The looming retirement disaster

Memo to President Bush and Congress: It's a retirement crisis, not merely a Social Security crisis.

All three legs of the stool on which America's workers are expected to rely for retirement — Social Security, employer-provided retirement plans and personal savings — are deteriorating, and are likely to be unable to support the weight that will be placed upon them in the next four decades.

The coming collision of demographics and economics is driving the United States toward a major pension crisis that will require creative solutions from government and employers, and behavioral changes from employees.

Without such changes, many retirees will be left living below — in some cases well below — their accustomed standards. It may already be too late for some older workers, since the first of the 77 million baby boomers will retire in 2008.

The reason is a nasty combination of poor pension legislation; significant pension plan underfunding, because of poorly designed accounting and funding rules and overly optimistic asset return assumptions; rising pension liabilities because of a rapidly growing retiree population; insufficient participation by employees in employer-sponsored defined contribution plans; a steadily eroding Social Security program; and paltry personal savings habits.

Pension experts, academics, economists and even politicians agree the time is ripe for reform. The question is whether regulators and legislators have the political will and intelligence to do it.

The numbers paint a bleak picture:

• More than half of the nation's 132 million workers do not participate in any kind of employer-provided pension or retirement plan.

• The number of employer-sponsored defined benefit pension plans has declined precipitously, to 31,200 in 2004 from 114,000 in 1985.

• More than 1,000 employers voluntarily shut down their defined benefit pension plans every year, leaving a shrinking pool of employers paying into the federal retirement insurance program that funds the pensions of bankrupt companies.

• People born in 1964 or later face a scheduled increase in the full retirement age for Social Security benefits to 67, as well as substantially higher taxation of those benefits even before a contemplated overhaul of the system that could cut benefits or raise taxes further.

• By 2030, when nearly the entire baby boom generation will be in retirement, Social Security will provide only 26.9% of retirement income, down from 38% for current retirees. With life expectancy continuing to rise, some older Americans will be in retirement for 20 or 30 years.

• The national savings rate, which had already plummeted to an anemic 3.5% by the mid-1990s from a robust 12.3% in 1950, has continued to decline to as low as 1%, by some estimates.

• People in their mid-50s and early 60s had median savings of just $42,000 in retirement accounts in 2001, well below the estimated $1 million that financial advisers estimate the average household with an annual income of about $40,000 will need to ensure a comfortable retirement.

• Escalating health-care costs threaten to exceed those of employer-provided retirement benefits. In 1960, 48.6% of every dollar companies spent on employee benefits went toward retirement benefits, and 12.4% went toward health care for employees and retirees. By 2003, employers spent 42.8% of every dollar of employee benefits on retirement and 31.9% on health care, including retiree health care, according to the Washington-based Employee Benefit Research Institute.

"We're moving into uncharted waters. That's what I'm concerned about," said Sylvester J. Schieber, director of research at employee benefits consultant Watson Wyatt Worldwide, Washington, and a prolific author of retirement studies.

Alicia H. Munnell, director of the Center for Retirement Research at Boston College was even more blunt.

"People are going to find themselves in their 80s and 90s with inadequate savings," she warned. "People are really going to have to change their work life. This business of retiring at 62 …will just not work."

The employer-sponsored pension system is arguably the weakest leg of the retirement stool. While Social Security receives a lot of headlines, it will not run into serious financial strain for another 36 years. And although personal savings rates are the lowest ever, many people have savings locked up in their houses; and some economists, counting assets in 401(k) plans and individual retirement accounts, argue that personal savings rates are in the 8% to 9% range.

But defined benefit plans have been in steep decline for nearly 20 years as employers have sought to shift responsibility onto employees amid rising costs of administering the plans.

The number of corporate defined benefit plans insured by the Pension Benefit Guaranty Corp. declined to just 29,512 in 2003, and by some estimates has declined further to as few as 23,000 in 2005, from a peak of 112,208 in 1985. The biggest decline was in plans with fewer than 100 employees. The number of these plans dropped by 80% between 1985 and 2003. But the wave of terminations has reached larger plans. The number of plans with more than 10,000 employees peaked at 644 in 2001 and had dropped to 624 in 2003. More terminations have occurred since.

