Amy Feldman - Subpage Collage
Subheader - Portfolio

Money, February 2003


Ken Kam used to manage a high-flying tech fund. Now, with Marketocracy, he lets regular folks compete for the right to run money. It’s an odd idea, but so far is seems to be working.

Ken Kam is either very smart or very crazy. The one-time fund manager believes, with an almost religious zeal, that the current way of managing a mutual fund is wrong. Instead of hiring smart young people with degrees in finance or business and teaching them how to analyze stocks, Kam thinks fund companies should draw their managers from the ranks of croupiers, actuaries, unemployed geologists and anyone else who might have a knack for picking stocks. This is no theoretical discussion: With a few million dollars of his own money and backing from venture capitalists, Kam has founded a firm called Marketocracy to see whether his novel idea can turn the fund industry on its head.

For the past 2-1/2 years, Kam has been tracking the results of 65,000 virtual funds set up on the Internet by average Joes. And for the past 14 months, he has been operating a real mutual fund, the Marketocracy Masters 100, which consists of all the stocks in the top 100 virtual funds. From its inception through Dec. 27, 2002, the fund, with assets of $17 million spread over 1,000 stocks, is beating the market by a wide margin, losing 2% vs. the S&P 500’s 21% drop.

Is Kam onto something? Let’s state this flat out: It’s far too soon to know whether his system is anything more than the latest fund gimmick. One year’s results are not meaningful. Morningstar has not rated Masters 100. And institutional investors have yet to invest a cent. But Kam and his concept are definitely worth watching.


It’s a rainy Monday in December in the Silicon Valley suburb of Los Gatos, Calif., where Kam and his execs have set up shop. Kam, a tall and affable 42-year-old, is in full sales mode, pulling out color-coded graphs to illustrate the potential of his idea. One shows the Masters 100 fund outpacing the market through mid-December by 14 percentage points with half the risk (as defined by beta, a measure that compares a portfolio’s volatility with the market’s). Another details the elaborate methodology that Kam uses to choose the investors who pilot his fund.

That methodology is what makes this project so intriguing. Kam is convinced that there are great stock pickers in all walks of life. The trick is to find them, and he has developed a painstaking process for doing that, which we describe below. Kam also believes that different people will do better in different market environments. So rather than rely on a static group of stock pickers, as an ordinary fund would, he does a new ranking every month to determine who will “manage” the fund; usually, about 10 of the virtual funds are replaced. Kam is convinced that he has found a better way to run money, one that could alter the basic risk/reward equation and improve the odds for investors. “This could change people’s thinking about the way the mutual fund industry should be organized,” he says.

Indeed, Kam’s boosters say his approach has implications for the industry as profound as anything since the 1976 advent of the index fund. “It is one of the most interesting experiments I have seen,” says Hersh Shefrin, a professor of finance at Santa Clara University and author of Beyond Greed and Fear, who joined Marketocracy’s advisory board. “Don’t underestimate the difficulty of filtering the skill component. But Ken has put together a business design that gives us an extremely fine mesh for doing so, and the odds of filtering out skill based on a mesh that fine rise substantially.”


When you know Kam’s life story, it’s easy to understand why he would be obsessed with an unconventional approach to money management. Asked about his influences, Kam cites his grandfather, a longtime pineapple picker who became a successful entrepreneur in Hawaii. The elder Kam opened a Chinese candy store across the street from a movie theater, a smart move, since in those days moviegoers could not get sodas and sweets inside. Kam worked in the shop every day after school from age six until he left the island. “I think I learned more about business from my grandfather than from business school,” he says.

While Kam plays down his own drive, friends describe him as ambitious, hard working and very focused. In high school, he joined Junior Achievement, an after-school program that teaches teens entrepreneurial skills, and won a partial college scholarship in its national competition. At Santa Clara University in California, Kam was elected student body president and came up with ideas that helped to straighten out the student body’s finances, recalls Father William Rewak, then the school’s president, who worked closely with him.

