CalPERS overhauls tracking-error formula

CalPERS' investment committee voted to switch to what pension fund officials call "actionable tracking error," a risk-control method that now will only take the plan's public assets into consideration.

According to a newly revised investment policy statement containing the tracking-error changes, the $491.8 billion California Public Employees' Retirement System, Sacramento, uses tracking error to define staff discretion to deviate from the pension fund's policy benchmarks. The change, approved Monday, removes a requirement that the asset allocation for its entire portfolio be managed within a target forecast annual tracking error compared to its benchmark of 0.75%.

Instead, the total fund has an active risk target "consistent with forecast tracking error" of up to 1% relative to its benchmark. The methodology removes alternative investments from its calculation as well as eliminates annual tracking-error limits.

The policy requires staff to inform the board of a tracking-error deviation and to develop a plan to move the tracking error back to the target range "if deemed advisable, taking into account strategy horizon, transaction costs, and liquidity conditions."

Separately, as part of its current asset liability study, CalPERS took the novel approach of adopting capital market assumptions that take a both short-term and long-term views. The investment committee adopted capital market assumptions for five years and 20 years to incorporate both a long-term and near-term view of the market. CalPERS usually uses a single time period.

James Sterling Gunn, managing investment director of CalPERS trust level portfolio management program, told the committee that "at some point we think the market might return to a more normal state" than it is now. Although staff does not know when that might be, "five years may be a reasonable horizon for when that inflection point occurs," he said.

The result is that CalPERS would have a "two-step portfolio"— one for the near-term and one for the long-term, Mr. Gunn said.

He said the change reflects the belief that it will give the fund a better outcome than if it just used a single portfolio for the entire period.

Board President Henry Jones asked whether the approach contradicts CalPERS' positioning as a long-term investor. But Mr. Gunn said that while CalPERS is still a long-term investor," we're showing there is a path that involves two portfolios rather than just a single portfolio."

The potential portfolios presented during the capital market assumption presentations also showed test portfolios with and without total fund leverage. Mr.Gunn said that adding more leverage to the total fund portfolio could be used to diversify the portfolio. CalPERS could borrow money to invest in new asset types during a drawdown, for example.

"Today we're dominated by growth,'' Mr. Gunn said. ''If CalPERS adopts the added leverage approach, then strategically we may be able to grow into a more diversified portfolio through use (of) leverage," he said.

Separately, the committee discussed its fiscal year 2021 total portfolio return, which underperformed its benchmark in all time periods. For example, CalPERS' one-year net return for its fiscal year ended June 30 was 21.3%, underperforming its benchmark by 42 basis points.

CalPERS board member Margaret Brown asked staff about CalPERS' returns and why CalPERS is last on Pensions & Investments' list of public pension plan fiscal year 2021 returns.

"There are currently now 66 pension funds on there and they all beat us," Ms. Brown said. "So why did we miss so bad?"

Dan Bienvenue, CalPERS acting CIO, said that the difference in performance has to do with its asset allocation, which aims to protect the portfolio from big drawdowns — a decision that could give up potential positive returns.

"By putting in a 10% allocation to U.S. Treasuries, we know a 10% allocation to U.S. Treasuries is going to underperform … but the decision the board took was that it was worth giving up some of that upside so we could avoid these big drawdowns," he said.

"It's really hard to compare plans, one to another, because really, it's the allocation that drives returns and that comes from" each plan's approach to risk, Mr. Bienvenue said.