Plans with between 5,000 and 9,999 employees peaked at 656 in 1998 and dropped to 509 by 2003.

Other large plans have not been terminated but are closed to new employees, meaning they will eventually wither.

In addition, economists at Morgan Stanley, New York, estimate the companies in the Standard & Poor's 500 index have a combined pension shortfall of about $170 billion. Overall, the PBGC estimates of corporate pension plans' underfunding exceed $450 billion.

The accounting rules companies use to measure pension assets allow them to smooth changes over a five-year period to eliminate volatility, but critics argue that this practice also allows them to overstate assets. Liabilities are discounted by the interest rate on high-quality corporate bonds.

"There are two essential reasons why defined benefit plans are losing ground very rapidly," explained Richard Berner, Morgan Stanley's chief U.S. economist. "One is the lack of portability. The second one is accounting and funding rules."

Plan sponsors "used to think that because you could invest in equities, which have a higher expected rate of return, the cost of defined benefit plans was lower. But what was being ignored was that the reason equities have a higher expected rate of return is because there is a risk premium," he Mr. Berner said. "That risk has got a cost. When you factor in the risk cost, the price of defined benefit plans is much higher than everyone was assuming. Consequently, corporations are moving out of defined benefit plans."

While many employer representatives place the blame of declining pension plan coverage squarely on the Employee Retirement Income Security Act — the 1974 law that established the quasi-governmental PBGC and set funding rules and other guidelines — lawmakers and regulators were fully aware of the calculated risks in enacting the protections for plan participants.

The tradeoff lawmakers and regulators settled for was a better funded pension system, from which smaller employers and others unwilling to bear the increased costs of offering greater protection to participants were expected to exit.

And they did. By the thousands.

By the end of the 1970s, numerous plan sponsors had shut down their pension plans without having amended them to comply with ERISA's minimum vesting standards or paying a dime in insurance premiums to the PBGC. Thousands of former participants in those plans sued for benefits from the PBGC; the federal pension insurer has to date paid out more than $1 billion in benefits to more than 87,000 participants.

Adding to the regulatory burden was the Tax Reform Act of 1986, which imposed a penalty on companies wanting to siphon off surplus pension assets. In the federal government's fiscal year ended September 1987, the PBGC recorded 10,865 fully funded plans that were shuttered. A 1990 study by the PBGC acknowledged that administrative costs of defined benefit pension plans for small companies doubled between 1981 and 1991.

James A. Wooten, a professor of law at the State University of New York, Buffalo, noted that the effect of the major ERISA reforms — creation of minimum vesting and funding rules, as well as termination insurance — was the same as any other consumer protection law: driving up costs.

"Protective measures increase costs, which may lead employers to shut down a plan, or reduce benefits. At the same time, federal policy is committed to the view that pension plans should not make promises they cannot keep," he noted in his book on ERISA, "The Employee Retirement Income Security Act of 1974, A Political History," which was published last summer.

Unquestionably, the financial pressure on the PBGC has been heightened by labor unions that frequently accept improved benefit formulas from financially shaky employers, Mr. Wooten said. They take that risk because even if the cost of those benefits ultimately drives the company into bankruptcy, the PBGC will act as the ultimate guarantor, he added.

In an interview, Mr. Wooten described the mindset of labor unions: "Five years from now, the company could go down the tubes and, in that case, we'll collect more benefits from the PBGC. Or, if things get better, maybe the company will be able to afford the higher benefits."

The gamble did not work for United Airlines' members of the International Association of Machinists and Aerospace Workers, who bargained for a 45% increase in benefits months before the airline filed for bankruptcy in December 2002, Mr. Wooten said. (The PBGC's guarantees for benefit increases phase in over five years. But the PBGC announced plans to shutter this plan before the end of the third year after the benefit increases had been made by UAL, thus covering only part of the increases promised to the mechanics. )

Labor unions from the airline and steel industries teamed up with employers to successfully lobby Congress for a stopgap measure in 2004 that let employers defer more than $80 billion in contributions, further weakening the funding status of already underfunded plans.

