8-Ks and 13Gs from the last year or so. If I know anything specific about the company's past – like there was a scandal, a proxy battle, legal case, unconsummated merger, etc. – I will read the 8-Ks and other documents surrounding that time period even if it was many years ago.
Also, I usually read the oldest 10-K as well as the newest. For example, if a company last filed its 10-K for the 2011 year, but it has been filing with EDGAR since 1996, I will read the 2011 10-K and the 1996 10-K. But unless I have a reason to I will not read all the 10-Ks in between. I will however read any shareholder letters ever written that are still available. So, if the 10-Ks included part of an annual report sent to shareholders like some do, I would read that part of the 10-K for all 15 or 16 years or however long the company has been sending them to EDGAR.
A few companies keep old annual reports, shareholder letters, etc. on their company website. I always check both EGDAR and the company website. In probably 19 out of 20 cases, there is nothing more in terms of corporate communication on the website than you can find on EDGAR. So I end up only using EDGAR.
Of course, companies that sell things have very useful information on their websites about their products, prices, sales methods, etc. which are often not reported to the SEC. Most companies are vague about these things. Although I've seen 10-Ks (especially for smaller companies) where they honestly list the exact names of competitors, the exact names of products, and the actual sales prices. This is rare for larger companies. But in all cases internet searches will provide that kind of information. So you end up doing some Googling along with reading EDGAR.
Otherwise, I do not have a one size fits all approach to how much of the past filings I read. I read the financial statements themselves for all years. And I enter balance sheet, cash flow, and income statement data into Excel sheets I have for every year the company reported to EDGAR. If the company has long been public this will be something like 14-17 years of data (it took a few years for EDGAR to be phased in. Different companies started at slightly different dates). I always have at least 10-year data. And this is the data I use for direct comparisons between companies. But in addition to this I fill out summary balance sheets, cash flow statements, and income statements for every year the company has reported using EDGAR.
Usually, you will end up with more than 10 years but less than 20 years of historical data. Like I said, some companies go further back. Obviously, they will mention data from before this period – especially if you read the oldest 10-K – but complete financial information is rarely available before the 1995-1997 period.
No. I wouldn't say I'm focused on the Management Discussion and Analysis (MD&A). Sometimes it is useful. But it depends very heavily on how honest and complete the company chooses to be. Many times, the MD&A is formulaic. It is very cut and paste. There is no logic to this. Probably the best disclosures I have seen in that area come from very tiny companies. At times it seems small companies don't bother to truly separate SEC reporting from the rest of their business. And so you get some descriptions thrown in there that were probably created for other purposes. For example, I've seen descriptions of new products and things like that which were almost certainly prepared originally for internal use. This is most common when the largest shareholder of the company is also the chief executive. Of course, there are also situations with small companies where almost no disclosures are made. Large companies tend to be the most formulaic in their reporting. Often none of the prose feels like it was written by a human being.
The things I look at most are actually the notes to the financial statements. Together, the general business description, the historical financial data (I take from the financial statements and enter in Excel) and the notes to the financial statements provide the vast majority of my understanding of a public company.
This information will sometimes be supplemented by data sources like competitors reports, trade publications, information presented to customers, interviews, news reports, and things like that.
The two most important bits of information that are usually not provided in the 10-K (or any SEC report) are the management perspective and the customer perspective. What is customer behavior in the industry? Why do customers choose what they choose? And then what does management really think? What are they really like?
Interviews, old news reports, etc. can help with the management perspective. So can (the more candid) shareholder letters. Customer behavior is usually best found in data rather than descriptions. Both customers and companies are badly biased in describing why they think customers behave as they do. Usually, data gathered by the industry, government, etc. is useful. Especially useful is the frequency and size of orders, who makes the purchasing decision, etc.
Sometimes studying the history of the industry and other companies in similar situations helps. For smaller companies, it is often possible to learn about the same industry through many different regional examples.
But, overall, I'd say learning about the customer is very, very important. And it is very hard to do just from the SEC reports. Most are badly lacking in this respect. And my lack of comfort with my understanding of a company's customers is often a reason I pass on buying a stock. I tend to prefer investing in companies where I feel I understand customer behavior best.
How you get this information varies. And different investors will choose to get it differently. We are verging on the Phil Fisher scuttlebutt realm of stock research here. It’s important. But it’s often harder to lay out the process as rigidly as when you are doing Ben Graham type research.
One thing I recommend to all investors is reading the notes to the financial statements. These are very important. They tell you a lot about the company.
And some investors don’t read the notes at all.
Talk to Geoff About Reading 10-Ks [email protected]
