State Pension Funds: What Went Wrong

The conventional wisdom was that employers and employees could put less money into pension funds, and get much larger pensions in the future, because investment earnings would cover most of the cost.

Now in bad economic times CalSTRS, CalPERS and UC Retirement have huge debts and need big increases in government funding, leaving less money for other programs and increasing the pressure for tax and fee hikes.

This year a wave of state and local government pension reforms began reversing the boom-years trend – increasing payments into the funds and rolling back pensions for new hires. Pensions promised current workers are protected by the courts.

Why had it seemed that investment earnings would provide the best of both worlds, lower contributions and higher pensions?

Proposition 21 in 1984 allowed public pension funds to shift most of their investments from predictable bonds to stocks and other investments, with higher yields but also higher risks.

The lid came off as stocks began to soar during a  historic two-decade bull market. The boom gave the three state pension funds surpluses of different sizes, but the response was the same:

Lower annual contributions from employers and employees to the pension funds, and higher pensions for retirees through more generous formulas based on years of service and percentage of final pay or other means.

Bill analyses show that the Legislature was given questionable information about whether key measures for the California State Teachers Retirement System and the California Public Employees Retirement System would increase state debt.

In the case of CalPERS, critics soon began to say the higher pensions threatened government budgets, but there was no change. Then historic pension fund losses in the stock market crash two years ago triggered alarming forecasts of soaring pension costs.

Here’s a look at what happened to the three state pension funds, beginning with the California State Teachers Retirement System, which has received less public attention than the other two funds.




CalSTRS had an unusually low funding level in the mid-1970s, about 30 percent of what was needed for future obligations. A booming stock market and legislation increasing contributions boosted funding to about 110 percent by 2000.

As soon as CalSTRS neared full funding, a half dozen bills increased pensions in various ways. In a time of teacher shortages, advocates said, better retirement benefits would be attractive and keep experienced teachers on the job longer.

Contributions to the CalSTRS pension fund were cut for 10 years by a bill that expired Jan. 1. A quarter of the teacher contributions that would have gone into the pension fund have been diverted to a supplemental program.

The Defined Benefit Supplement Program gives teachers a lump sum or annuity in addition to their pensions.

Under legislation that expired at the end of the year, AB 1509 in 2000, a quarter of the teacher contribution, 2 percent of pay, went into the supplemental program rather than the CalSTRS pension fund, which has had a shortfall for several years.

“No (state) general fund effect and no effect to the solvency of STRS,” said the analyses of AB 1509, unusually brief for legislation shifting billions of dollars. “The STRS surplus will absorb the cost of DBSP (Defined Benefit Supplement Program).”

When AB 1509 moved through regular committees, it dealt with a different issue, credit cards. The cut in contributions to the CalSTRS pension fund apparently was negotiated during the state budget process, presumably with powerful teacher unions.

The supplemental fund holds about $7.4 billion for 544,000 active and retired teachers, a CalSTRS spokesman said. The fund has guaranteed earnings based on the 30-year Treasury bond rate, and can get more when CalSTRS investments do well.

By 2006 the CalSTRS surplus was gone, after a dip in the stock market, and the funding level had dropped back down to 86 percent.

The CalSTRS board, which unlike most public pension funds cannot set employer contribution rates, began an unsuccessful attempt to get legislation giving the board rate-setting power.

The rate policy adopted by the teacher-friendly CalSTRS board would have taken the smallest bite from teachers: increases of up to 0.5 percent of pay for teachers, 1.25 percent for the state, and 4.75 percent for school districts and other employers.

The current total contribution to the CalSTRS pension fund is 20.75 percent of pay. Teachers contribute 8 percent of their pay, school districts and other employers 8.25 percent of pay, and the state general fund 4.5 percent of pay.

After the stock market crash and a decision to lower the earnings forecast from 8 to7.75 percent a year, CalSTRS is now said to need a rate increase of 15 percent of pay to reach full funding in 30 years.

