The Magic of the Magic Formula

August 25, 2022 Printer Friendly Printer Friendly

The Magic Formula is an investing strategy promising that humble retail investors can beat the market (handily) by following a few simple rules, based on a few simple metrics. Sounds magical indeed, right? This highly inviting proposition is what is laid out in Joel Greenblatt’s 2005 bestseller The Little Book That Beats The Market, and reinforced in a 2010 sequel The Little Book That Still Beats The Market. In this article, we’ll explore the premise of the Magic Formula and what teachings it holds for fundamental investors.

Greenblatt is a hedge fund manager and a professor of finance at Columbia University. He is a value investor who believes in buying company shares at well below their intrinsic value (using Benjamin Graham’s concept of a margin of safety), and then holding the stocks until their value is recognized by the market.

That is the classic value paradigm. In his book, Greenblatt attempts to efficiently formula-ize it so that anyone can systematically beat the market, even those don’t have the financial prowess to estimate a company’s value. There is a catch, of course, but we’ll get to that.

Greenblatt backtests his model to show what spectacular returns can be achieved by following it to the letter. According to his testing, the Magic Formula results in an average return of 30.8% over a period of 17 years, an alpha of 18.4% over the S&P 500. When only selecting from the largest 1,000 US-traded companies (which is more realistic for the typical retail investor), the formula provides a return of 22.9%, or 10.5% alpha. Others who have tested the model have produced varying results but generally agree that it produces alpha. Okay, Greenblatt, you have our attention!

Magic Formula Premise

At the crux of the model are two metrics, a version of earnings yield and return on capital (ROC). Greenblatt uses a special version of these metrics that incorporate EBIT (earnings before interest and tax), which means that his model ignores capital structure and focuses only on a company’s ability to generate revenue from its operations. (In Stock Rover, you’ll find the metrics he uses as Greenblatt Earnings Yield and Greenblatt ROC; these are Premium metrics). He wants to find companies with a combination of the highest earnings yield and highest ROC.

A high return on capital not only suggests that a business model is working, but provides further opportunity for a company to invest capital at high rates of return. While the dynamics of capitalism will likely intervene (competition will increase), Greenblatt argues that companies with the very highest ROC have—generally speaking—a special competitive advantage that will benefit them over a long-term period. In other words, they are really good companies.

Of course, a really good company is not necessarily a really good investment. Price matters. Greenblatt’s earnings yield, which is EBIT divided by enterprise value (EV), is meant to find companies that are undervalued. A higher earnings yield means your dollar of investment goes further.

A high earnings yield alone only means that a company is inexpensive, possibly for good reason. But when you combine a high earnings yield with high ROC, you are much more likely to buy, as Greenblatt puts it, “above-average companies at below-average prices.”

There are a few other trimmings to the formula that try to ensure that these metrics do their job, but, in terms of stock selection, that is almost all there is to it. He then provides rules for when and how long to hold the stocks.

The Formula, Already!

Without further ado, here is the Magic Formula (courtesy of Wikipedia):

  1. Establish a minimum market capitalization (usually greater than $50 million).
  2. Exclude utility and financial stocks.
  3. Exclude foreign companies.
  4. Determine company’s earnings yield, which is EBIT / EV. Note that in Stock Rover this special version of the earnings yield is called the Greenblatt Earnings Yield.
  5. Determine the company’s return on capital, which is EBIT / (net fixed assets + working capital). In Stock Rover this is called the Greenblatt ROC.
  6. Rank all companies by highest earnings yield and highest return on capital (ranked as percentages).
  7. Invest in 20–30 highest ranked companies, accumulating 2–3 positions per month over a 12-month period.
  8. Rebalance portfolio once per year, selling losers one week before the year-mark and winners one week after the year mark.
  9. Continue over a long-term (5–10+ year) period.

Did you spot “the catch”?

Arguably, there are a few. One is that the formula is used to buy a sizeable portfolio of stocks, not for individual stock picking. And not everyone can afford a portfolio of 20-30 stocks. Transaction fees can seriously hurt your cost basis if you aren’t buying a solid number of shares of each stock. Let’s call this more of a caveat than a catch. Although the 20-30 stocks guideline is not in reach of everyone, the more important point is that the formula is used to select a basket of stocks, with the idea that they will collectively outperform. Keep this in mind if you plan to adapt the formula in any way to your own investing strategy. It is not a recipe for picking individual stocks; it is a recipe for a portfolio.

Here’s the real catch. Greenblatt says to allow 3-5 years for the strategy to work. In other words, this is a strategy that requires patience and discipline. Patience, because of the long time horizon, and discipline, because the strategy doesn’t always work. In fact, on average it underperformed the market 5 months out of every 12 and it failed to beat the market one out of every four years. As Greenblatt says in the book, “Think it’s easy to stick with a formula that hasn’t worked for several years?…I assure you, it is not.”

