FORTUNE -- Before clocking a $100 billion loss in early 2009, the California Public Employees' Retirement System, known as Calpers, had the swagger of a hedge fund and the certainty of a saint. Other pension funds followed its lead, loading up on leverage, investing in unrated CDOs, shoving money into high-priced private equity deals and barreling into commodities and real estate.
The question now is whether a loss of nearly 40% of its market value -- the worst loss in the system's 77-year history -- has brought Calpers sufficiently back down to earth to avoid another such debacle, and whether other chastened pension systems have followed suit. In truth, not all of the evidence of a rebirth at Calpers is comforting. And in the case of some other underfunded pension funds, their latest financial bets look downright scary.
"Some pension plans are evidently hoping to make up losses by taking more risk," says Olivia Mitchell, executive director of The Pension Research Council at the University of Pennsylvania's Wharton School, whose research has shown many pension funds to be poor asset pickers. But, "pension plans that take a risky position to try to 'earn their way out of underfunding' are quite likely to bear big downside risk when the market tanks."
This is not to say that Calpers, the nation's largest pension fund, hasn't made some strides in the right direction. Facing billions in unfunded liabilities and increasing anger from California taxpayers who are ultimately stuck with the bill, it would be tempting for Calpers to double down. But Clark McKinley, a Calpers spokesman, says the pension system "took some bitter medicine" and has learned important lessons from the recent financial crisis.
Calpers is preparing a new asset allocation strategy after finding that its diversification efforts failed to cushion much of the stock market's fall. Calpers is also reining in its exposure to equity derivatives, moving to reduce leverage in its real estate portfolio, terminating under-performing partnerships, tightening the review process for real estate investments and bringing together its diverse investment groups to poll their knowledge about investment risks and opportunities.
Before, McKinley says, information tended to be stove-piped, meaning the retirement system's fixed income department was selling off mortgages even as the real estate department was shoveling money into the sector.
Years of bad bets catch up
Most of Calpers investment losses came from its largely passive investments in baskets of equities, which still account for about half of the system's assets. But the retirement system also got into trouble by adding leverage, reducing oversight and by chasing other hot markets.
After maintaining a low-risk real estate strategy for decades, studies commissioned by Calpers show that it switched gears in 2002, embracing higher levels of risk even as the real estate market began to top out in 2005. By mid-2009, Calpers had a one-year loss of 48.8% in its real estate portfolio and was reporting among the lowest returns of any large pension funds in the country.
In early 2006, it said it would invest $6 billion in commodities, particularly through index futures, news that caused Grants' Interest Rate Observer to respond: "On the timing of this demarche, we hand Calpers the gold medal for Being Late."
Calpers showed even worse timing in the mortgage market. Just before the market tanked, it invested approximately $140 million in unrated collateralized debt obligations (CDOs) and $1.3 billion in complex buckets of loans and debts called structured investment vehicles (SIV). A Calpers lawsuit puts the SIV losses at "perhaps more than $1 billion."
If the investment losses weren't enough, questions have also been raised about why certain asset managers were chosen by Calpers during this period. The state brought fraud charges against the pension system's former chief executive, Fred Buenrostro, and a former board member named Alfred Villalobos, who later became a money-manager placement agent trying to steer pension system business. Both men have denied the allegations.
Calpers is clearly a better managed pension system these days, and has gained with the market rebound over the last year -- although its $202 billion market capitalization still falls roughly $60 billion short of its 2007 high and the plan remains under-funded by many measures.
But Calpers isn't alone in facing funding difficulties. In one influential study last year, researchers Robert Novy-Marx and Joshua Rauh, of the National Bureau of Economic Research, found that pensions are undercapitalized by $3.12 trillion if one assumes the states have to make good on all of their obligations.
Faced with such possible shortfalls, some pension funds have begun to swing for the fences again. Last year, North Carolina passed legislation allowing its state pension to invest in assets such as junk bonds, commodities and real estate. Wisconsin recently decided to increase its pension's bond exposure by levering the portfolio, with one proposal to add leverage up to 120% within two years (in theory this could reduce the portfolio's stock risk and slightly increase overall returns, but only if interest rates stay low). The San Diego County Employees Retirement Association -- which lost about $150 million when Amaranth Advisors went bust -- is among other public pensions ramping up similar schemes; it plans to borrow up to 35 percent of its assets and invest the money in treasury futures, and sock other funds in emerging market debt.
And a recent review of the Illinois Teachers Retirement System by Alexandra Harris of the Medill Journalism school at Northwestern University found that more than 80% of the system's funds were now seen as risky and that it had added leverage by taking the predominantly risky side of over the counter derivative contracts, such as credit default swaps.
Although Calpers has remained active in many potentially risky markets, so far it has refrained from turning to the sort of concentrated bets undertaken in Illinois or Wisconsin. That said, economists say Calpers could still do more to safeguard its assets.
The land of leverage
Despite lowered target allocations, in practice Calpers remains over-leveraged in its real estate portfolio, and over-exposed to the vagaries of the stock market. It is also considering moving money into volatile emerging markets. And it is continuing to hand over money to high fee private equity firms and hedge funds, including a move to invest in more distressed assets with managers such as Apollo Management LP.
Investing in the securities of distressed companies is particularly tricky since the securities, as hedge fund guru Seth Klarman wrote, "are analytically complex and often illiquid," while the "reorganization process is both tedious and highly uncertain." Even the once mighty Apollo reportedly picked up nearly $2 billion in loans that are now halved in value, while one of its large acquisitions, Linens 'n Things, has filed for bankruptcy. Calpers other private equity investments are a mixed bag, with Calpers documents showing that some are showing steeply negative internal rates of return.
Stanford's Institute for Economic Policy Research questions whether Calpers has really benefitted from taking on such risks at all. In an April brief, it noted that "Historically, if Calpers had simply invested in investment_grade corporate bonds, the fund could have earned 7.25%, only 0.66% less than it has earned with its highly volatile portfolio."
However, Calpers is unlikely to put that theory to practice given that it already projects too rosy an outlook in its assumed rate of return. One of its main advisors, Blackrock, notes that Calpers will face sub-par performance for many years. This is particularly worrisome since Calpers' investments have generally made up 75% of every pension dollar needed to cover existing liabilities.
Although Calpers is debating a reduction in its assumed rate of return, doing so would reveal a far larger hole for California's already-strapped taxpayers to fill, as liabilities would explode. Calpers recently requested an additional $600 million in funding from debt-soaked California, which would bring the state government's contribution to the pension system up to about $3.9 billion a year.
The Stanford study says much more will be needed, because "there is less than a 20% likelihood that Calpers' investment returns are sufficient to pay for all existing pension obligations."
Calpers has vigorously disputed the Stanford study's methodology as well as its results. But the pension system is clearly facing some tough funding questions ahead, and is looking at all its options.
In February, shortly after Wisconsin approved leveraging its core fund, a Calpers' board member wanted to discuss using such an idea at Calpers as well. Although the staff presented a primer on the investment, McKinley says that the Calpers board "hasn't pursued the idea, to date."
But with funding short and more than 1.6 million workers expecting retirement and health benefits to come due, the matter will probably come up again. No one knows what tomorrow's liabilities might bring.