How much additional money is that? One indication is that with the current contribution rate of nearly 21 percent of pay, the total received by CalSTRS in the fiscal year ending in June 2009 was $5.3 billion.

Although the 2 percent of teacher pay diverted for a decade resumes flowing into the CalSTRS pension fund this year, the Defined Benefit Supplement Program will continue to receive some money.

Another bill, AB 2700, expanded retirement credit to teaching beyond a standard year (for example, summer school and overtime) and puts the employer and employee contributions for the work into the supplemental fund. There is no expiration date.

A longevity bonus also enacted a decade ago did expire at the end of the year. The bill, AB 1933, adds $200 a month to the pensions of teachers who serve 30 years, $300 a month for 31 years, and $400 a month for 32 or more years.

Last fiscal year 56,567 retirees were receiving longevity bonuses, costing $19.4 million initially. Continuing to grant the bonuses to new retirees until 2018 would cost $600 million, a CalSTRS staff report estimated.

A package of bills in 1998 began the policy of spending the surplus by cutting contributions and increasing pensions. A reduction in the state contribution to CalSTRS, AB 2804, was expected to save $577 million in the first year.

The contribution cut was linked to another bill, AB 1150, that boosted pensions. A cap at 2 percent of final pay for each year served at age 60 was raised to 2.4 percent at 63. The increase was expected to cost $380 million in the first year.

Yet another bill, AB 1102, made a smaller boost, allowing unused sick leave to count toward years of service. Members who joined CalSTRS prior to 1980 already had the unused sick leave provision.

Two years later retired CalSTRS members got a boost in their pensions. AB 429 in 2000 gave members who retired in 1997 a 1 percent raise in their pensions, which increased with time retired to 6 percent for members who retired before Dec. 31, 1974.

The retroactive pension increase went into the base pension that gets an annual inflation adjustment of 2 percent. A legislative analysis said the increase, similar to one given CalPERS members a year earlier, had a “total present value cost” of $889 million.

* * *

CalPERS sponsored legislation authorizing a major trendsetting pension increase when it had a surplus in 1999, famously telling legislators that state costs would not rise above the previous year’s rate for a decade.

The CalPERS actuaries made an accurate forecast of how state costs could soar if, as happened, investment earnings fell below the target. But a 17-page brochure CalPERS gave to legislators about SB 400 only showed the optimistic scenario.

“This is a special opportunity to restore equity among CalPERS members without it costing a dime of additional taxpayer money,” William Crist, the CalPERS board president at the time, said in the brochure.

CalPERS had strong investment earnings, 20 percent in each of the previous two years and an average of 13.5 percent during the past decade. The part of the pension fund covering most state workers had a funding level of 110.7 percent.

The powerful CalPERS board, which sets employer contribution rates, had given the state a contribution “holiday.” State payments that had been $1.2 billion in fiscal years 1996 and 1997 were dropped to $159 million in 1999 and $157 million in 2000.

State workers, whose contributions are set by legislation, did not get a holiday. Most state workers continued to contribute 5 percent of their pay to CalPERS, and others as much as 8 or 9 percent.

The lack of a contribution holiday for workers was part of the rationale for the pension increase. But the sweeping legislation, SB 400, also allowed workers hired after 1991 to opt out of lower benefits given new hires under former Gov. Pete Wilson.

The legislation created a more generous pension formula for most state workers: 2 percent of final pay for each year served at age 55, increasing to 2.5 percent at 63. The previous formula was 2 percent at 60.

What became the best-known part of SB 400 was a formula negotiated by the Highway Patrol union, 3 percent at 50, up from 2 percent at 50. A patrolman with lengthy service could retire at age 50 with a pension paying up to 90 percent of their final salary.

The new Highway Patrol pension set a standard that spread to many local government police and firefighters in labor negotiations. Critics who say public pensions are “unsustainable” often point to the “3 at 50” pensions.

