Sunday, 22 September 2013

Joel Greenblat and Value Investing

I wrote a short posting before about Joel Greenblatt, this time I like to present more material.

The first time I heard about Joel Greenblatt is when a friend recommended this book a long time ago to me:

"You Can Be a Stock Market Genius: Uncover the Secret Hiding Places of Stock Market Profits"

It is a great book about "special situations" (mergers, spinoffs, arbitrages, etc.), definitely not meant for a beginner in stock market investing. The only minor point for people interested in Asian investing is that it is 100% focused on the US market (as unfortunately the large majority of books about investing). It is written in 1985, but still relevant and I would highly recommend it, the logic presented is very compelling.

Twenty years later he followed with:

"The Little Book That Beats the Market"

It provides a rather simple formula to chose and pick shares. The returns, backtested on a reasonable large sample, looked very promising. Basically the formula choses shares with a high ROE at a relative low PE multiple.

[As "K C" rightly pointed out in the comments, Greenblatt actually uses EBIT (Earnings Before Interest and Tax) and EV (Enterprise Value), I use myself ROE and PE, I think the two methods are quite similar]

Since no formula is perfect, users of the formula are encouraged to pick about 20 to 30 different companies, to diversify the risk.

Five years later he followed this up with"

"The Little Book That Still Beats the Market"

An updated version based on the latest data.

I have to admit, I haven't read this latest book, I found the proof in his previous book compelling enough.

Also, the proof that his formula is still working can be found here, a presentation Greenblatt held at the 2009 Value Investing Congress:

The link to the presentation slides can be found here.

Quite amazing that such a simple formula is enough to beat the market, while more than half of the US fund managers are trailing the relevant index.

The Magic Formula's website can be found here.

The holdings of the "Formula Investing US Value Select A (FNSAX)" which uses the formula for its stock selection can be found here.

Big question for Asian investors: would this formula also work, say in Malaysia or Singapore? My guess is it would indeed work. But unfortunately the data to test this assumption is not readily available, like in the US.


  1. Buying great companies at cheap price. How not to work with this type of strategy?

    I have been doing my personal investment this way. The 5 year compounded annual return of my portfolio has out-performed the broad Bursa market by more than twice. ie 27% Vs 12% of KLSE. I am quite sure if anybody back-test Bursa stocks for the last 5-10 years, the results would be quite similar. Bursa is more inefficient where extra-ordinary profits are more likely to be found by the Magic Formula.

    Greenblat actually used ROIC and enterprise value/Ebit for the firm instead of ROE and PE ratio for the equity holders only. He ranked his portfolio of stocks using the two metrics, and rebalance the portfolio every year.

  2. Thanks for sharing, appreciated, and agree with what you write.

    Yes, Bursa is pretty inefficient, I invested myself very actively from 1994 to 2008, also outperforming the index by a large margin (about the same as you describe).

    Focused, good quality companies, not too much debt, high ROE, some dividend, no CG issues, they are often seen as too boring, but not by me (and I guess neither by you).

    Yes, I was aware of EBIT and EV, thanks for pointing out, I used ROE and PE myself, I think not that different, but should have pointed that out to the reader.

  3. Two companies in exactly the same industry but with total different capital structures can have the same ROE, and trading at the same PE. But in term of enterprise value, they are totally different, one can be trading at more than 50% higher in EV than the other. If you were to consider purchasing one of them, which would you prefer to buy?

    EV=market cap+total debt+MI-excess cash

    Example two companies below:

    Company A B
    No. of shares, m 100 100
    Share price 1.00 1.00
    Market Cap, m 100 100
    Total debt, m 50 10
    Excess cash 10 20
    Book value 100 100

    Company A B
    EBIT 15.8 13.8
    Interest -2.5 -0.5
    EBT 13.3 13.3
    Tax @ 25% -3.3 -3.3
    Net income 10.0 10.0
    ROE 10% 10%
    ROTC 11.3% 15.3%

    Enterprise value, m 140 90
    PE 10% 10%
    EV/EBIT 8.9 6.5

    1. Thanks, I agree, also decent example, I normally don't go for D/E above 50%, so first example is (just) inside my limit.

      Another way to look at it is to completely separate the operational part of a company(making the products or delivering the service, generating revenue, COGS, administration, etc.) from the financial part (equity, borrowing, paying interest/dividends, etc.).

      Many companies (many hundreds of Malaysian listed companies) are not good in the first part, they probably should not have been listed.

      Some companies are good in the first part, but not so good in the second (like Panasonic), which is a pity, it is not exactly rocket science.

  4. I agree with KC, using P/E and ROE is fundamentally different from using EV/EBIT and ROIC. What about companies that require large amounts of working capital and fixed assets, which they have funded with debt? What about the differences in accounting assumptions that can change the level earnings below the operating line?

    I'm using a similar screen (EV/EBITDA - so as to take out any differences in accounting treatment across countries) over a majority of the ASEAN countries (Malaysia, Singapore, Vietnam, Philippines, Indonesia). However, I'm also trying to add a level of analysis to the screen because I've found that the screen picks up some dodgy companies (S-Chips, technically listed companies, and subsidiaries of parent companies that are heavily leveraged) and I also want to add a macro overlay, taking into consideration the shifts in FX, which will be material for ASEAN investors (n.b. Indonesia recently). I'm confident that the formula could be applied to SEA markets but I believe there would be more ups and downs caused through corporate frauds, since there isn't the same level of corporate governance / accountability in Asia as the States.

    You can check out the analysis and some of the results at

  5. Thanks, great feedback, will post separately about the issues mentioned.

  6. What I was trying to share in the table is that even two simialr companies with the same ROE and PE ratio, they are not performing the same (ROIC), and their price is not the same too (EV/ebit).

    If you take the return of capital (ROIC), not just ROE, company B did much better with ROIC of 15.3% vs 11.3% of A. Why is ROIC important as you are only interested your return as an equity investor? Because you borrow money to get the same return of other with leverage, definitely it is better as it is less risky. Leverage can cut both ways, it will also amplifies a company's losses if there is a down down.

    In term of value, market enterprise value of company A at 8.9 times ebit is also much expensive than 6.5 times of B. B can replicate the capital structure of A and still get the same ROE and PE ratio by borrowing 40m from bank and pocket it. Then B will have the same net debt of 40m as B.