For a portfolio based on a fundamental error, my R&D winners picks of April 2004 haven't done too badly. The five stocks in the list were up an average of 32.4% during the past three years. That beat the 27% return on the S&P 500 index, the 21% return on the Nasdaq Composite Index and the 20% return on the Dow Jones Industrial Average for the period.

Imagine what my new R&D winners portfolio might do for the three years ahead, now that I've corrected a basic mistake in my thinking about how research-and-development spending turns into company revenue and earnings growth.

My mistake -- and it's a pretty common one among investors -- was to believe that there's a straightforward connection between the amount of money a company spends on research and development and how fast it can grow revenue and profit. The logic goes that more money spent researching and developing new products means more future revenue from those products and faster earnings growth, since new innovative products command a higher profit margin from the market.

On that principle, I built my R&D winners screen and picked my five stocks on April 14, 2004: Applied Films (acquired by

Applied Materials

(AMAT) - Get Report

in May 2006),

Avid Technology

(AVID) - Get Report

,

Donaldson

(DCI) - Get Report

,

Johnson & Johnson

(JNJ) - Get Report

and

Schlumberger

(SLB) - Get Report

. Returns ranged from a 24% loss on Avid Technology to a 128% gain on Schlumberger.

It Wasn't All Bad

Fortunately, I ran enough fundamental checks on these five companies to make up for the one tiny flaw in my screen: "There are no significant statistical relationships between R&D spending and the primary measures of financial or corporate success: sales and earnings growth, gross and operating profitability, market capitalization growth and total shareholder returns."

That's consulting company Booz Allen Hamilton talking, and I find its conclusion convincing. It has run the numbers twice now -- once in 2005 and again in 2006 -- for a group of companies it calls the Global Innovation 1000. And it has run them with a rigor that no individual investor can hope to match, analyzing seven performance screens from 2000 through 2005.

Each annual study has come to the same conclusion: What counts isn't how much a company spends but how it spends it. The companies that get the biggest bang out of their R&D spending are those that manage research and development as part of an ongoing process that includes customer feedback, design, manufacturing, marketing and -- certainly -- innovation and commercialization.

The annual study -- the most recent version came out in November 2006, and you can download a copy

here -- punches holes in some basic assumptions about R&D. For example, patents don't drive profits. Higher R&D spending can increase the number of a company's patents, but the relationship between the number of patents and a company's financial performance is weak.

Most critically for investors, Booz Allen found that spending more on R&D isn't necessarily better. As the study puts it, "Deep pockets can be dry wells." Just quickly looking at Booz Allen's list of the top 25 R&D spenders in 2005 confirms that point. On the list are such R&D heavyweights -- yet stock market underperformers -- as

Ford Motor

(F) - Get Report

,

Pfizer

(PFE) - Get Report

,

General Motors

(GM) - Get Report

,

Intel

(INTC) - Get Report

and

Sony

(SNE) - Get Report

.

Although there's no way that I can duplicate the statistical heavy lifting that went into the Booz Allen study, I can stand on its shoulders by adding my two tablespoons of investing perspective. The result is my new R&D winners portfolio for 2007.

How did I generate this portfolio?

I began with a list of 94 companies that Booz Allen calls "high-leverage innovators." These are companies that were among the top 1,000 spenders on R&D in the world in 2005 but got more bang for their research spending than their peers.

These companies spent less on R&D as a percentage of overall sales than the median for their industry, yet they got higher sales growth, gross margins, gross profit growth, operating margins, operating income growth, total shareholder returns and market capitalization growth over the past five years. (The full list of 94 is published in the study available on the

Booz Allen Web site.)

Then I went through that list of 94, adding investing perspective. Specifically, I looked for signs -- either technical or fundamental -- that the stocks of these companies were likely to outperform the market average in the next 12 to 18 months.

For U.S. companies, I used MSN Money's

StockScouter tool to get a read on a stock's prospects during that time period. Turns out there are six U.S. stocks with StockScouter ratings of 10 -- the top of the Scouter scale -- among the 94 in the Booz Allen list. They are

Apple

(AAPL) - Get Report

,

Caterpillar

(CAT) - Get Report

,

MEMC Electronic Materials

(WFR)

,

Newmont Mining

(NEM) - Get Report

,

Smith International

( SII) and

Weatherford International

(WFT) - Get Report

. I added all these to my new R&D winners portfolio.

A Look Overseas

Adding investing timeliness for the foreign companies on the list was more difficult for two reasons: StockScouter doesn't rank most overseas companies, and many performance metrics don't translate easily from U.S. to foreign stocks. So I went through the 46 foreign companies on the Booz Allen list one by one, looking at everything from macro market trends to micro internal earnings trends, in a search for stocks that were likely candidates for outperformance in the medium term.

My list of four overseas stocks:

  • Smith & Nephew (SNN) - Get Report, which just acquired Plus Orthopedics. That deal moves the London company into the No. 4 slot in the worldwide orthopedics market. As in most markets, the biggest share of the profits go to the top players, so the company's move up the orthopedics food chain will be a significant driver of increased profit margins in an orthopedics market that will see big growth as the world's population ages.
  • Komatsu, which will benefit from the global infrastructure boom I wrote about in my March 22 column, " Build Your Inflation Hedge With Infrastructure."
  • Hon Hai Precision Industry is the largest electronics contract manufacturer in the world, passing Flextronics International (FLEX) - Get Report in 2004. The Taiwan company will reap the rewards of the current new product cycles from Apple, Cisco Systems (CSCO) - Get Report, Nokia (NOK) - Get Report and Sony. A recent deal to take a large stake in Simplo, the world's largest notebook PC battery packager, moves Hon Hai a step closer to becoming a notebook computer assembler.
  • Companhia Vale do Rio Doce (RIO) - Get Report has made four acquisitions in the past year, the biggest being Canadian nickel giant Inco, just in time to catch the next wave of growth in commodity demand and prices.

At the time of publication, Jim Jubak owned or controlled shares of Schlumberger. He did not own short positions in any stock mentioned in this column.

Jim Jubak is senior markets editor for MSN Money. He is a former senior financial editor at Worth magazine and editor of Venture magazine. Jubak was a Bagehot Business Journalism Fellow at Columbia University and has written two books: "The Worth Guide to Electronic Investing" and "In the Image of the Brain: Breaking the Barrier Between the Human Mind and Intelligent Machines." As an investor, he says he believes the conventional wisdom is always wrong -- but that he will nonetheless go with the herd if he believes there's a profit to be made. He lives in New York. While Jubak cannot provide personalized investment advice or recommendations, he appreciates your feedback;

click here

to send him an email.