Pension experts agree that lawmakers will ultimately have to choose between making pension plans more secure, with the realization that only the fittest survive, or loosening funding rules to let more employers keep their underfunded plans afloat for now.

Either way, executives at some companies will not be pleased.

"It usually turns out that the companies in the worst shape are the ones that have the biggest pension problems," Morgan Stanley's Mr. Berner said. He noted that a big part of the problem is what he called legacy costs from the pension funding "holidays" in the 1980s and 1990s, when pension plans were fully funded and companies did not make additional contributions.

(At that time, because of federal law, companies could not contribute to their pension funds on a tax-favored basis once the value of their pension assets was 150% of the value of the liabilities owed to employees if the plan was immediately terminated. Many companies reached that level in the bull market of the 1990s and halted contributions.)

Today's weakest companies do not have the cash needed to make up the funding they skipped in years past, Mr. Berner said, noting they're also often the ones that made the most extravagant benefit promises to their workers, in lieu of pay.

But under funding laws at the time, companies were not allowed to contribute to plans that were fully funded on an accumulated benefit obligation basis. When stock prices and interest rates declined, these plan sponsors did not have a big enough cushion to maintain their fully funded status.

The split between strong and weak companies in the pension community is likely to widen as a result of reform.

"If Congress were to come out with a clear view that the (PBGC) deficit was not going to be paid by taxpayers but was going to be paid by other corporate pension plan sponsors, you'd probably see a split between strong companies and weak companies," said Michael Peskin, managing director, global capital markets, at Morgan Stanley. "The weak companies will want to keep the funding rules flexible, and the strong will say the weak have got to fund their own plans."

"On the other hand, if taxpayers were going to finance the bailout, strong companies would also want weak funding rules because they'd rather have flexibility," he said.

Messrs. Berner and Peskin are quick to note that there are "lots" of defined benefit plans that are healthy.

"They don't want reform if it's going to mean they have to pay for the sins of others who didn't manage their affairs very well," Mr. Berner said.

The Bush administration is taking an all-or-nothing approach to funding rules to prevent the PBGC's $23.3 billion deficit from getting larger. But many observers say that's a mistake. Instead, they say, the administration could have developed a targeted approach to dealing with the problem industries — steel and airlines — that alone account for about 75% of the insurer's deficit.

In an op-ed article in The Wall Street Journal at the end of last year, former PBGC Executive Director Steven A. Kandarian suggested targeted relief for the airlines that would permit them to freeze their defined benefit pension plans and make up the funding shortfall over a specified number of years, while providing new benefits to employees through defined contribution plans.

"It would be very helpful if the law gave the various stakeholders the flexibility to develop solutions tailored to the unique pension problems facing particular industries," said Harold Ashner, a former PBGC assistant general counsel and a partner with the Washington law firm of Keightley & Ashner LLP, which specializes in PBGC issues.

The Bush administration decided not to provide the PBGC with this flexibility.

"People within the administration might have been uncomfortable with giving the PBGC more authority," said a former administration official who did not wish to be identified.

Rep. Earl Pomeroy, D-N.D., who is a member of the powerful House Ways and Means Committee and is active in pension issues, said that rather than the reform the Bush administration proposed, it could have developed a solution to the PBGC's funding crisis "that solicits the best thinking of employers, employees and others. There's an awful lot of creative thinking to advance the ball."

Mr. Pomeroy pointed to alternatives being developed by the trade associations representing plan sponsors. These alternatives capture the actuarial accuracy of the yield curve for measuring liabilities, as proposed by the Bush administration, but smooth the impact of fluctuations in funding requirements over time, he explained.

Not surprisingly, employers and lobbyists have threatened, time and again, to use the only weapon they have against regulation — withdrawal.

But competitive forces play a large role in why employers offer benefits, including retirement plan coverage. Companies are more inclined to offer such plans if their competitors do, as a way to attract and retain quality employees. In addition, companies that shut down retirement plans also deprive their top management of powerful tax incentives.

