Editor’s note: First in an occasional series.Editor’s note: First in an occasional series.
Changes loom for Washington state pension system
Seattle Times business reporter
As public pension plans come under fire around the country for being overgenerous, underfunded, loosely managed budget-busters, Washington state’s massive system looks pretty good by comparison.
But that may not be good enough.
An analysis by The Seattle Times suggests that the system’s promised benefits are much bigger, and its real assets smaller, than official numbers indicate.
The analysis, using market-based data and methods, pegs the total gap between the present-day value of future benefits and assets on hand at more than $31 billion.
By that measure — one advocated by many outside experts and economists — the system’s nine major defined-benefit plans together have only about 64 cents in assets for every dollar of liabilities, rather than being 100 percent funded as the official figures indicate.
State officials defend the current way of valuing the pension system’s assets and obligations, saying it’s the industry standard.
But the pension industry, prodded by regulators and bond-rating agencies, is slowly moving toward a market-based approach — treating pension promises much like state debt, and recognizing changes in the market value of pension assets immediately rather than smoothing them out over several years.
Advocates of that approach say it paints a truer picture of the pension plans’ financial condition. But adopting it also could increase pressure to rein in benefits or make cities, counties, schools and other public employers pay more into the system — presenting officials at all levels of government with unpalatable choices.
“When the school district has to make higher contributions to the pension system,” said Andrew Biggs, a researcher at the American Enterprise Institute who’s written extensively about public pensions, “that means taxes go up or class sizes go up or kids don’t get new computers.”
Washington’s pension system is one of the largest and most complex in the nation. One of every 14 Washingtonians either receives checks from the state system or is in line to when they retire.
The system has about $73 billion in total assets, split among 11 separate defined-benefit plans and five 401(k)-style defined-contribution plans.
Though managed by the state, the system extends far beyond Olympia. Except for the cities of Seattle, Tacoma, Everett and Spokane, nearly every city, county, port, school system and fire, library or mosquito-control district takes part.
Among the beneficiaries is Dave Proebstel, who worked 26 years as an electrical engineer for the Clallam County Public Utility District before retiring in June. He and his wife, who live outside of Sequim, rely on his state pension for about half of their $6,000-a-month retirement income — enough, he said, to let them live much as they did before he retired.
“We have three kids scattered around the country, and we occasionally visit them,” he said. “We go to Hawaii, and some other places.”
Proebstel said he appreciated the pension’s stability and predictability, as opposed to the stocks in his deferred-compensation account. It takes a lot of the worry out of retirement, he said, and he hopes the pension will be around as long as he is.
“That’s a tough one, given all the uncertainties that pop up once in a while,” he said. “But I’m kind of hoping the Washington employees’ system is guided by smart people.”
Making sure retirees such as Proebstel get what they’ve been promised is largely a matter of matching assets and liabilities — the stocks, bonds, office buildings and other investments owned by the pension system, versus payment promises that stretch many decades into the future.
The state actuary’s office, which does the heavy number-crunching for the pension system, projects that over the next century it will have to pay out some $440 billion across all its defined-benefit plans. (Those plans pay a set amount per month based on the worker’s salary and length of service, which makes it possible to project their future obligations fairly accurately.)
Matt Smith, the state actuary for the past decade, explains that this stream of future payments has to be converted, or “discounted,” back into present-day dollars before it can be sized up against the plan’s assets. The trick is picking the most appropriate discount rate; the bigger the rate, the smaller all those future obligations look in today’s dollars.
If the plan’s discounted liabilities are bigger than its assets, the plan is underfunded.
According to Smith’s current analysis, all but two of Washington’s nine major defined-benefit plans are fully funded, with combined liabilities worth $60.2 billion and assets worth $60.7 billion.The exceptions are two plans covering state workers and teachers that were closed 35 years ago.
Indeed, a recent report from the Pew Center on the States, using 2010 data, found Washington’s pension system was the fourth-healthiest in the nation.
But a growing number of economists, academics and outside pension experts say the standard approach is fundamentally flawed. Most plans, they say, use discount rates that are too high, making the present-day value of their pension promises look smaller than they really are.
“I’ve seen people get so red-faced about this issue,” said Alicia Munnell, director of the Center for Retirement Research at Boston College.
Public pensions such as Washington’s operate under special accounting rules, one of which allows them to assume a long-term rate of return on their investments. Most plans have picked a rate between 7 and 8 percent; all but one of Washington’s plans assume 7.9 percent.
That assumed return is significant, because another special rule lets public plans use it as their discount rate — something corporate pension plans were forced to abandon nearly two decades ago.
Critics such as Munnell and Biggs say this rule ignores the fact that pension benefits are effectively almost as guaranteed as state bonds. That, they say, means they should be valued similarly to bonds.
“The way to value a stream of promised benefits is with an interest rate that reflects the riskiness of the promised benefits themselves, not the expected returns,” Munnell said.
Critics also look askance at another special rule that allows public pensions to smooth out abnormal gains and losses on their investments over several years, rather than recognizing them right away.
Washington smooths returns over as many as eight years, meaning the market plunge of 2008-09 isn’t yet fully reflected in the plans’ official valuations.
So while the actuary’s office put the system’s assets at $60.7 billion as of June 2011 (the date of the most recent assessment), their market value was $57.4 billion — a difference of $3.3 billion, or more than 5 percent.
The current approach is defended by those responsible for making the pension system work.