After college he found work at Chevron, consulting firm Marakon and medical device maker Novoste Puerto Rico. He got an MBA from Stanford in 1986 and gained money to fund new businesses from the 1993 sale of Novoste, of which he was a part owner. Kam’s first attempts to enter the money-management business ended with his being rebuffed by several firms because, he says, his career path didn’t fit the standard mold, His transition to fund manager started with an investment club called the Winners Circle that he set up with his buddy Kevin Landis. He and Landis subsequently launched Firsthand Funds to capitalize on their direct knowledge of medical devices and semiconductors. The new fund shop stormed onto the scene in 1994 and, while tech was hot, took off on one of the industry’s longest winning streaks. As money poured into Firsthand, Kam wondered how the fund would find enough talented people to invest it all. He tried asking job candidates to bring in their own brokerage statements but couldn’t figure out how to tell if they were showing him only their best accounts. He wanted to try a radical way of opening up the field to those with unconventional backgrounds, to people more like himself. In September 1999, unable to persuade his partners to go for his idea, Kam left to do it himself. (Landis stayed at Firsthand picking tech stocks, but that’s another story.)

As Kam says today: “We had a big disagreement over the direction of the firm. Even friends have disagreements.”


At the time that Kam launched Marketocracy in July 2000, there were already at least two mutual funds trying to use the Internet to tap the wisdom of the masses. The StockJungle Community Intelligence Fund (now defunct) and IPS iFund (down 36% this year) were buying stocks suggested by ordinary people. But Kam had grander ambitions than running a stock popularity contest. He teamed up with Bruce Horn, a Ph.D. software developer who helped create the Apple Macintosh, to hunt for gifted stock-pickers. They lured individuals with the chance to run their own virtual funds. Anyone could sign up for free (at and set up as many as 10 portfolios, each of which would start out with $1 million in play money. Virtual managers had to play by real rules on diversification (no more than 25% of assets in one stock and at least half of the assets in positions of 5% or less) and were charged hypothetical commissions (5 cents a share) on trades. No shorting. No commodities. No obscure foreign stocks. Funds that did well would be rewarded with fake cash inflows, while those whose performance lagged would see outflows. Each fund’s net asset value would start at $10 and be recalculated daily.

Kam says he had hoped for 5,000 responses but was deluged with people who wanted to play. Marketocracy now has 54,000 members running 65,000 funds, and it’s adding another 150 newcomers a week. As all those real people plugged away at fake trades, Kam got what he wanted: a treasure trove of data. With Horn, he analyzed it exhaustively, looking for the precise combination of factors that would produce the best results over time. For example, he found that a fund would perform better using the top 100 portfolios, not the top 10 or 250. He learned that ranking the virtual portfolios on the basis of the previous 12 months’ returns was effective and that reranking them once a month was ideal. He later adjusted each portfolio’s performance using software that filtered out the effects of market movements and the managers’ sector and style biases, and found that the top 100 portfolios based on adjusted data did better than the top 100 based on raw figures.

On Nov. 5, 2001, Kam launched the Masters 100 fund, filling the portfolio with the picks of the top 100 virtual funds. The virtual fund managers are all free to make trades in their own portfolios when they wish. Each day, a computer spits out a spreadsheet that shows the updated holdings in the aggregate virtual portfolio and compares it with the real fund’s holdings. Then Kam and his trader simply make the trades necessary to align the real-life portfolio with the virtual one.


Here’s what the top 100 funds looked like in December: They were run by 88 men and one woman. Seventy-four of the managers were from the U.S. (representing 30 states) and 15 from other countries (Australia, Canada, France, Luxembourg, Switzerland and Thailand). Their median age was 38. Kam originally thought the best stockpickers would be people like himself: nonfinancial types with special insight into their own industries. But the largest chunk of top managers had financial or stock market-related jobs. The others ran the gamut: attorney, chef, croupier, engraver, geologist, inventory auditor, journalist, nurse, stone cutter, student and unemployed. When I spoke with nine of them, including a New York City psychiatry resident, a copyeditor for and a radio network executive who does technical analysis, there were few common threads, except that not one has personally invested in the Masters 100 fund. Some didn’t even believe in the concept of mutual funds.

This multifarious group of people has produced a truly eclectic portfolio. It’s diversified across industries, with heavy doses of basic-materials and consumer stocks. At a time when small-cap value stocks have been hot, it has a small-cap value feel: its average market cap is under $500 million (vs. the S&P 500’s $16 billion) and its price/earnings ratio is 6 vs. the market’s 19. Among the top 10 holdings at the end of November, gold stocks predominated, a sector that soared in the latter part of 2002.

So has Kam really developed a system for identifying gifted money managers? Or is something else behind the fund’s apparent success? Perhaps continuous analysis of the massive amounts of performance data generated by 65,000 virtual portfolios has allowed Marketocracy to ride the market trends before they become apparent. Or perhaps the fund is essentially using a momentum strategy, jumping onto stocks and sectors and bailing as soon as they start to flag.

