Joel Greenblatt’s Magic Formula, as described in The Little Book that Beats the Market relies on ranking stocks based on two simple concepts. The first one measures the quality of the business: return on capital, a concept I’ve covered in a previous article. The second statistic measures how cheap a stock is against trailing earnings: earnings yield. But what is earnings yield? It’s not easily available in any run-of-the-mill stock screener. And why not use the more ubiquitous P/E ratio when valuing stocks against trailing earnings?

The first thing we need to cover is how to calculate earnings yield. To do this, we’ll first look at the simple way, and then examine how Greenblatt devised his earnings yield calculation for ranking stocks for the Magic Formula screens. The simple way is just so, simply taking the inverse of the P/E ratio and turning it into a percentage:

Earnings Yield = Net Profit / Market Cap

So, for example, a stock with a P/E ratio of 5 has an earnings yield of 1/5, which is 0.20, or 20%.

By turning P/E into a percentage we’ve already accomplished something useful. In theory, this percentage represents the return on every dollar invested that should be earned by the company, assuming earnings remain flat (a questionable assumption to be sure). This percentage can then be compared directly against the returns offered by alternative investments, such as interest on a bond or savings account. The utility is greater than that provided by the P/E ratio. This is one reason why earnings yield is better than P/E.

Magic Formula earnings yield is a bit more complicated than this. The formula used by Greenblatt to rank stocks is:

MFI Earnings Yield = Operating Earnings / Enterprise Value

One side of this, enterprise value, was discussed in detail in the article linked here. Using enterprise value penalizes companies that carry a lot of debt and little cash, and rewards firms with a lot of cash and little if any debt – a useful distinction not reflected in the P/E ratio. Enterprise value is lower than market cap when a firm carries more cash then it has in debt, and higher than market cap when the debt burden is higher than cash, meaning earnings yield will be higher in the former case, given a constant value for kasegu-hyouban001 operating earnings.

Operating earnings is simply profits before non-operating items like interest, goodwill write-offs, taxes, and so forth. This is also referred to as “earnings before interest and taxes”, or EBIT. By using operating earnings instead of net, we get a better picture of the ongoing profits earned by the company, without the distortions caused by recognized tax benefits or large goodwill write-downs that have very little to do with the company’s profitability.

It’s always instructive to go through a small example that illustrates the value of the Magic Formula way of doing things. For this we’ll use recent MFI stock Pacer International (PACR). Pacer certainly wouldn’t show up on any P/E screens, as it’s net income over the last 12 months is reported as negative $208 million! At a current market cap of $143 million, the P/E would be about -0.7.

But in this case, P/E is misleading – Pacer over the past 12 months is both a profitable and cash flow positive company. To calculate the Magic Formula earnings yield, first we find the enterprise value. For more details on how this is calculated, see the enterprise value article cited earlier.

Enterprise Value = Market Cap + Debt – Excess Cash

= $143 + $80 – $5 = $218 million enterprise value

Now we look at operating earnings and we see the misleading part: close to $290 million dollars in goodwill write-downs account for the entire net loss. While goodwill write-downs are never a good thing (they indicate overpayment for past acquisitions), they do not involve any real, cash operating losses. Aside from these, and interest/taxes, Pacer has booked $42.4 million in operating earnings over the past year. So, the company really is profitable in that time period, despite what the P/E ratio say