Yet this short-term performance variability is also key to the formula’s success. It is precisely why, Greenblatt argues, the strategy will continue to work—because not everyone has the stamina or orientation to follow it. According to the Little Book, if you can outlast Mr. Market’s tempestuous mood swings, you’ll find that he is, in the end, a “very rational fellow” who will get the pricing right. Therefore, if you’ve bought good companies at bargain prices, you stand to gain.

Magic Formula Screening

Here is the main Magic Formula Screener in Stock Rover:


As you can see, this screener sets the formula’s parameters of minimum market cap, acceptable sectors, and US-based companies. The passing stocks are ranked according to Greenblatt earnings yield and Greenblatt ROC, which are each given 50% weight.

You can easily change these criteria (for example, upping the minimum market cap) and set the number of ranked stocks you want it to return. Although it’s not shown in the image above, this particular screener returns only the top 30 ranked stocks, which comes from the formula’s guideline of buying 20-30 stocks. But that number can be modified so that you can see and select from a larger population. In the Little Book, Greenblatt shows that the top 10% of 2,500 ranked companies (i.e. 250 stocks) collectively outperform every other decile, so it’s not strictly necessary to use only the top ranked 30. In fact, there is a fair amount of choice as long as you are more or less within that top decile.

We have created a few ranked screeners based on this formula, which you can download from our library.

Screener Library

Is the Magic Formula for You?

If you are swayed by what you read here, then it’s probably worth picking up Greenblatt’s book, which I can assure you is a very quick and easy read. You can then make a judgment about the degree to which you follow the strategy. If you are already a Greenblatt disciple, let us know in the comments what your experience with the Magic Formula has been.

Even those intrigued by Greenblatt’s argument or the formula’s results may not be inclined to sign up for it, for one perfectly good reason or another. Nonetheless, the Magic Formula still holds some pragmatic teachings for all fundamental investors. Here are a few that we can name:

  1. Be mindful of the collective performance of your stocks. Diversify.
  2. Sell losers before the one-year mark (the loss can offset any taxable gains).
  3. Sell winners only after a year (for the lower long-term capital gains tax).
  4. If your strategy is based on a long-term premise, then you need to evaluate its success only after it has had enough time to be realized.
  5. EBIT / EV (AKA Greenblatt Earnings Yield) and return on capital (AKA Greenblatt ROC) are useful metrics for assessing valuation and profitability, respectively. As mentioned earlier, Premium users have access to the Greenblatt Earnings Yield and Greenblatt ROC in Stock Rover. You can also use the ROA (return on assets) as a stand-in for ROC and you can use operating income as a stand-in for EBIT if you want to calculate the metrics on your own.

This is all common sense. Yet it’s easy to lose our way when Mr. Market is in one of his moods, especially a protracted bad mood. That’s why it’s a good idea to have something to believe in, a strategy (or two) to dispassionately follow and test and sharpen over time. The Magic Formula is just one such strategy, and it could be a worthy experiment for anyone who’s got the time and a long-term mindset.

Editor’s Note: This post was originally published in November 2016 and has been updated for freshness, accuracy, and comprehensiveness.


JProducer says:

I also read the interview – Joel Greenblatt’s Investing Secrets Revealed – Barron’s Oct 15, 2016. (Barron’s had also done previous articles on his formula – 2006, 2009, 2014.)
After reading the Barron’s interview, I got a library copy of his updated book on The Magic Formula. Interesting concept, and thanks for putting it to work in Stock Rover.
Your insights and caveats are also helpful.

Victor says:

How some one can set up Scan to search good Stock? And is there Pre set up Scan for that?

Howard Reisman says:

In the Stock Rover library there are 4 screeners that screen for Greenblatt stocks, each with a different variation – for example Greenblatt and market cap > 1B.

Derp McGillicuddy says:

One thing to point out is that SR’s filters score the Greenblatt metrics on the trailing twelve months (TTM) while Greenblatt himself appears to use the last reported quarter. This will cause significant differences in the stock rankings.

Kenneth Goodman says:

Is there a way to combine screening data. For example, screening for Greenblatt Magic Formula companies with market caps above $1Bil, and Piotroski F-scores of 7 or above, and sort them by momentum factors like relative strength?

Howard Reisman says:

Yes – Stock Rover can do this (with a paid subscription).

Leave a Reply

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

We value your privacy and will not display or share your email address

This site uses Akismet to reduce spam. Learn how your comment data is processed.