The landmark SB 400 also gave retired CalPERS members the increase later given CalSTRS retirees. In increments based on time retired, the increase went from 1 percent for persons retiring in 1997 to 6 percent for retirees in 1974 and earlier.

In addition to boosting state worker and non-teaching school pensions, the labor-friendly CalPERS board urged the more than 1,500 local governments in CalPERS to boost their pensions, offering to inflate the value of assets to cover the higher costs.

CalPERS said the SB 400 benefits would be paid for by using “excess” assets, an accounting change that recorded assets at 95 percent of market value instead of 90 percent, and amortizing excess assts over 20 years rather than 30 years.

An Assembly floor analysis of SB 400 said of the CalPERS plan to pay for the benefit increase: “They anticipate that the state’s contribution to CalPERS will remain below the 1998-99 fiscal year for at least the next decade.”

The 17-page CalPERS brochure given to the Legislature said the same thing, but in large boldface type: “CalPERS fully expects the state’s contribution to remain below the 1998/99 fiscal year for at least the next decade.”

The brochure said the state contribution to CalPERS in fiscal 1998-99 was $776 million. SB 400 sailed through the Legislature with little debate or opposition, passing the Senate on a 39-to-0 vote and the Assembly 70-to-7.

The stock market fell soon after SB 400 passed. By 2003 there were media reports that the Legislature had mindlessly rubber stamped a massive pension increase that could cost the state billions. By 2004 the state CalPERS contribution was $2.5 billion.

Former Gov. Arnold Schwarzenegger and other critics pointed to the soaring increase from the contribution four years earlier, $157 million, not to the less dramatic increase, in percentage terms, from the $1.2 billion pre-holiday payment in 1996 and 1997.

Schwarzenegger briefly backed a plan in 2005 to switch all new state and local government hires to 401(k)-style investment plans. CalPERS tried to avoid future rate shocks by adopting a radical “smoothing” plan in 2005 that spreads gains and losses over 15 years.

But CalPERS investment earnings have been on a rollercoaster. The portfolio peaked at $260 billion in the fall of 2007, dropped to $160 billion after the stock market crash in the fall of 2008, and is now back up to about $220 billion.

In another smoothing plan, contribution increases to cover the big CalPERS losses are being phased in over three years. Last June CalPERS set a contribution rate for the current fiscal year, $3.9 billion.

Remarkably, that was almost precisely the amount CalPERS actuaries estimated for 2010 a decade earlier, if the SB 400 rosy scenario was undermined by lower investment earnings. The darker scenario was not shown to the Legislature.

Early last year an official of the trendsetting Highway Patrol union signaled they would negotiate cost-cutting pension changes. One worry was that soaring pension costs could fuel a drive for an initiative to switch to 401(k)-style plans.

The Highway Patrol and several other unions negotiated labor contracts with what has become a common way to cut pension costs: higher worker contributions and lower pensions for new hires.

Schwarzenegger used a record 100-day state budget deadlock to prod the largest state worker union, SEIU Local 1000, into a similar contract. Worker contributions go up from 5 to 8 percent of pay. New hires get the pre-SB 400 pension, 2 percent at 60.

Last month CalPERS cut the state contribution for the current fiscal year by $200 million, reflecting the new agreements by state workers to pay more toward their pensions.

* * *

The UC Retirement Plan seemed to be living the dream of pension fund earning power, going 20 years without contributions from employers or employees while raising pension benefits several times.

Putting no more money in and getting more money out – what could be better?

When University of California regents declared a contribution holiday in 1990, the plan had 137 percent of the funding estimated to be needed to pay for pensions promised current workers in the decades ahead.

Remarkably, after a decade without contributions and with several pension increases the funding level soared even higher, peaking in 2000 during a high-tech boom at 156 percent.

It’s been downhill since then, however, as the stock market slowed during the last decade, finally crashing in 2008. By last July the estimated funding level had dropped to 87 percent.

