A Quant Fund with a Human Touch


By Carol A. Wood BlackRock Investment Trust Portfolio/Inv A (CEIAX) has been around in various forms since 1993. The large-cap blend fund got off to a good start, managed in a traditional, fundamental style by PNC Financial Services Group (PNC). In 1998, it was taken over by PNC unit BlackRock (BLK), an investment-advisory firm previously known for its fixed-income funds that has moved increasingly into equities. From 1995 through 1999, the fund racked up double-digit increases. But the tech bust in 2000 set off three years of losses ranging from -15.3% to -26.4%.

In 2003, BlackRock decided that enough was enough and ordained a more disciplined, quantitative-driven management style for the fund. To do that job, it hired a five-man quantitative team from Weiss, Peck & Greer, led by Frederick Herrmann and David Byrket. The two men restructured the BlackRock portfolio in March, 2003.

In January, 2005, BlackRock acquired State Street Research & Management Holdings and merged two of the latter's funds into the revamped BlackRock fund. Renamed BlackRock Investment Trust, the fund retained similar goals and strategies as it held previously.

CHECKING THE INDEX. In short, the aim of BlackRock Investment Trust is to beat the Standard & Poor's 500-stock index by following the sector and capitalization weightings of the benchmark while picking the best-performing individual stocks in each sector. It also strives to resemble the index in terms of style composition (growth vs. value), and doesn't engage in market-timing maneuvers. The fund's R-squared, a statistic that measures how closely a portfolio follows the market, is high, revealing that the fund's performance is closely tied to its benchmark.

Its top five holdings as of Sept. 30 were General Electric (GE

; 3.5% of the fund), Microsoft (MSFT

; 3.2%), Exxon Mobil (XOM

; 2.8%), and Intel (INTC

; 2.5%). The top five sectors were financial services (19.9%), technology (17.1%), health care (12.9%), energy (10.1%), and consumer noncyclicals (7.9%).

As of Sept. 30, the $552.5 million fund had fractionally exceeded the index over three years, registering an average annualized return of 16.9%, vs. 16.7% for the S&P 500 and 14.9% for the average large-cap blend fund. However, the results were achieved with less volatility than its peers and the benchmark, revealing how management has added value. For the one-year period ended last month, the fund was just a hair above the index, returning 12.3%.

"NO BLACK BOXES." BlackRock Investment Trust, ranked 4 Stars by S&P, holds 118 stocks and carries a 1.26% expense ratio, compared with an average of 1.07% for its large-cap blend peers.

The management team uses two computer models to screen for top candidates while removing sector and other biases from the process. "We're allowing the model to tell us which stocks to avoid, as opposed to becoming emotionally attached to names because we made a buy decision on them," Byrket says. Herrmann adds, "Though we do use a quantitative process, there are no gurus, no black boxes -- the foundation is quantitative research."

The team first employs a Barra risk-management model to winnow out chancy stocks from its universe of approximately 800 issues, comprising most of the S&P 500, plus 300 more large- and mid-cap stocks. Stocks outside the index are included if they're similar to those in the index but have the potential to outperform.

AVOIDING THE EXTREMES. Next, the team applies its own proprietary stock-selection model, which grades securities on a varying number of factors broadly related to earnings expectations and relative valuation. In constructing this second, proprietary model, Byrket says his team looks for factors that can predict results consistently over a significant period of time, as demonstrated by back testing. Factors also should be grounded in sound financial and economic theory so the managers can understand what's causing any pricing "anomaly" and how to benefit from it. The model ranks stocks from 1 to 10 relative to their peers. The managers choose those ranked 8 and higher.

The final decision to buy or sell, however, is made by the fund's analytical team, not a computer model. "We take this other step, which differentiates us from a pure quantitative shop," Hermann says. By retaining this element of human judgement in its quantitative methods, the managers eschew overreliance on technology as well as emotion.

Once a stock's pricing anomalies have been recognized, Byrket, Herrmann, and their four analysts review the data to ensure accuracy and to look for "non-model insights" -- events ranging from accounting changes to breaking news. "Not everything you need to know is in the score," Byrket says. "If a stock with a negative change is on our 'buy' list, we will forget about it."

RAPID TURNOVER. For each stock, the team's analyst who covers that sector evaluates the signal that has changed and tries to understand what's behind it. The decision to buy or sell a stock will depend on the severity of the news.

The managers rebalance the portfolio every two weeks, seeking to increase the expected return of holdings by replacing lower-ranked stocks with higher-ranked ones in the same sector. While large-cap equities tend to be priced efficiently due to the amount of information available on them, discrepancies do exist. "We would index the fund if the category were completely efficient," Byrket says. The portfolio's turnover is 72%, compared with 54.4% for its large-cap blend peers.

The managers believe following the benchmark's lead for industry weightings and market cap is the most effective way to reduce the risk of underperforming. Byrket notes: "We focus on specific stock selection as a source of excess returns. All other risks minimize or outweigh the potential return."

Wood is a reporter for Standard & Poor's Fund Advisor


Best LBO Ever
LIMITED-TIME OFFER SUBSCRIBE NOW
 
blog comments powered by Disqus