COLUMN-CalPERS' commodity programme loses 11 percent: John Kemp

(John Kemp is a Reuters market analyst. The views expressed are his own)

By John Kemp

LONDON Aug 7 (Reuters) - The California Public Employees' Retirement System (CalPERS) lost 11 percent investing in commodity derivatives in the last financial year, according to a performance report published on Monday, making it the second worst-performing asset class in the portfolio.

Only the fund's investment in forestland proved more disappointing.

In the 12 months ending on June 30, the fund's commodity investment programme lost 11 percent on an allocation of just over $3 billion, compared with a long-term expected return of 5 percent a year. Losses in the $2 billion forest programme were also 11 percent, which was even more disappointing against a higher expected annual return of 7 percent.

Overall, the fund squeezed out a return of just 1.0 percent on its $233 billion portfolio in 2011/12. Big gains in real estate, infrastructure, inflation-linked bonds and private equity helped offset losses in public equity, commodities, forestland and absolute return strategies, according to reports to be presented to the investment committee on Aug. 13.

A small gain in the commodities programme during June, rising 1.2 percent, was not enough to offset a horrendous May, when the programme shed 12.2 percent in a single month.

Reviewing the poor performance of the programme in his annual report for 2011/12, CalPERS Chief Investment Officer Joe Dear wrote: "Commodities had a difficult year due to global economic uncertainty, natural disasters, and a China slowdown in economic activity."

He failed to mention many developments that have boosted commodity prices, including unrest across the Middle East and North Africa, outages in North Sea oil fields, the shutdown of Japan's nuclear industry and fossil fuel buying, drought, and the continued interest in "real assets" amid a welter of fears about both inflation and deflation.

CalPERS's failure to make money in commodities does not seem to be a short-term problem, and it is not restricted to the last year. Overall, the programme has struggled with exceptional levels of volatility since its inception in October 2007.

While the programme was always expected to deliver variable returns (it has the highest expected standard deviation of any asset class in the CalPERS portfolio) it has failed to achieve consistent underlying performance (Charts 1-2: link.reuters.com/vuh89s).

The commodity programme has lost money heavily since inception. A formal assessment will be published when CalPERS reports performance for the third quarter. But a review of previous investment reports shows the programme had a negative rate of return of 6.9 percent a year between October 2007 and June 2011, and has now capped it off by losing another 11 percent in the last 12 months.

INDEXING RETURNS VANISH

CalPERS is one of the most sophisticated and influential institutional investors. Its early move into commodity derivatives helped win broader acceptance of commodities as a new asset class among traditionally cautious institutional money managers.

CalPERS has only a small team researching and implementing the commodity investment programme, but it has been able to leverage its size and influence to secure the best advice around from sell-side banks, hedge funds and investment consultants.

CalPERS has also followed the currently recommended best practice, allocating roughly 75 percent of its funds to a passive index-based strategy designed to track total returns on the Goldman Sachs Commodity Index with the rest allocated to an actively managed overlay.

The actively managed component is designed to enhance index returns, mostly by under-weighting derivatives stuck in a contango structure and over-weighting those more prone to backwardation.

Indexing with an actively managed overlay is currently advocated by many banks and investment consultants.

But if even CalPERS cannot make money from an enhanced indexing strategy, there must be doubts about whether any other pension funds will be any more successful.

The index-based programme was set up to capture systemic returns from passively investing in a broad basket of commodity derivatives.

Index returns comparable to equities were formally estimated in a paper on "Facts and Fantasies about Commodity Futures" published by Gary Gorton and Geert Rouwenhorst in 2004-2005. But they have since vanished, perhaps because too many investors have chased the same returns, leaving pension funds struggling with losses.

CalPERS' quarter allocation to alpha-generating strategies has helped it outperform the GSCI, but only slightly, and not by enough to avoid the crippling effect of the contango. In effect, the fund has not been able to generate enough alpha to outrun the contango drag on performance, which appears to have doomed the strategy to failure.

TIME FOR A NEW STRATEGY

CalPERS has rejected much of the criticism of last year's poor overall investment performance, countering that many critics fail to understand that it is a long-term investor. "As a long-term investor we fully expect a range of possible returns every year ... we use investment strategy to reach our long-term goals," according to a response posted on the fund's website last month.

"The key to have having a strategy is working with it," CIO Joe Dear warned. "The worst mistake is to abandon the strategy when it appears to have some trouble."

Actually, that is the second-worst mistake. The worst one is to stick with a strategy long after it has proven to fail, persisting with it over and over again and hoping for a different outcome.

CalPERS may be right about the correctness of its overall strategy, but in the commodities segment it is time for a serious rethink.

First generation "passive" index strategies are now widely understood to have failed. With its active management overlay, CalPERS is already beyond this stage.

But there are good reasons to be sceptical about whether second and third-generation "dynamic" and "enhanced" indices will be any more successful. The large amount of institutional money chasing the same sources of systematic return in relatively shallow markets is likely to cause dynamic and enhanced returns to disappear as well.

Second-generation indices have already begun to struggle this year, as the Financial Times noted back in April ("New style commodity indices underperform" April 4).

CalPERS faces a tough choice. It could axe the commodities programme altogether. Or it could switch to a fully active strategy that would see the pension manager behave more like a hedge fund in the commodity sector.

Active management would have its own pitfalls. Not least CalPERS would risk being criticised for fuelling food and energy price swings, though it would actually remain modest by the size of current commodity hedge funds.

But if the giant public pension fund wants to remain in the commodities asset class, while earning positive returns, it may have to take the risk and embrace a far more active approach in future. (Editing by Anthony Barker)