The filters we use keep us focused on the factors we consider important rather than responding to someone else’s agenda. Our models rank the universe by quintiles of attractiveness. The factors we consider important fall in three general categories: quality, valuation and catalysts.

Q:  Can you give us some examples of how your process helped you to select certain stocks?

A: Cognizant Technology is a stock that we purchased back in the summer of 2002, when people thought that demand for IT was falling apart. At the same time, there was some controversy about the future margins of the business because of worries about the response of larger competitors who stood to lose market share. But the stock appealed to us because of the huge market potential. You only had to look at the number of total consultants and the pricing of their services relative to the value delivered, to see the enormous potential for a lower-cost competitor to take share.

Cognizant also has a business model that was capital-efficient and highly cashgenerative. The other aspect was the labor pool, which consists of an English speaking, educated, and very technically-skilled population. We also looked at Cognizant's vertical markets, key accounts, and at how deeply they could penetrate these accounts. Then we made projections by verticals and by potential customer within these, and arrived at some fairly large numbers.

We don’t own the stock now in the small-cap portfolio because its market cap grew to about $8 billion, but we still hold it in some of our other products.

Another example would be Resmed, which we have held for a long time. It's a company I’ve known for 10 years. The market they participate in has historically grown 20% per year. We looked at the gross margins, because that's an indicator of the value added, and the margins run in the 60% to 65% range. So it’s a profitable business. Of equal importance, Resmed generates a tremendous amount of operating cash flow relative to net income, so they have been able to fund expansion out of their cash flow, not through a credit line.

The company’s devices and masks address the sleep apnea market. Sleep apnea is highly correlated with obesity and very under-diagnosed. There is a strong and growing body of evidence that sleep apnea contributes to heart disease and stroke, among other things. Resmed is an R&D leader and I’ve watched them go through a number of product cycles. They’ve often been the first to innovate a new mask or a new pressure device that is smaller, quieter, or more efficacious.

The industry has consolidated and become a two-horse race between Resmed and one other company, both of which managed to distance themselves for a number of reasons. In Resmed’s case, they’re a thought leader within the medical community and have superb R&D. Their competitive advantages map onto an under-diagnosed disease within a growing population of overweight people. Ultimately, it meets the requirements of being in a big, fast-growing market, having a capital-efficient business model, and having visibility on sustainable growth for a period of time.

Q:  What are the benchmarks of the portfolio construction process? What is the number of stocks that you own, the exposure limits, and the turnover?

A: Our portfolios hold between 100 and 150 stocks for purposes of diversification and liquidity, which is a big issue in small-caps. In that area trading volumes are lower, so when you're buying or selling stocks, you have to consider the price impact on the market that may erode the alpha between the idea generation and the execution. If you’re trying to put a lot of money to work, you'll impact the price, unless you’re doing it at news events, which is when information gets priced into the stock and liquidity happens.

Another reason for the relatively large number of holdings is that we want the returns to be sustainable as assets scale. From inception, the process and portfolio construction were designed to appeal to sophisticated clients who would want the manager to be able to handle a bigger asset base. The returns we’ve generated aren’t the result of a highly concentrated, nimble product resulting from a low asset base.

We compete against the Russell 2000 Growth Index and our sector over- and underweight tend to be less than 500 basis points around the index weights. The target tracking error for the portfolio is 4% to 8%. About 80% of that is attributable to stock selection and 20% to sector overand underweights, so the majority of our performance comes from stock picking.

The turnover rate is approximately 100%, but the name turnover is lower than the dollar turnover because we trade around positions. We have both upside and downside price targets, and we think about how much we want to own along that spectrum. It's basically owning more towards the downside and less toward the upside. We tend to trade around positions when stocks become temporarily overvalued or undervalued within the context of longerterm solid investment opportunities.

Q:  What are your views on risk controls?

A: I’m very focused on risk management. This gets reflected in how much of a stock we own at any given price, how much we own given the volatility, what active weights we have, and how diversified the portfolio is. We also do risk assessment along many different factors, such as currency exposure or exposure to size, to name a few.

We have a set of analytical tools that help us monitor risk in real time and more extensively than just looking at sector exposures. Our quantitative strategies group has built tool sets to drill down into our portfolios and help us understand if we have, for example, a currency or interest rate bet.

The general idea is that we don’t want any unintended bets in the portfolio. Everything is intentional, which doesn't mean that we’ll always be right, but we’re making investment decisions that are clearly informed by what the risks are. Unintended bets may cost you a lot more than you expected.