Still, the use of the tax subsidy to encourage retirement plans has largely failed because rank-and-file employees do not see a tangible flow of benefits from the tax deferral, which is one reason participants have generally tended to discount the value of the employer-provided pension plan.

Small changes in defined benefit regulations might increase coverage and participation rates and give employers incentives to maintain their pension plans, various experts said.

Suggestions include permitting employers to add and subtract distribution options such as taking pension payouts in a lump sum.

Many employers added a lump-sum option when there were no restrictions about removing it, said Nell Hennessy, president of Fiduciary Counselors Inc., Washington, and a former deputy executive director and chief negotiator at the PBGC during the Clinton administration.

Ronald Gebhardtsbauer, senior pension fellow at the American Academy of Actuaries in Washington, suggested that employers be allowed to replace lump-sum options with a 20-year "certain and life" annuity, which permits participants to leave their benefits to heirs. Participants without heirs would be able to trade in those annuities with one-time payouts from life insurance companies.

Dan McGinn, a retired actuary, said giving small employers a tax exclusion for contributions, instead of a tax deduction, would be a more powerful incentive.

In a roundtable discussion with the Senate Health, Education, Labor and Pensions Committee, Donald E. Fuerst, senior retirement consultant for Mercer Human Resources Consulting, Denver, offered two proposals to shore up the defined benefit industry. One is giving companies that offer both defined benefit and defined contribution plans a tax credit in addition to the tax deduction for contributions. The other is allowing employees to exclude from taxable income a portion of the benefit that's paid from a defined benefit plan as an annuity.

The American Council of Life Insurers, Washington, has offered similar suggestions.

In an interview, Mr. Fuerst said his suggestions for bolstering defined benefit plans stem from his belief that employers' relying too much on personal savings — even through defined contribution plans — is bad for the economy because it hampers their ability to manage the work force and will eventually stifle consumption when retirees find they are running out of money.

Money in defined contribution plans is effectively individual savings because as soon as a contribution is made — whether by an employee or employer — the investment risk and the longevity risk are transferred to the individual, he said.

"We're shifting toward overreliance on one leg of the stool, and that is private savings," Mr. Fuerst said. "Unless substantial changes are made, the overreliance on this one leg will be detrimental to the economy, to the ability of corporate America to manage the work force properly."

Companies that rely exclusively on defined contribution plans have a harder time managing their work force. That's because a company often wants to lay off workers or encourage workers to retire — at the same time 401(k) accounts are declining and people are less likely to want to retire, he said.

"For companies with defined benefit plans, it's much easier to accomplish that transition because people have a stronger guarantee of what their benefit is going to be," Mr. Fuerst said.

"We've changed laws to substantially encourage people to contribute to 401(k) or any type of defined contribution plan, yet there hasn't been any form of legislation that encourages more participation in defined benefit plans," he said. "Nothing."

Mr. Fuerst considers his proposals major fundamental changes and said other ideas that offer bits and pieces of reform are not likely to work over the long haul.

Other pension experts agreed.

"I don't see too many opportunities for tweaks that could help materially," said Andrew Oringer, a partner at Clifford Chance U.S. LLP, a law firm in New York. "This (defined benefit) thing has grown to be quite the program, and I think you're beyond changing a subsection (of ERISA), hoping it has some kind of dramatic effect."

While legislators and regulators struggle to find defined benefit reform that is palatable to most, if not all, constituencies, many economists, consultants and other pension experts believe the fundamental crisis is about pension accounting.

The solutions they propose deal with getting the funding right from an economist's point of view.

"Number one, we want clarity as to what the right cost of pensions is," said Morgan Stanley's Mr. Peskin "Without that transparency, we're not going to get the right action."

He said such clarity would lead to better asset-liability management, a key step in putting the defined benefit industry on firm ground.

In fact, the Bush administration's pension reform proposal, issued in January, includes a provision for increased disclosure of plans' funding status and funding trends. It would also replace multiple pension liability measures with one measure; alter the premium structure by which companies pay into the PBGC; and allow companies to make additional tax-deductible pension contributions during good economic times.

