By JACK HOUGH
Anyone can call a stock cheap on a gut feeling. To screen for stocks that are cheap based on evidence, however, it's useful to compare their prices to some measure of underlying value that isn't altered by investor fervor. Common measures for the job include the money a company makes (earnings, sales, cash flow), the money it returns to investors (dividends) and the estimated price of stuff it owns (book value). Oddly enough, there's also an expense category that works surprisingly well: research and development spending.
Money spent on corporate overhead subtracts from investor returns, but R&D is different. Companies spend today to create products and processes that will earn top dollar tomorrow. Not all experiments produce riches, of course, but research-intensive companies learn to spend money on the surest things, so R&D turns out to be a pretty good predictor of earnings growth. A 2004 study published in the Journal of Finance found that between 1951 and 2001, companies that suddenly boosted their R&D spending experienced faster profit margin growth and better stock returns than those that didn't.
If R&D predicts earnings, and if a low price-to-earnings ratio can help find attractive shares, a low price-to-R&D ratio should work, too. A 2001 paper that examined the matter found that it does; stocks with low price/R&D ratios outperformed those with high ones by six percentage points a year during the study period.
The three companies below are research-intensive, spending at least 10% of sales in the lab. They've increased their R&D spending by at least 10% over the past year. They also have price/R&D ratios below 30, the median for U.S. companies that spent on research over the past year.
P/R&D ratio: 20
R&D growth, past year: 13%
P/R&D ratio: 20
R&D growth, past year: 36%
Drug sales for Abbott Labs (ABT)
P/R&D ratio: 22
R&D growth, past year: 12%
Analog Devices (ADI)