The big losses are not yet fully reflected under this common reporting method, which spreads gains and losses over five years to “smooth” volatility. Using the market value of assets, the estimated funding level last July was 73 percent.

What would have happened if the UC regents had not declared a contribution holiday and had ignored the growing surplus, mindful of big market swings in the past and standing fast if accused of hoarding or worse?

“Hypothetically, had contributions been made to UCRP during each of the prior 20 years at the Normal Cost level, UCRP would be approximately 120 percent funded today,” a UC staff report said in September.

But now UC faces years or even decades of financial pain. In a plan approved by regents last month, contributions resumed at a low level last April will be increased in steps in an attempt to get back to a full funding level of 100 percent.

The employer contribution, restarted at 4 percent of pay, is expected to reach 20 percent of pay by 2018. One indication of how that translates into dollars is the current “normal cost” (not counting unfunded debt): 17.6 percent of pay or $1.4 billion.

UCLA Chancellor Gene Block told Regents last fall that 20 percent of pay, where contributions would remain for years, is a “breath-taking cost” that could cut faculty recruitment, research and other programs.

The state share of the cost, a third of the total, is said to be the equivalent of 700 faculty positions. The other two-thirds of the $20 billion UC operating budget comes from medical centers, federal sources, grants and other operations.

After a two-decade absence, it will be difficult to get UC pensions back into a state budget that is in crisis and facing years of projected deep deficits. So UC is looking at restructuring debt and borrowing from its own short-term investment pool.

This year, said a staff report, “Student fees had to be used to partially cover the state share of the contributions and campuses had to redirect resources to cover the 4 percent (of pay) university cost.”

Employee costs were restarted at 2 percent of pay and 4 percent of pay above the amount ($106,800 this year) not taxed by federal Social Security. The current employee contributions are expected to peak in 2014 at 8 percent of pay.

New employees hired after June 30, 2013, will get less generous pensions, working five years longer to get benefits under the current formula. The new workers will contribute 7 percent of pay.

In broad terms, the plan returns contributions to the general range of their historic peaks before the long holiday.

“Between 1976 and 1990, contributions to UCRP varied,” said the September staff report. “Employees paid between 5 percent and 7 percent, and the employer contribution went as high as 16.37 percent.”

Beginning a big escalation of pension costs during a severe economic recession adds to the difficulty. And an estimated $6.4 billion pension debt or “unfunded liability” run up during the contribution holiday must be paid off.

Another part of the regents’ plan cuts the UC contribution to most retiree health care insurance, dropping the average from 89 percent of the cost to 70 percent. There is no change for nearly half of employees, who have long service or are near retirement.

The regents approved the plan on a split vote, 14 to 3. Opponents think more should be done to cut retirement costs, perhaps even a switch to the 401(k)-style individual investment plans now common in the private sector.

UC officials said pensions are an important recruiting tool, particularly in competition with elite private universities that only offer unpredictable 401(k)-style retirement plans that can rise and fall with the market.

The trend in the private sector is away from “defined benefit” pensions and toward “defined contribution” 401(k)-style plans, avoiding long-term debt and annual pension costs that can soar, if investments fall short.

Pensions are theoretically less risky for government employers, which are backed by taxpayers and unlikely to go out of business. Losses can be spread over a longer period of time.

Government pensions operate under different accounting rules than private pensions, notably with a higher earnings forecast that lowers the actuarially required contribution needed to fully fund the system.

Part of the rationale for a government pension has been to offset wages that tended to trail the private sector. As UC began the contribution holiday, regents made several increases in pensions, presumably to make working there more attractive.

For example: In 1990 the minimum retirement age dropped from 55 to 50. In 1992 the pension formula was changed from 2.4 percent of final pay for each year served at age 63 to 2.4 percent at 60. In 2001 the formula was changed to 2.5 percent at 60.

In 1999 the regents agreed to allow pensions higher than the federal cap of $245,000 a year if a federal waiver is approved, three dozen top UC executives contend in a letter to the regents last month, the San FranciscoChronicle reported.

