By 
Now we've piqued your curiosity, let's see how this academic managed to generate a net return of 30% per year via his hedge fund Gotham Capital from 1985 to 1994.
As mentioned in the introduction, this incredible performance is supposed to come from a formula made up of two ratios, allowing companies to be selected according to profitability and return criteria. The final goal is to find undervalued stocks with high returns on the capital they invest.
But first, let's go back to the two ratios mentioned above. EBIT/EV and ROIC.
EBIT/EV is supposed to be a measure of return on earnings. So the higher the multiple, the better for the investor. Greenblatt uses EBIT (Earnings Before Interest and Taxes) as opposed to net income because it puts companies with different debt levels and tax rates on an equal footing when comparing their returns. ROIC (Return on Invested Capital) is a measure of return on invested capital, either in terms of benchmarking the economic performance of several companies or in terms of value creation by a company. By comparing the ROIC of different companies, it is possible to identify those that generate the highest return on invested capital.
The secret of the magic formula
It is not enough to simply apply these two filters to all companies. A few adjustments are necessary before implementing this formula.
First, the market capitalization of the selected companies must be at least $50 million. This amount, set almost 20 years ago, is no longer really relevant. We therefore advise you to select only companies with a minimum market capitalization of $1 billion.
Financial stocks, i.e. banks, holding companies and investment funds are excluded. Utilities are also excluded. Among these categories we can count water, gas, electricity or energy suppliers in general.
ADRs (American Depositary Receipts) are not eligible for selection. ADRs are foreign shares listed on the US market via a certificate issued by a US bank. In other words, these are foreign companies listed in the US. Famous ADRs include Alibaba, Tencent or Shopify.
In addition, select companies with pre-tax earnings yields above the average for their sector. To do this, simply calculate the EBIT/Enterprise Value and keep only the companies with the highest results.
Another rule is that one should only buy companies with returns on invested capital (ROIC) that are higher than the average for their sector. Again, the calculation is relatively simple: ROIC = EBIT / (net assets + working capital). Keep only the companies with the highest results.
Finally, once these two calculations are done, rank the selected companies according to the highest returns (EBIT/VE) and the best return on investment (ROIC). Greenblatt proposes to do this as follows: if a company is the tenth highest performer related to EBIT/VE (10) and third highest related to ROIC (3) then we give it a score of 10+3=13.
Once all these steps are done, it is enough to buy two to three positions each month in the top 20 to 30 companies and sell each position after one year of holding.
Some studies have also shown that combining the magic formula with momentum filters greatly improves returns. Research has even shown that the total return over the period 1999 - 2001 increased from 235% to 784%
.
If this sounds tempting and very simple to reproduce, wise investors will know how to take a step back from these beautiful promises. In hindsight, it is relatively simple to find a set of criteria that will beat the market. To be wary of these magic formulas is completely normal and proceeds from good sense. That's why we will analyze an open source backtest posted on Reddit a little over a year ago. This backtest will select each company according to Greenblatt's criteria and will resell them at the cost of one year of ownership. So without further ado, let's get to the analysis of the results
As this backtest shows perfectly, the method seems to work perfectly. Between 2003 and 2015. The Greenblatt Formula offered an annualized return of 11.4% (Sharpe ratio of 0.60), against 8.7% for the S&P500 (Sharpe ratio 0.54). Indeed, when we take a step back from this formula, it simply applies some fundamental filters. Even if the skimming remains relatively rough, it already allows you to remove some bad records. To summarize, the performance is clearly superior to the S&P, but so is the volatility. However, the most informed investor knows that the period selected for this kind of backtest plays a very important role. For example, if we select the period 2003 to 2019, in addition to a much higher volatility, the performance of the Greenblatt formula clearly underperforms its benchmark.
An important lesson to be learned from this paper is that there is no magic formula that will ALWAYS beat the market. In hindsight, it is easy to find a set of criteria that beat the market over a period of time. Nevertheless, you can definitely use Greenblatt's criteria of ROIC and EBIT/EV to select above-average stocks, thereby reducing the risk of investing in very bad companies.