Ronald J. Ryan, chief executive officer of Ryan ALM Inc., New York, has long argued that plan sponsors need to manage their assets in relation to their liabilities.

In addition, he said, the current practice of smoothing assets over five years probably overstates true pension assets by 20% to 29%, while current accounting rules leave liabilities 30% to 45% undervalued.

"In 1999, most plans had a surplus, and I was begging them to match assets to liabilities and move the surplus to a separate, distinct fund with a separate, distinct asset allocation," Mr. Ryan said. "But their motivation was to support earnings.

"There's nothing wrong with that. That's why a stockholder buys (the company's stock)."

But accounting rules — specifically the smoothing rules — "confuse everybody about what the true economics" of the pension plan are, Mr. Ryan said.

His suggestion is to define target areas of the liabilities — short, intermediate, long and very long — and manage assets in a similar fashion. Mr. Ryan has created a custom liability index that divides liabilities into such time frames.

"Until that's in place, the asset side cannot function properly," he said. The custom liability index provides the yield of liabilities for each time frame.

For example, he said, if the 10- to 15-year time frame had a 4.5% yield, that would mean the liabilities in that area would grow of 4.5%, on average, to their maturity date.

"I now have an absolute return benchmark," he said. "The asset manager now knows the return they need to match or beat the objective. If I told you to beat 4.5%, you'd behave differently than if I told you to beat the S&P 500."

This idea of managing a pension fund's assets with an eye toward the liabilities is gaining traction.

Rob Blackwell, managing director of research and strategy at Russell Investment Group, Tacoma, Wash., said a combination of the bear market in stocks, low interest rates and low investment return expectations led many plan sponsors to look more seriously at the liability side of their ledgers.

"They started saying ‘maybe we need to think differently strategically … maybe we should be taking the idea of starting with liabilities even more seriously than we had always done,'" Mr. Blackwell explained. "We've become more and more convinced that that is at least the right conceptual starting point."

At its annual client conference in mid-March, Russell unveiled a new product called the Russell Pension Navigator, a web-based system that allows corporate defined benefit pension plan executives to analyze the relationship between assets and liabilities. The tool also can forecast the impact of capital markets and corporate policy on a plan.

"It's designed to be an additional piece in the overall asset allocation/asset-liability service set," Mr. Blackwell said. "It will help plan sponsors answer ‘what if' questions more quickly. We expect it will lighten our workload and put us in a position to spend more time with clients, exploring what different kinds of strategies they should consider."

For the last six months or so, the consulting unit at Wilshire Associates Inc., Santa Monica, Calif., has been implementing a new optimization model with public and private plan sponsors, based on the idea that the core business of any pension plan is to pay benefits. The model establishes a relationship between the risk of not having enough assets to pay benefits and the long-term cost of running a plan.

"The resulting cost-risk efficient frontier presents the budgetary consequences of risk-taking in the most straightforward way," said Dimitry Mindlin, a managing director at Wilshire. "The efficient policies produced by Wilshire's methodology maximize the safety of pension benefits at a given level of cost and, at the same time, minimize the sponsor's cost at a given level of risk added."

He said the policies thus represent the best interests of both the plan participants and the sponsors.

In the public sector, retirement programs are dominated by defined benefit plans that are not necessarily in good shape.

According to Wilshire's annual state retirement system funding report, the combined assets of the 64 state plans that submitted 2004 data to the firm increased 14%, to $778.9 billion, while liabilities rose 5%, to $942.3 billion (Pensions & Investments, March 21).

But, as in the case with private defined benefit plans, accounting rules used by plan sponsors understate liabilities and costs, according to some pension experts. Some estimate that public pension plans are underfunded by a combined $1.2 trillion.

"The dead moose in the center of the table that no one is talking about — that is state defined benefit plans," said Andrew Turner, managing director of institutional investor services at Russell.

"Already many of them are underfunded, and if you had to go back and revalue them on a real economic basis, my bet is people would be stunned by the size of the promise that's been made to state employees."