The waiver was approved in 2007, but UC President Mark Yudof opposes lifting the cap, which would add millions to UC pension costs. The executives, threatening to sue, accused UC of an ethical breach that jeopardizes the ability to recruit top employees.

* * *

Back-to-bonds investment strategies, considered by CalPERS and CalSTRS after the stock market crash, might be a way to avoid the risk of catastrophic losses in bad times and the temptation of pumping up pensions in good times.

If a portfolio of bonds or similar financial instruments could be structured to produce the earnings needed to pay future pension obligations, some of the current pension problems might shrink or even vanish.

Government budgets would not be at the mercy of markets, which can give or take away, causing annual pension costs to rise or fall. Unrealistic earnings forecasts would not conceal massive debt or “pay for” pension increases later said to be “unsustainable.”

Pension funds could put more money into bonds rebuilding the nation’s crumbling infrastructure, not into overseas markets where future growth is expected or into private equity firms that critics say eliminate jobs and enrich the chosen few.

Pension funds would not be as politicized, no longer using shareholder clout to push “corporate governance” reforms they say are needed to enhance returns, but which are opposed by business groups who say the reforms are a labor-driven agenda.

If done properly, some experts say, a bonds-based strategy could prevent pension fund deficits and surpluses, making annual contributions to the fund predictable.

But there is little sign of a back-to-bonds movement among U.S. public pension funds, even though what some call “liability-driven investing” reportedly is gaining favor among corporate pension funds and some European public pension funds.

Among the problems mentioned by skeptics: matching inflation, future pay increases and other variables, probably through hedging with derivatives, and getting enough yield from bonds to pay for the current level of pension benefits.

But what makes a bonds-based strategy difficult, if not impossible, for most public pension funds is the cost of making the switch. The current structure of the funds assumes that pensions promised current workers will be paid for with strong earnings.

Until shortly before Proposition 21 in 1984 lifted the limit that allowed no more than 25 percent of the portfolio to be in stocks, CalPERS got most of its revenue from employer-employee contributions not investment earnings.

In fiscal 1981 CalPERS received $496 million from members, $1.36 billion from employers and $1.46 billion from investment earnings. Then stock investment earnings during a two-decade bull market that followed brought dramatic change.

In fiscal 1997, two years before the SB 400 benefit increase, CalPERS received $1.4 billion from members, $2.3 billion from employers, and $23.5 billion from investment earnings.

Public pension funds now commonly expect to get two-thirds to three-quarters of their revenue from investment earnings to cover normal costs. After the stock market crash, most of the funds also must make up for big losses.

At a workshop last November on investment allocations, the CalPERS board was shown the option of returning to an all-bonds portfolio. But it was a non-starter. A staff report said the change would require “unrealistic contributions to meet funding goals.”

In September 2009 CalPERS actuaries said a small survivor benefit program with a $63 million surplus offered a “unique opportunity” for a bonds-based test of “immunization” against big cost increases.

The actuaries said the survivor program “could then be used as a model for immunization possibilities in the future should surpluses ever arise in other programs.” The CalPERS board rejected the proposal, preferring current investment methods.

Last July the CalSTRS board heard a pro-and-con presentation from two experts on “liability-driven investing.” A staff report suggested this type of risk management may be better suited to corporate pensions that operate under different accounting rules.

“Finally, given the historically low interest rates and current underfunded liability, the board may wish to re-visit the LDI framework as a viable option to managing the portfolio’s exposure to risk assets after CalSTRS achieves an acceptable funding status level,” the staff report concluded.

* * *

Restructuring board membership was one of the responses in San Diego to the worst-case example of boosting pensions while cutting contributions. Seven of the 13 board members are now financial experts appointed by the mayor.

San Diego is still struggling with pension costs that soared after two deals in 1996 and 2002 linked cuts in the city contribution to higher pensions. State and federal charges against eight former pension officials were dismissed after costly legal battles.