State Sen. Steve Conway, who chairs the Legislature’s Select Committee on Pension Policy, said smoothing asset values makes for more stable employer contributions, which means less year-to-year bumpiness in state and local government budgets.
“It also stabilizes expectations” from public-employee unions, the Tacoma Democrat added.
Others see the push toward market-based valuation as a stealth attempt to weaken support for public pensions, by making them seem too expensive.
“They’re really trying to step on the air hose of public pensions,” said Steve Hill, who retired in January as director of the state Department of Retirement Systems (DRS), which manages the various plans. “They’re arguing about something that’s not really the issue.”
The real issues, Hill says, is that the bills are now coming due for expensive (and now closed) plans for state workers and teachers, as well as for previous benefit expansions in open plans.
All this might be so much academic quibbling if the pension system’s investment returns were consistently 7.9 percent or more. But they aren’t: The financial crisis and the subsequent recession and bear market pushed down the system’s 10-year average return to as low as 3.87 percent.
Though long-range returns have since recovered and are currently well above 8 percent, the crisis demonstrated that they can’t be taken for granted.
Elsewhere, most public pensions discount their future payouts based on market interest rates. The Netherlands, for instance, has set its rate at 2.42 percent; if Washington operated under conservative Dutch rules, its pension gap would swell to $105 billion.
Back in the United States, the critics’ arguments are slowly gaining favor:
• The California Public Employees’ Retirement System (CalPERS), the nation’s biggest public-pension fund, now uses the market values of its assets to determine its funded status, saying that market value “represents the true measure of the plan’s ability to pay benefits at a given point in time.”
• Last year Indiana’s public pension fund became the nation’s first to drop its assumed rate of return — and hence its discount rate — below 7 percent, citing low bond yields and volatile stock markets.
• Moody’s, one of the three big bond-rating agencies, is considering changing the way it analyzes state pension plans. Its proposed new method, which would use market values of assets and a discount rate based on an index of high-grade corporate bonds, would nearly triple the total unfunded liability of the nation’s state and local pensions, from $766 billion to $2.2 trillion.
• Even the Government Accounting Standards Board, which sets rules for public pension plans and has long defended the actuarial approach, is changing its tune. Under new rules the board issued last year, underfunded plans will have to start discounting their “excess” liabilities using municipal-bond rates.
So how would Washington’s pension system look under a market-based approach?
After consulting with several economists and pension experts, The Times decided to use discount rates derived from yields on long-term state general-obligation bonds, matched to each individual plan’s duration. The rates ranged from 3.48 percent to 6.26 percent.
Those rates were used to compute present-day values of each plan’s projected future payments, using data provided by the state actuary’s office. The values were compared to the market value of each plan’s assets, as disclosed in their financial statements.
Looked at this way, funding levels varied widely from plan to plan. Though none was fully funded, the two plans that cover local police and firefighters came closest: They were about 83 percent funded, generally considered a reasonably healthy level.
On the other hand, the gap between assets and liabilities in the original (and now closed ) state workers’ plan grew from $3.7 billion to nearly $10 billion. The gap in the original teachers’ plan, also now closed, widened from $1.8 billion to $6.8 billion.
Other analysts, using similar approaches, have come up with similarly large estimates of Washington’s “real” pension-funding gap.
Biggs, in a recent study, pegged Washington’s total unfunded liability at $51.8 billion. Joshua Rauh, a finance professor at Stanford’s business school, estimated a $42.9 billion gap.
Washington already has begun decoupling investment assumptions from the rate used to discount future obligations: Under a law passed earlier last year, the discount rate will be gradually lowered, to 7.7 percent by 2017.
State actuary Smith had urged lawmakers to bring the rate down even further, to 7.5 percent, but called the new law a good start.
“I call it ‘reasonable conservatism’ or ‘being on the right side of wrong,’ ” he said. “If you miss your assumptions, is your plan going to be better or worse off?”
But even small reductions to the discount rate — not to mention the larger cuts that a full market-based approach would entail — will widen the gap between the system’s assets and liabilities. That could force employers, workers or both to pay more into the system.
Biggs’ study, for example, estimated that lowering the discount rate to 7.5 percent, as Smith recommended, would increase required employer contributions to the open PERS plans by nearly 25 percent.
But cash-strapped local governments, which already have had to squeeze higher pension contributions into their budgets, likely would resist such a move.
In 2003, for example, King County paid $11.3 million toward employee-retirement plans. This year it expects to pay $69.5 million; next year’s bill is estimated at $83.9 million.
Nor would it be easy to reduce benefits, at least for current employees or retirees.
Dallas Salisbury, president of the Employee Benefit Research Institute in Washington, D.C., cautioned that moving to market-based valuation carries its own risks.
“If interest rates shoot up to 10 percent, for example, it would suddenly appear that every pension plan in the United States was well-funded,” he said. “You’d have pension plans all over the country saying ‘Hey, now we can do big benefit increases.’ ”
Ultimately, Salisbury said, it makes sense for pension managers and overseers to run the numbers multiple ways, under multiple assumptions, but to always err on the side of caution — such as by not permanently increasing benefits during what usually turn out to be temporary bull runs.
But, he added, “no matter what numbers you give people, if they’re inclined to make stupid decisions they’re going to make stupid decisions.”
Drew DeSilver: 206-464-3145 or firstname.lastname@example.org
This story, published March 3, 2013, was corrected March 5. An earlier version indicated Steve Hill was still director of the state Department of Retirement Systems. He retired in January.