How does it work?

Greenblatt sets out his formula in the classic "The Little Book that Beats the Market" - a short and well written book that is a must for any investor's bookshelf.

His approach is very simple to understand - it's all about interest rates. The best businesses out there are the ones that earn the highest rates of interest on the money they invest.

You work this out by calculating a company's return on capital (ROCE). Greenblatt defines this as a company's trading profits divided by its tangible capital invested (stocks plus debtors less trade creditors plus fixed assets). ShareScope and SharePad include all money invested (including intangible assets) when calculating ROCE.

In terms of the price you pay for them, you look for ones with the highest earnings yield - what the business earns compared with the purchase price of the business. This is defined as a company's trading profits divided by the market value of the business (its market capitalisation plus net financial debt). Companies with higher earnings yields (a higher interest rate) are better than those with lower ones.

Greenblatt focuses on trading profits (also known as earnings before interest and tax - or EBIT for short) so that he can look at the profits that the whole business is making. This means that he can easily compare companies without worrying about the big distortions caused by debt levels and different tax rates that show up in measures such as EPS and PE ratios.

There are no predictions of future profits in this formula. All selections are made on the basis of current profits and prices - in other words, what is known now.

That's all well and good but how do you go about finding great companies at cheap prices? It takes a long time to do this with a newspaper and a notepad; you ideally need the help of a decent stock screening product such as ShareScope or SharePad.

Here's the theory:

• Select a universe of stocks with a minimum market value of say £50 million. This is to make sure the shares are likely to be big enough to buy and sell.
• Set a minimum return on assets or return on capital of 25% - rank them in order with the company with the highest returns having a rank of 1.
• Rank companies by their EBIT yield (EBIT/EV). The company with the highest yield will have a rank of 1.
• Ignore all utilities and financial stocks. This formula doesn't work well for these companies due to the way they are financed (with lots of debt).
• Then add the rankings together to get a combined score (the lowest scores are the best).
• Buy 5-7 top ranked companies. Repeat every 2-3 months until you have a portfolio of 20-30 stocks.
• Sell each stock after you have held it for a year and replace it with a new magic formula stock.
• Continue this process for at least three to five years but preferably for as long as you can.

The good news is that you don't have to do all this number crunching. SharePad and ShareScope do all the legwork for you. ShareScope will allow you to sort a list of shares by their Greenblatt rank, whereas SharePad will allow you to sort by their combined magic formula scores.

In SharePad, add a column and select Greenblatt Magic Formula from the Ratios tab. This gives the score for each qualifying stock. In ShareScope the process is almost the same but you'll need to add a Result column. In this case, you'll get the rank for each stock. Sort your column to find the top ranking/lowest scoring stocks. If you need any help with this, please call the Support team on 020 7749 8504.

Does the formula work in the real world?

Greenblatt's book sets out an impressive market-beating track record. Between 1988 and 2009 a magic formula portfolio delivered average annual returns of 23.8% per year compared with 9.6% for the S&P 500 index. It had some stellar years with returns of over 70% in 1991 and 2001 and over 80% in 2003. However it had a bad year in 2008 losing nearly 40% which was worse than the S&P 500.

Its recent performance is not as stellar. Greenblatt's company set up some hedge funds where the magic formula forms a significant part of the investing strategy. As of 9th June 2015, the Gotham Absolute Return Fund has delivered total returns to investors of 44.1% compared with 58.78% for the S&P 500. Greenblatt might argue - as he did in 1999 - that the approach does not beat the market when stock markets are seen to be expensively valued as many think they are today. Over the long run he continues to believe that this strategy will reward patient investors.

Other things to look out for

No strategy works all the time. Blindly investing to a formula takes a lot of emotion out of investing but is not always a good idea. The Magic Formula may lead you to lots of shares that are heavily concentrated in a particular industry that are out of favour making for an unbalanced and risky portfolio.

The return on capital calculation in Greenblatt's formula also ignores intangible assets (SharePad's does not) and fails to adjust for companies that rent - like lots of retailers - rather than own assets or have big pension fund shortfalls. This could lead to companies looking good when they are not.

Enterprise values are usually based on end-of year net debt or net cash figures when a company wants to look its best. These sometimes flatter a company as average debt levels throughout the year can be a lot higher than they are at the year end. This will lead to a lower enterprise value and higher EBIT yield figure than might be warranted. Housebuilders and travel companies are good examples of this.

Watch out for trading costs too. Buying and selling up to thirty companies a year - if you want to try the strategy out - will cost you a lot in dealing commissions and this will reduce portfolio returns, particularly if you are investing smaller amounts of money.

Personally, I might not build a whole portfolio on the magic formula rules. However, the reasoning behind it makes a lot of sense to me and I use ROCE and EBIT yield as two of my key yardsticks when buying shares.