Powerful Evidence: Companies with High Returns on Equity Outperform the Market

Over the weekend, I read a book entitled Buffett and Beyond by Dr. J.B. Farwell, who is both an investor and professor of finance. The book presents compelling evidence that stocks of companies with high returns on equity materially outperform the market.

Buffett has been a longtime proponent of investing in these types of companies, which make up the bulk of Berkshire’s portfolio. Also, Joel Greenblatt has presented research which shows that cheap stocks (as measured by the ratio of EBIT/Enterprise Value) with a high return on invested capital outperform the market by a large margin over the long-term. Greenblatt’s approach has been popularized as the “magic formula”.

Buffett and Beyond chronicles Dr. Farwell’s journey as an investor and how he came to focus on stocks of companies that earn a consistently high return on equity.

In order to compare different companies, Farwell normalizes earnings in his calculations by omitting non-recurring items in the income statement and certain non-cash items that would affect a company’s equity, such as a large write-off of a future pension liability. He calls these adjustments the Clean Surplus Accounting Return on Equity. (See the book for exact details.)

The Test

Dr. Farwell tested his ideas by comparing the performance of the eight stocks in the Dow 30 with the performance of the entire DOW 30 and the S&P 500.

Here is his methodology.

“The Portfolio of 8 Dow stocks for any one year was selected by taking the 8 Dow stocks with the highest ROEs for the previous year. Fourth quarter of the previous year was comprised of estimated earnings and dividends, as these numbers are not known with absolute certainty on January 1st.

The 1987 portfolio consisted of the eight stocks out of the Dow 30 with the highest ROEs of 1986.

The calculations were performed on the first day of the year. Thus on January 1st of 1987, the 1986 ROEs were calculated. The eight stocks with the highest ROEs became the 1987 portfolio.

All eight stocks were held for the entire year.”

The Results from 1987 to 2002 – Average Returns

Dow 30 – 13.50% ($100,000 becomes $658,012)

S&P500 – 12.35% ($100,000 becomes $522,651)

DOW TOP 8 18.74% ($100,000 becomes $1,286,085)

Did Farwell Find a Correlation Between the ROE’s and Returns of Stocks?

Dows 30 (average ROE from 1987-2002)

ROE – 14.00%

Total Returns – 13.50%

Dow 8 Stock Portfolio (average ROE from 1987-2002)

ROE – 22.00%

Total Returns – 18.74%

Farwell’s data show a high correlation between a stock’s return on equity and its total return. Keep in mind that the ROE calculations were done using Farwell’s Clean Accounting method.

These are the types of stocks that are included in my watch list. I will add to the watch list each week. The watch list will continue to focus on stocks that have averaged a return on equity of 18-20% over the past 10 years.

Overtime, we can expect Mr. Market to offer shares of some of these companies at compelling bargains. One key to exploit these opportunities is to keep regular tabs on your watch list so you’re not sleeping when opportunity knocks. You will also need to do your own valuation homework so you have the conviction to pull the trigger when the time comes.


7 thoughts on “Powerful Evidence: Companies with High Returns on Equity Outperform the Market

  1. Andrew Schneck

    There is something to keep in mind when analyzing return on equity. The more debt the company has, the less the equity on the balance sheet (in relation to the assets). Considering that the assets, not the equity, are what drive profits, you could very well end up with a highly-leveraged company earning low returns on their assets, but with such a small sliver of equity, the returns on the equity still look quite high.

    Buffett lists at the bottom of his annual report (taken from 2009) that he is looking for companies that:

    (1) Large purchases (at least $75 million of pre-tax earnings unless the business will fit into one of our existing units),
    (2) Demonstrated consistent earning power (future projections are of no interest to us, nor are “turnaround” situations),
    (3) Businesses earning good returns on equity while employing little or no debt,
    (4) Management in place (we can’t supply it),
    (5) Simple businesses (if there’s lots of technology, we won’t understand it),
    (6) An offering price (we don’t want to waste our time or that of the seller by talking, even preliminarily, about a
    transaction when price is unknown).

    Take a look at number 3. Buffett is clearly talking about what I just described; he wants a high return on equity, but doesn’t want it to be because of financial engineering of having high debt loads.

    Here are the equations to consider:

    Return on Equity (ROE) = net income (NI) / Shareholder’s equity (SE)

    Return on Assets (ROA) = NI/SE

    Leverage = Assets/Equity

    If you use algebra, you can split up Return on Equity into two parts; Return on assets & leverage. So, if you do what Buffett is looking for (high ROE, low debt), you end up with a high return on ASSETS.

    The reason Buffett likes ROA so much is that it demonstrates the efficiency of a company’s assets and the intelligence of management for having invested in such high-return assets. Return on equity may be important, but without a ROA of at least 12% (what I like to see), you may end up with an average company that just simply has a lot of debt.

    My equations listed above can best be seen when looking at what is called a “Du Pont” model. (has nothing to do with the company Du Pont). Here is a link to view a Du Pont model:


    you can break up return on assets further to determine where the asset efficiency comes from, but for purposes of this discussion, just keep in mind that debt affects the ROE calculation.


  2. Bob Woronoff

    I enjoy reading your blog. One interesting anomaly to this point is Berkshire Hathaway. The ROE (actually they list the calculation as Return on Shareholders Equity) for BRKB over the last 10 years according to Valueline are:

    2010-7.5% Estimate

    I believe this data supports Andrew Schneck’s point in the previous commenter’s post.

  3. Greg Speicher Post author

    Andrew, great point about debt. I am much more cautious when debt is being used to achieve a high ROE.

    Bob, I think part of the problem with looking at Berkshire’s ROE is that it does not include the earnings from the company’s massive equity holdings. Also, Buffett tends to hold a lot of cash which reduces ROE. In any given year mark-to-market losses on Berkshire’s derivative’s can skew the calculation. Finally, due to Berkshire’s size and it large cash earnings, Buffett has increasingly turned to more capital, lower (but acceptable) ROE businesses such as MidAmerican and Burlington Norther Santa Fe.

    It is worth noting that the ROE of most of Berkshire’s equity holdings is high. Also, the return on average tangible net worth (equity minus goodwill and other intangibles) for the manufacturing, service and retailing operations using normalized earnings is probably in the high teens or low twenties.

  4. Greg Speicher Post author

    Giuseppe, thanks for the link to Jenson. I looked at it briefly and they appear to have a very sound investment philosophy and process. I intend to look at their site in more detail.

  5. Jim Allen

    Andrew, is there a typo here?

    “Return on Equity (ROE) = net income (NI) / Shareholder’s equity (SE)

    Return on Assets (ROA) = NI/SE”

    It looks to me like you mean that Return on Assets ought to be NI/Assets, or something like that.

    Or am I missing something?


Leave a Reply

Your email address will not be published. Required fields are marked *