On Researching Industries

From a reader:

Hi David,

I’ve been a classic “bottom-up” investment advisor for a few years now, but I agree with your assessment that industries, in general, are under-analyzed by the masses.

What is the best way to learn about a particular industry? Are you aware of any comprehensive publication that sheds light on both the qualitative characteristics of an industry and the appropriate valuation methods?


There are several ways to learn industries.  I’ll try to explain:

1) You can choose a bunch of companies in an industry, email the investor relations area, and ask for packet equivalent to what they send buy-side analysts.  I’ve done that at various points in time for industries I wanted to learn.  Compare and contrast.  Who is doing well, badly and why?  In the mid-90s, I did this for the trucking industry, and learned a ton of information.  I also talked with some trucker friends of mine who gave me on the ground data.

2) You can read industry publications.  When I was a buy-side analyst for the insurance industry, I read those regularly.  They exist for almost every significant industry.

3) You can go to industry meetings.  Almost every industry has meetings where they discuss industry conditions.  Just don’t be too pushy in trying to get information.  Be interested in the industry as a whole, and don’t try to gain material nonpublic information.

4) Value Line & Morningstar both provide industry analyses.  So do most major investment banks.  You can review those and compare and contrast.

5) You can use the quality screen to look at what industries have a rising ratio of gross profits from operations, versus a falling ratio of gross profits from operations.  Here is a chart from the last seven years:


The colored field reading “Chg” is the difference between the average of years 1-3 and years 5-7.  Profits are noisy, that’s why I did an average.

Gross profits from operations as a fraction of assets [GP/A] is a good measure of the quality of an industry, and whether their sustainable competitive advantage is is improving or declining.

Now, when I look at a measure like that, I do one of two things:

  1. I buy cheap companies with strong balance sheets among those industries where GP/A has fallen hard, and buy them, knowing that they are survivors, and will rebound.
  2. I buy moderately strong companies in industries where GP/A has been improving, and after research, the trend is not well understood.  It helps if the industry is dull, and few people follow it.

That’s what I do.  Whatever you do, size it to your own abilities, or the abilities of your firm.  Beyond that, look at cheapness of a company relative to normalized earnings, i.e., average earnings over a full market cycle.


  • Greg says:

    David – Just curious of your reason for using operating earnings (profits) instead of GAAP earnings?

    Both numbers have their flaws of course. But GAAP earnings are (roughly) an apples to apples comparison, the same GAAP flaws apply equally to all.

    “Operating” earnings involve a huge amount of management discretion… some wall street insiders like to mock EBITDA as “earnings before expenses and management mistakes”. The definition of “on-going business units”, which are the only ones that “count” toward operating earnings is not standardized … the company / shareholders are still on the hook for all the losses and problems of the non-continuing business units unless and until they are sold.

    GAAP has plenty of flaws too — but those flaws apply equally to all companies (they are standardized). And equally important, the flaws are well known and well documented … an investor knows the limitations of the model (or can easily look them up).

    I like your idea of looking at 7 years of data — roughly one typical full business cycle.