In any case, it’s simply not possible to draw firm conclusions from such a short spurt. For starters, you’d want at least three years of returns (the fund won’t turn three until November 2004). And it’s easy to imagine all sorts of things that could foul things up before then. If the fund’s assets continue to expland, moving in and out of stocks could get more difficult. Also, it remains to be seen how well the system works if the market abruptly changes direction or something entirely unexpected happens. One need only think back to the implosion of Long-Term Capital Management to be reminded that the history of investing is filled with smart people whose elaborate strategies were undone by the unforeseen.

Some outside observers scoff at the entire undertaking. “It’s a mobocracy,” sneers Roy Weitz, who runs the FundAlarm website. “It looks like a bizarre index made up of amateur picks.” But others are not so quick to dismiss Kam’s creation. “The key is the broadness of the data and their ability to cull out with quantitative methods what is working and what is not,” says Fred Weiss, a pension adviser with Madison Portfolio Consultants, who recently heard Kam speak at an industry gathering and has since been trying to make sense of the fund’s workings. “I think he has captured something that has some persistence. Exactly what he’s got, I’m still trying to get my hands around.”


In hours of conversation with Ken Kam, at the offices and over crab-and-cheese sandwiches and coffee at a local restaurant, it’s clear to me that he’s smart and that he’s spent a lot of time thinking about this concept. He could have kicked back after Firsthand and done something easier, and you can’t help but respect him for choosing to do something hard. “There’s no way to shortcut this,” he says. “We just have to run it for some more time.”

Kam’s ultimate goal is to have a series of funds run by computer-selected investors. Each would be based on different ways of slicing and dicing the data. He’s looking at a Masters 5 fund, where the top five investors would pick stocks, or a Masters 10 fund, limited to the top 10, but he believes he needs much more data for both. He’s considering sector funds and funds that would fall into each of the nine mutual fund style boxes as defined by Morningstar. He’s even considering using the data on the worst pickers to short stocks. He’d like to open at least one new fund in 2003. “Every year that goes by, we will have more data and more knowledge about how to use it,” Kam says. “We’re going to let the data drive the decisions.”

But for now, even the Masters 100 fund is just a big experiment. And for investors, the fees, 1.95% vs. an average of 1.26% for U.S. equity funds, make it an expensive one. If Kam is right, Marketocracy could become to mutual funds what eBay is to retailing: a radically different way to run a stodgy business. If he’s wrong, there is no shortage of fascinating but ultimately flawed investing ideas that it could join in the graveyard. Whichever turns out to be true, I, for one, can’t wait to see what this massive database will show about how investors pick stocks and whether they can do so with anything close to true skill.

Kevin Fravel is one of Marketocracy’s champs. His virtual fund, up 127% since its inception in August 2000, is currently ranked No.2. You could consider Fravel, a 34-year-old unemployed computer programmer from suburban Philadelphia, a fluke. “I started it just to have fun and to teach myself to be a better investor,” says Fravel. When he opened his virtual fund, Fravel put 10% of the assets in each of five picks. One of his favorites that subsequently soared was a little drug firm called Bradley Pharmaceuticals that makes tar shampoo and prescription vitamins. That pick, plus bets on a number of home builders, have been good for Fravel’s personal finances, too: His own portfolio consists of many of the same stocks in his virtual fund. In the working world, he’s been less successful. Out of a job for seven months, Fravel says he’s considering teaching math or working for the government. Ultimately, Fravel hopes he’ll be hired by Marketocracy to run his own fund. “It’s just a hobby,” he says, “but one that I take very seriously.”

Christian Straessle is one of the few professional investors playing the Marketocracy game, and he’s got two virtual funds in the Masters 100. The 30-year-old runs institutional money for ValueFocus Equity Management in Wil, Switzerland. Straessle says he uses the same value strategy for his online picks at Marketocracy and his real portfolios, and rarely trades. His favorites include restaurant chain Applebee’s International and Ross Stores, a cutrate apparel retailer. Why play the game against amateurs? “My colleagues say, ‘I don’t want to compete with these nonprofessionals.’ It can look very bad if you deliver bad performance,” says Straessle. “I like the competition. I want to show that I can manage portfolios, virtual and real.” Bio photo