Every other issue of The Validea Hot List newsletter examines in detail one of John Reese’s computerized Guru Strategies. This latest issue looks at the Peter Lynch-inspired strategy, which has averaged annual returns of 11.2% since its July 2003 inception vs. 6.4% for the S&P 500. Below is an excerpt from the newsletter, along with several recent top-scoring stock ideas from the Lynch-based investment strategy.

Taken from the March 27, 2015 issue of The Validea Hot List

Guru Spotlight: Peter Lynch

During the 13 years he headed Fidelity’s Magellan fund, Peter Lynch became known for his ability to hone in on rapidly growing companies whose stocks were ready to pop. But while Lynch did indeed love his “fast-growers”, he didn’t disregard companies growing much more slowly. Lynch had room for them in his portfolio, too.

Because they were lacking in growth, however, “slow-growers” — stocks of companies growing earnings per share at less than 10% a year over the long term — had to offer something else for Lynch to take notice. That something else was dividends.

For slow-growers, Lynch wanted a high dividend yield, and the model I base on his approach requires dividend yield to be higher than the S&P 500 average and greater than 3 percent.

Lynch famously developed the PE-to-growth ratio as a way to value stocks, dividing a stock’s P/E ratio by its long-term growth rate. His idea was that when a company was producing strong growth, you should be willing to pay a higher multiple for its earnings. For slow-growing stocks, he tweaked the ratio a bit. Since slow-growers often were large firms that paid out nice dividends, he adjusted the “G” portion of the PEG ratio to include dividend yield. For example, consider a stock that has a P/E ratio of 10, EPS growth of 8 percent, and a 4 percent dividend. To find the PEG, you’d divide the P/E (10) by the total of the growth rate and yield (8+4=12). That gives you 10/12 = 0.83 To Lynch, anything under 1.0 indicated that the stock was indeed a good value.

As he did for other stocks, Lynch looked at the inventory/sales ratios of slow-growers, which my Lynch-based model wants to be declining, and the debt/equity ratio, which should be below 80%. (For financial companies, it uses the equity/assets ratio and return on assets rates rather than the debt/equity ratio, since financials typically have to carry a lot of debt as a part of their business.)

The final part of the Lynch strategy includes two bonus categories that apply to all types of stocks: free cash flow/price ratio and net cash/price ratio. Lynch loved it when a stock had a free cash flow/price ratio greater than 35 percent, or a net cash/price ratio over 30 percent. (Lynch defined net cash as cash and marketable securities minus long term debt). Failing these tests doesn’t hurt a stock, however, since these are only bonus criteria.

Lynch’s slow-grower approach is a great example of the fact that you don’t have to target high-growth dynamos to make money in the market. In fact, because so many investors are drawn to stories of explosive growth, oftentimes you can find some great buys among slower growing firms that are getting overlooked. But remember one word of caution Lynch mentioned in his One Up On Wall Street when it came to buying slow-growers: “If you do plan to buy a stock for its dividend, find out if the company is going to be able to pay during recessions and bad times. If a slow grower omits a dividend you’re stuck with a difficult situation: a sluggish enterprise that has little going for it. A company with a 20- or 30-year record of regularly raising the dividend is your best bet.” He says to avoid over-indebted high-dividend stocks (which my Lynch model should do), and adds that cyclical stocks are not always reliable dividend payers.

With all that in mind, here’s a sampling of stocks that pass my Lynch-based model, have dividend yields of at least 3.5%, and have five-year annualized growth rates under 10%. (Note that the PEG ratios listed here are not dividend-yield-adjusted.)

Slow-Growing, High-Dividend Lynch-Based Model Picks (as of April 3)