Meet Joel Greenblatt

When we talk about Value Investing, it is difficult to go through the subject without hearing at some point or another about Joel Greenblatt. Professor at the Columbia University Gradual School of Business, he started in 1985 the hedge fund Gotham Capital, and returned from 1985 to 2006 compounded returns of 40% per year  to his investors. (The period 1985-1995 is even more impressive, with annualized returns of 50%). 

Beyond his fantastic results, he is also mainly known for two of his books:

  • You can be a Stock Market Geniuswhich focuses on investing in special situations (spin-offs recapitalizations, merger securities…)
  • The Little Book That Beats The Market, in which he develops a systematized way to invest in good businesses at a cheap price, calling his system “The Magic Formula”

By now it should be clear that Joel Greenblatt is much better at investing than at finding good book titles that don’t sound cringeworthy… But thank God the books’ contents are way better than their cover’s titles! In this article I will focus on his Magic Formula. We will then implement an experiment by composing a virtual portfolio according to Greenblatt’s principles, and we will be able to monitor, month by month and year by year, how this portfolio is performing.

What’s the Magic Formula?

The Magic Formula is based on a simple principle: We want to find good businesses, and buy them at a cheap prices. Warren Buffett applied this principle during most of his career and became one of the wealthiest persons in the world. Greenblatt tried to figure out a systematized way to apply this principle. He then decided to rank companies, using two criteria: 

  • One ranking for figuring out how good companies are
  • A second ranking to reflect how cheap they are selling at

How good is a business?

What is a good business? From an owner point of view, a business is good if it provides a lot of operating returns compared to the capital employed by its operations.
Let’s say we have two companies, Company ABC and Company XYZ, and both need 500’000 CHF to operate. If at the end of the year Company ABC returns 250’000 CHF of operating profits and Company XYZ only 10’000 CHF, it is obvious which company is better from the point of view of their owner.
Company ABC returns 50% on capital, while Company XYZ returns only a meager 2%. So Company ABC is by far a much better business!

But how do we measure Return on Capital? Greenblatt calculates it in the following manner :

Return On Capital = \frac{EBIT}{ Net Working Capital + Net Fixed Assets}

EBIT means Earnings Before Interest and Taxes, and is also often called Operating Earnings. It allows to measure how much the business is generating, before being possibly distorted by the capital structure (debt levels) and tax status of the company. Using EBIT allows us to make apples-to-apples comparisons to know what is the intrinsic earning power of a business. 
Intuitively we do the same thing for instance when we want to invest in a rental property. First we look at how much rent we would collect, and how much costs we would incur with management charges and maintenance (the difference is the EBIT). Only if the EBIT is satisfying do we choose if we are going to buy the property in cash or with a mortgage.
The denominator measures how much capital is needed to operate the business.
Net Working Capital is used because a company has to fund its inventories and accounts receivable, but does need to lay out money for its accounts payables (you can interpret it as a interest-free loan provided by the suppliers of the company). Excess cash is not taken in consideration.
Net Fixed Assets are mainly Property, Plants and Equipments. Intangible Assets, and in particular Goodwill, are dismissed because they represent an excess expense the company has made in the past. In order to continue its operations, the business only needs to replace its tangible assets. Goodwill is an historical cost that does not need to be replaced.

So all companies are ranked by their return on capital and are assigned a note (lower is better): the best company gets note 1. If a company has the 124th best return on capital, it gets note 124.

What is a cheap company?

Now that we know what is a good business, we have to define what is a cheap company. A company is cheap if it has a high Earning Yield, i.e if its business generates a lot of earnings compared to the price we would have to pay if we were to purchase it.

Earnings Yield = \frac{EBIT}{ Enterprise Value}

We already defined EBIT, now let’s explain the denominator. Here Greenblatt does not take the Price of Equity but the Enterprise Value, which represents not only the equity of the business, but also the debt used to finance it.  To continue the comparison with a rental property, let’s say you want to buy a property that sells for 500’000 CHF. The price is still 500’000 CHF whether you pay all in cash or use a mortgage to finance part of it. It’s the same thing for a business. If you want to buy the whole company, its “price”, or Entreprise Value,

Greenblatt ranks then all companies using the Earnings Yield as a second criteria and gives them a second note according to how well they rank : the company with the highest Earnings Yield gets note 1, and so on. If a company is the 485th cheapest one, then it gets note 485.

