Have you ever wondered if there was a magic formula that you could use to find winning stocks and invest in them and make handsome gains?
Well, the good news is there is indeed such a formula. In the book The Little Book That Beats the Market (Little Books. Big Profits) Joel Greenblatt layed out a simple quantitative formula. He backtested the formula over 15 years and it beat the market handsomely with lower risk. So, how does the formula work.
The formula tries to keep it very simple. Its intended goal is to buy “good” companies for a “cheap” price. Sounds simple and sensible right. You want to pay less for the best companies. So how does he define Cheap and quality.
1) Cheap : Companies that have high earnings yield. He defined Earnings yield as Earnings Before Interest Tax (EBIT) / Enterprise Value(EV). Now, most of us are familiar with the Price / Earnings ratio or PE. The PE is the multiple that you pay for 1$ in post tax earnings. Example Apple as of the writing of this post is trading at a P/E of 23. That means an investor is paying 23$ for every 1$ in Apple’ post tax earnings. Obviously, paying a smaller multiple is better than paying a higher multiple for the same earnings. i.e I would prefer to buy Apple at 10x earnings than 30x earnings, all things same.
The inverse of P/E is Earnings / Price and is often referred to as the earnings yield. (Similar to a bond yield or a dividend yield.). Professional investors often look at the earnings yield of stocks as it lets you compare different types of investment such as bonds and real estate. You can compare the earnings yield of a stock to a bond yield or the yield on a CD/fixed income investment. Since, the earnings yield is the inverse, we prefer higher earnings yield than lower earnings yield.
Now, Greenblatt does not use Earnings (E/P) but instead uses EBIT/EV. The reason he uses EBIT is that it allows you to compare companies that have different tax rates and different amounts of leverage and hence interest payments. Similarly, he uses Enterprise value instead of the market cap of the equity. Since, EBIT could be higher due to leverage used, using EV makes it a fair comparison
Enterprise Value = Market Cap + Long-Term Debt + Minority Interest + Preferred Stock – Excess Cash.
2) Quality company: Companies that have high returns on capital (ROC). ROC is defined as EBIT / (Net Working Capital + Net Fixed Assets)
# Net Working Capital = Total Current Assets – Total Current Liabilities – Excess Cash
(if Total Current Assets exceeds Total Current Liabilities, otherwise it is zero.)
# Excess Cash is determined. If Total Current Assets are greater than 2 * Total Current Liabilities, then Excess Cash is determined to be the lesser of Cash And Short Term Investments or Total Current Assets – 2 * Total Current Liabilities, otherwise it is zero,
# Net Fixed Assets = Total Assets – Total Current Assets – Total intangible assets
The best part is that he has created a simple screener that anyone can use from the website http://www.magicformulainvesting.com
Remember, both Earnings yield and ROC are backward looking metrics. He does not the forward P/E based on consensus earnings that you will find on many sites. He uses the the EBIT from the last 12 months. Similarly, ROC is based on the recent 4 quarterly financial statements. As you may have often heard, past performance is no indicator for the future performance. However, he argues that making accurate future estimates is difficult even for the professional investors. Plus, their estimates are often over optimistic. Hence, he uses the past to measure both cheap and quality. And he then says to pick 30 such companies in the portfolio.
He also says that the formula does not always work and does not year in year out. There were periods of under performance in his back testing. You may wonder that if everyone were to follow this formula, then it would stop working. Greenblatt says that since the formula has periods where it does not work and under performs the benchmark, people will stop using it. He advices that an investor using the formula give it a minimum of 3-5 years for it to work.
Hope you enjoyed the post. Your feedback is welcome.