Last month San Jose made outside financial experts a majority on the two city pension boards. Most pension boards not only control investments, but also have the power to set annual contribution rates that must be paid by the government employer.

A consulting firm hired by San Jose in Mayor Chuck Reed’s drive for pension reform argued that boards dominated by “stakeholders” representing management and labor are an outdated model.

Cortex Applied Research of Toronto said amateur boards lack the expertise to properly oversee complicated new ways to manage pension funds, such as “liability-driven investing” and “immunization.”

In addition, said Cortex, a stakeholder board has obvious conflicts of interest. Managers might tend toward decisions that reduce employer contributions, while employee representatives lean toward decisions that boost retirement benefits.

An early adopter of an independent pension board was the Oregon Public Employees Retirement System, which gave three of the five seats to outside financial experts in 2003 after a severe pension funding problem.

The CalPERS and CalSTRS boards arguably are dominated by labor representatives and their allies. But the 26 UC regents, 18 appointed by the governor, are not a typical stakeholder board.

How many regents had financial expertise when the two-decade contribution suspension began in 1990 is not clear. Several current regents have a strong financial background. One of them, Richard Blum, was an early advocate of resuming contributions.

“The idea of having a defined benefit and not paying into it is insanity,” Richard Blum told his fellow Regents at a meeting in November.

* * *

Pension debt moved into the spotlight last April when Stanford graduate students reported that the three state pension funds (CalPERS, CalSTRS and UC) have a combined shortfall of more than $500 billion, not $55 billion as they had been reporting.

Why the tenfold difference?

The students, following the views of some financial economists, used an earnings forecast based on a “risk-free” bond rate, 4.1 percent, to offset or “discount” pension obligations in the decades ahead.

The pension funds, following accepted actuarial practice, use a discount rate based on their forecast of what they expect their assets to earn, which at the time ranged from 7.5 to 8 percent among the three state funds.

Critics say the pension fund earnings forecasts (lowered by CalSTRS from 8 to 7.75 percent last month) are far too optimistic in this era and conceal massive debt owed by taxpayers.

The pension funds ran up the big debt, say the critics, despite a provision in the state constitution that says state debt above a minimal amount must be approved by voters.

The new way of looking at pension debt may be partly acknowledged by the Governmental Accounting Standards Board with a “blended” discount rate – the actuarial forecast for liability covered by assets, and a risk-free rate for any remaining shortfall.

Former Gov. Schwarzenegger’s pension advisor, David Crane, told a GASB hearing last fall that more than a dozen lawmakers who voted for the SB 400 pension boost, all he could contact, did not understand that they were creating state debt.

“I can tell you there are 14 legislators in our state Legislature right now who were there in 1999, who tell me to a person had they known the truth in 1999 they would never have supported that pension increase,” said Crane, a recently appointed UC regent.

Pension reform legislation obtained by Schwarzenegger in October has a “transparency” requirement. When setting the rate paid by employers, CalPERS must include a forecast of higher rates needed if investment earnings are below the forecast.

CalPERS may drop its current earnings forecast, 7.75 percent, to 7.5 or 7.25 percent next month. The additional forecast required by the legislation is based on 6 percent or 1 percent below the earnings assumed by CalPERS, whichever is lowest.

A political battle is expected in Congress over a bill introduced last month by U.S. Rep. Devin Nunes, R-Tulare, requiring state and local pension funds to report their debt using a risk-free bond rate, or become ineligible to issue tax-exempt bonds.

A CalPERS federal lobbyist told the board last month that some think the legislation is part of a move by Republicans to enable state and local governments to declare bankruptcy and undo labor contracts and pension and health care commitments.

“It’s about more than just openness,” Reuters columnist James Pethokoukis wrote. “Some Republicans hope the shock of the newly revealed debt totals will grease the way towards explicitly permitting states to declare bankruptcy.”



Reporter Ed Mendel covered the Capitol in Sacramento for nearly three decades, most recently for the San Diego Union-Tribune.