OWNERSHIP STAMP: Patents for investment processes still unusual

Father and son are among the few recipients

Richard O. Michaud, chief investment officer at New Frontier Advisors LLC in Boston, and his son Robert were awarded a patent for a mathematical process designed to optimize a portfolio by means of resampled efficient frontiers.

The patent is among a flood of business method patents granted since a landmark 1998 court ruling.

Other recent patents include those of Enhanced Investment Technologies Inc., which was granted a patent for a diversity-weighted index that automatically modifies a financial portfolio, and 401(k) Pro Inc. of Los Angeles, which was granted a patent for a software product that allows businesses to run their own 401(k) plans on their computer.

More ahead

Observers believe developments in recent years may lead to even more computer-related business method patents in the financial services industry, but not necessarily more patents for investment processes.

Patent attorney Michael D. Schumann, senior vice president at Merchant & Gould in Minneapolis, said patents for investment processes have been rare, and usually have been tied to a computer model.

In the past few years, the number of patents granted has reached an all-time high and the amount of computer-related business method patents has increased dramatically.

According to the U.S. Patent and Trademark Office in Washington, 169,154 patents were granted in 1999, eclipsing the record of 163,208 established in 1998. More specifically, the number of applications for computer-related business patents has gone from 700 in 1996 to 1,300 in 1998 to about 2,600 in 1999. The number of computer-related business method patents granted in this period went from 143 in 1996 to 583 last year, said Brigid Quinn, deputy director of public affairs.

While the patents encompass a range of ideas, those touching on investment processes in the financial services industry are few.

There are two key reasons for the increase in financial patents overall. One is the Internet and the arrival of e-commerce.

The other is the 1998 federal court ruling in State Street Bank & Trust Co. vs. Signature Financial Group, which basically stated that business methods are indeed patentable.

"What the State Street decision did," said patent attorney Mr. Schumann, "is increase the public's awareness of this issue."

In the State Street case, a federal appeals court overturned a lower court's ruling that involved a patent submitted by Signature Financial Group Inc., New York. Not only was Signature's software system determined to be patentable subject matter, the court also ruled that methods of doing business are patentable.

A sea change

While it was a secondary feature of the case, the method of doing business ruling got all of the attention. "That became the revolution, that became the sea change," said Steven L. Friedman, senior partner at Dilworth Paxson in Philadelphia, Signature's attorney in the case.

While some business method patents were granted prior to the Signature case, it became "a wakeup call to the business world that you can patent a method of doing business," Mr. Friedman said.

As for specific patents, New Frontier Advisors' Richard O. Michaud and his son were granted a patent Dec. 14 for portfolio optimization by means of resampled efficient frontiers. The mathematical process applied in the patent is based on the ideas in Richard Michaud's 1998 book, "Efficient Asset Management."

"The resampled efficient frontier is a fundamentally new way of thinking about portfolio optimality and practice," said Richard Michaud.

The patented process calls for the use of modern statistical data and resampling techniques to create a resampled efficient frontier that is less specific to given data sets than the traditional mean-variance efficient frontier. The resampling process, also known as Monte Carlo simulation, takes the same risk and return data over a specific period of time and looks at different ways they could have evolved. The resampled data create a different efficient frontier by taking the average of all of the alternatives.

Richard Michaud finds portfolio optimizers do the job too well by finding "the largest returns, the smallest risk, the most negative correlations, and it's exactly that type of data that's least likely to show up going forward."

He believes his portfolio optimization process is more intuitive and robust than the traditional models and "just makes more sense."

New Frontier acquired the rights to the patent and will offer the technology through a product called the ROM Optimizer. The firm has not yet licensed the product, but is building technology to offer it over the Internet. Prospective clients would dial up and pay for access to use the technology to build a resampled efficient frontier. He also hopes to form partnerships with brokerage firms or portfolio management vendors.

Balancing a portfolio

Prior to the Michaud patent, Enhanced Investment Technologies Inc., Princeton, N.J., received a patent for its diversity-weighted indexing investment process in 1998. The indexing process uses a mathematical process to weight the stocks in the Standard & Poor's 500 to a certain power, which automatically modifies a portfolio.

Essentially, explained inventor Robert Fernholz, the process is designed to balance the portfolio by automatically taking a little market capitalization weight out of the largest stocks and putting it into the smaller stocks.

Mr. Fernholz is chief investment officer of Enhanced Investment Technologies, a firm that manages about $6 billion in assets, including about $300 million in the patented diversity-weighted product. Because small-cap stocks outperformed large-caps last year, the product has not been as attractive, said Mr. Fernholz, but over time it should be effective as returns revert to the mean.

Because it is a passive indexing strategy, Mr. Fernholz said, "we had to get a patent in order to protect it."

But in general, he said, not many investment processes have been patented, particularly those that involve active strategies.

From a money management perspective, he said, actively managed processes are usually trade secrets that firms don't want to give away.

On the other hand, he said, patenting software that will manage a portfolio might make more sense.

Why few investment patents

Robert Haugen, author of several books including "The New Finance: The Case Against Efficient Markets" and founder of Haugen Custom Financial Systems, Irvine, Calif., agreed that the idea of investment process patents is different for active managers.

"If you're doing qualitative management," he said, it doesn't make as much sense to get a patent because "it's your own human capital that really is the product, and people can't steal that."

"If you're doing quantitative management, it becomes a question of keeping your methods secret. If they get out, you have competitors."

One other recent patent in the pension area -- related more to computers than investment process -- was awarded to 401(k) Pro Inc., Los Angeles, a third-party administrator. Its 401(k) Easy product is a software-based "run-it-yourself" plan administration system.

President James Gilbert hopes 401(k) Easy will "crack open the small plan market so that small companies can have 401(k) plans at an affordable price."

401(k) Easy can be downloaded from the web to a desktop computer. It allows companies to essentially create their own 401(k) plans by selecting a range of mutual funds or self-directed discount brokerage accounts. Once the plan is created there is a one-time processing fee and an annual fee that Mr. Gilbert said is "60% to 80% less than the next least expensive 401(k)."