Putting it all together

Once we have defined what is a good business and what is a cheap business, the Magic Formula will simply try to get the best compromise between good and cheap. Greenblatt will thus create a final score by just adding the two previous notes. For instance, if a company scored 50 in the Return On Capital ranking and 324 in the Earnings Yield ranking, its final note is 50+324 = 374.
By selecting the companies which have the lowest note in this final ranking, we are getting companies that are both good and cheap, or at least presenting the best compromise. Utilities stock and financial companies are removed from the list because of their peculiar balance sheet.

Joel Greenblatt will then buy a basket of around 20-30 companies having the lowest final score on the list, and hold them for one year, before restarting the process. That’s all there is to know to the Magic Formula.

To ease the process for retail investors, the author created a website,  https://www.magicformulainvesting.com/ , that gives you the current rankings of all US companies for the Magic formula.

How well did it work in the past?

In his book, Joel Greenblatt claims returns that are nothing short of impressive, given the systemized approach and the little work to perform.
Let’s judge by the numbers:  

Annual Performance: Magic FormulaS&P 500
198827.1%24.8%
198944.6%18%
19901.7%-16.1%
199170.6%45.6%
199232.4%11.4%
199317.2%15.9%
199422%-4.5%
199534%29%
199617.3%14.9%
199740.4%16.8%
199825.5%-2%
199953%36.1%
20007.9%-16.8%
200169.6%11.5%
2002-4%-24.2%
200379.9%68.8%
200419.3%17.8%
200511.1%4.9%
200628.5%15.8%
2007-8.8%5.5%
2008-39.3%-37%
200942.9%26.5%
Average Yearly Return23.8%9.5%

Annual returns of 23.8% over the period is really outstanding. Except for the years 2007-2008, the magic  formula has consistently beaten the S&P500. Note that the results mentioned above result from applying the Magic Formula on the 3500 biggest companies, with a market capitalization of at least $50 millions.

The Magic Formula Experiment

Curious to see if the Magic Formula would still work, I decided to create an experiment, by setting up a virtual portfolio of 30 stocks that rank best on the website www.magicformulainvesting.com, holding them for one year, and repeating the process every year. This would allow us to monitor the performance, and to verify if the returns claims are indeed impressive.

As a disclaimer, please note that at the time of these writings I am not invested in this portfolio (this is a virtual portfolio), but I will update this statement if in the future I want to implement it.

Magic Formula Portfolio, October 20th 2018

Here is the portfolio that I created at the beginning of the experiment. Let’s see how it will evolve over time!

Company NameTickerMarket Cap($ millions)Price, October 19th (USD)Current Price, Janury 18th 2019Performance
Acorda Therapeutics IncACOR841.9917.8616.26-8.96%
Advanced energy Industries Inc.AEIS1’784.2445.47485.56%
Applied Materials Inc.AMAT33’707.2734.2935.714.14%
Argan IncAGX657.3242.2240.69-3.62%
Block H&R IncHRB5’304.1925.8125.63-0.7%
BP Prudhoe Bay Royalty TrustBPT734.0234.3025.02-26.53%
Casa Systems IncCASA1’206.2613.9911.20-19.94%
Corus Entertainment IncCJREF773.763.724.2413.98%
Deluxe CorpDLX2’486.4052.2144.03-15.67%
DHI Group IncDHX93.511.742.0618.39%
Electro Scientific Industries IncESIO519.2615.0429.9999.40%
Entravision Communications CorpEVC428.184.823.71-23%
Finjan Holdings IncFNJN107.333.952.83-28.355%
FTS International IncFTSI1’366.2712.508.05-35.6%
Gilead Sciences IncGILD94’957.9173.2769.18-5.58%
Immersion CorpIMMR297.179.669.81.45%
Interdigital IncIDCC2’502.7071.972.340.6%
inTest CorpINTT77.697.417.13-3.78%
KLA-Tencor CorpKLAC14’170.7290.8195.735.42%
Mesabi TrustMSB381.9229.1128.88-0.79%
MSG Networks IncMSGN1’977.8126.5122.5-15.13%
Natural Health Trends CorpNHTC250.9522.0615.81-28.33%
Nektar TherapeuticsNKTR8’308.6848.1946.20-4.13%
NIC IncEGOV945.7614.2113.81-2.81%
Ophthotech CorpOPHT81.782.261.39-38.5%
Palatin Technologies IncPTN179.400.890.701-21.24%
Sequential Brands GroupSQBG86.921.360.9898-27.22%
Unisys CorpUIS958.6418.7613.11-30.12%
United Therapeutics CorpUTHR5’780.25122.95115.75-5.86%
Vaalco Energy IncEGY142.612.401.94-19.17%
Total Performance-7.20%