Quant firms tinkering with factor weightings

A growing number of quantitative equity managers are striving to better match their computer models to changing market environments as they fight to reverse their performance drought since the start of the financial crisis in mid-2007.

Before the crisis, the weightings quant firms allocated to the broad components of their alpha models for factor-driven strategies were “pretty static,” said Soonyong Park, a managing director and head of global portfolio solutions at investment consultant Rogerscasey LLC, Darien, Conn.

At the height of the crisis, however, many found the performance hit from short-term volatility simply “too great to bear,” Mr. Park said. As a result, more firms are working now to make the factor weightings “conditional on the market environment.”

Industry veterans predict those efforts could result in a clearer division of winners and losers than was apparent during the halcyon years leading up to the crisis. Back then, the value and small-cap equity signals favored by the majority of quant firms paid off broadly in terms of both strong investment gains and heavy net inflows.

Another development cited by some observers is a decline in public debate about the factors quant managers are incorporating in their models, reflecting greater efforts to preserve competitive advantages as long as possible.

In mid-May, executives at BlackRock Inc. — the New York-based bond giant that became a quant and index behemoth as well with its December acquisition of Barclays Global Investors — announced they were working to make their models more flexible.

Blake Grossman, vice chairman and head of scientific investments at BlackRock said in a letter to clients and consultants that the firm is “shifting our approach from relatively static tilts toward more dynamic portfolio positioning suited to the specific market context.” The move is one of several “enhancements” aimed at ensuring strong investment performance in the future, he said.

BlackRock has faced some difficult times with its active quantitative equity business, which oversaw $144 billion as of Dec. 31. Several of BlackRock's biggest “alpha tilt” quant offerings — including its $17 billion Alpha Tilts Fund (2% risk); its $6.6 billion Russell 1000 Alpha Tilts Fund and its $4.2 billion Russell 3000 Alpha Tilts Fund — have trailed their benchmarks for the one-, three- and five-year periods through March 31.

Rather than a prolonged stretch of underperformance, however, it was 2007 — when a number of its strategies underperformed by four or five percentage points —that weighed down the three- and five-year numbers. Many of those strategies outperformed their benchmarks the following year, as volatility was peaking.

Ken Kroner, who earlier this month was named BlackRock's chief investment officer and head of scientific active equity, declined to provide details about the changes in the firm's active quant business.

Investment consultants say the broad pressures facing quant managers now have led many to implement — or consider implementing — similar moves. “There were a number of managers that had relatively static models three years ago that are more dynamic today,” noted Keith H. Black, an associate with Chicago-based investment consultant Ennis Knupp & Associates Inc.

Quant giant State Street Global Advisors, which reported $86 billion in active quant equity strategies as of March 31, is one of them. In a recent interview, Richard Lacaille, global CIO, said SSgA has made a “pretty intense effort in the last several years” to study how “different parts of our evaluation process behave” in different market environments.

SSgA introduced some changes at the end of 2008, finding some “interesting evaluation signals that work at certain times very, very well,” but prove much less effective at other times, he said. Those changes appeared to help the firm's active European quantitative equity strategy rebound strongly in 2009, although the results have been less dramatic in other geographic regions, he said.

SSgA's Europe Alpha Equity Strategy outperformed its MSCI Europe benchmark by more than five percentage points in 2009 (after lagging by 1.7 points in 2008 and four points in 2007). That helped lift its annualized return for the three years through March 31 to -0.68%.

SSgA will continue to research the topic, said Mr. Lacaille, who predicted attempts to make models more adaptable to changing market environments will likely become “more important to us in the future.”

Ennis Knupp's Mr. Black said the prospects for benefiting from adjusting quantitative factor weightings to account for a shift in market environment could depend on whether those shifts are “reasonably stable over the medium term.” For example, if it takes three months to adjust the portfolio, and the shift in market leadership doesn't last that long, there's just as much chance that changing weightings could hurt performance, he said.

Some quant veterans predict continued broad industry efforts to make computer-based models more flexible. “I think it is the direction all quants are going,” said Churchill G. Franklin, executive vice president and chief operating officer of Boston-based quant firm Acadian Asset Management LLC. Acadian reported $49.4 billion in quant assets as of March 31.

And for a group of managers whose underperformance between 2007 and 2009 left them open to criticism that their models were all picking the same stocks, “it has the potential to add greater differentiation, as some will get these (models) more or less right and some will get them wrong,” Mr. Franklin said.

Yet another factor leading to greater dispersion could be greater efforts by quant firms to protect their secret sauces. One trend being seen now, said RogersCasey's Mr. Park, is that “these firms no longer publish about their factor insights.”

Being more secretive is one means by which managers can try to combat the “crowding out effect” as competitors figure out — and exploit — the same factors those managers are using to add alpha, Eric J. Petroff, director or research with Seattle-based investment consultant Wurts & Associates, said in a December paper.

“The problem here for investors is the conundrum of needing to be ignorant of how their chosen manager proposes to add value, while at the same time forming an educated evaluation of their ability to do so,” Mr. Petroff wrote.