The Wilshire report said 54 of the 64 funds that reported values on or after June 30, 2004, are underfunded based on the market value of their assets. Those plans are underfunded by an average of 23%.

In addition, the unfunded actuarially accrued liability, or UAAL, has increased over the last five years, which indicates "deteriorating financial health for most state retirement systems," the report said. But actuarial values are often smoothed, like in the private sector, to reduce volatility when calculating plan sponsor contributions.

"If the UAAL were calculated using market value of assets, the positive market return over the last year would have led to a decline in the UAAL … indicating improved financial health for most state retirement systems," the report said.

Mr. Ryan from Ryan ALM said smoothing is undervaluing the pension plans' assets, but the discount rate they use to price liabilities is too high, creating a lethal combination.

"The discount rate is somewhere in the neighborhood of 300 basis points too high above the market," he said. "So liabilities are undervalued by 30% to 45%."

He added that many plan sponsors, believing they are close to or completely fully funded, are raising benefits.

Several states and municipalities have issued pension obligation bonds in an attempt to return to a fully — or nearly fully — funded status, a move Mr. Ryan said makes sense because it buys the plan sponsor time to solve its funding problem.

"The potential for success here (among all funds) is extremely high," he said. "This is the best moment in the last 45 years for the asset side to win."

Mr. Turner said the defined benefit problems — and the debate over pension reform — are not exclusive to the United States.

"This is worldwide," he said, pointing out that in the Netherlands, "there's a lot of squealing going on" about new requirements that pension plans be fully funded.

In January, a new international accounting standard, IAS 19, took effect. Combined with a new supervisory framework due next January, it requires European companies to value liabilities at market rates, not unlike what some experts have proposed for U.S. plan sponsors.

"The most significant requirement internationally has been the Dutch requirement of full funding," Mr. Turner added. "They've said, ‘We've made these promises to workers and we're going to keep them, and the only way to do that is to ensure full funding.'"

He said the United Kingdom is also "a little bit out front" of other countries in tackling its pension troubles by recognizing the impact of marking to market pension liabilities.

In the area of Social Security, several countries have launched private account programs not unlike the accounts President Bush has proposed. The most notable is Chile, but others include Sweden, Argentina, Bolivia, Singapore and the United Kingdom.

The Bush administration has proposed defined benefit plan reform. Regulators are looking at accounting and other policies. Pension consultants and other experts are creating new ways to solve the funding problems. Still, many experts agree that the demographic trends, combined with the economics, will continue to drive plan sponsors out of this market.

In fact, a report released last week by Hewitt Associates LLC, Lincolnshire, Ill., found more than one-third of companies will either freeze their defined benefit plans to new employees or freeze them to all employees in the next two years.

Of the 200 financial executives at large companies that Hewitt surveyed, 16% said their companies will close their defined benefit plans to new employees; 15% said they will freeze their DB plans for all employees and give them defined contribution plans instead; and 5% said they will freeze their DB plans to all employees and not provide any additional retirement plan.

In addition, 65% said their top priority for the next two years will be to find ways to control growth in retirement costs. Almost half said they would focus on reducing pension cost volatility.

"The birth of a new DB plan is cause for public rejoicing," said Don Ezra, the New York-based director of strategic advice at Russell.

"One of the big uncertainties is whether or not — and what kind of — coordinated governmental response there's likely to be," he said, citing the range of players and proposals, including the Department of Labor and the PBGC, and reform proposals such as marking to market assets and moving to risk-based premiums.

"Whether or not those will be coordinated in a way so as not to completely push people out of the defined benefit world is the question," Mr. Ezra said. "A lot of economists believe there's a social welfare niche solved by defined benefit plans, and that they have a particular role in the labor market. We think that's true; but if the governmental response isn't coordinated, a lot of (corporate) treasurers are going to say, ‘This is too risky, I'm going to go a cash balance plan because I can't stand the risk.'"

Stephen P. Utkus, a principal at the Vanguard Center for Retirement Research in Malvern, Pa., said defined benefit plans are great, but under specific circumstances.

"If you stay at the same employer for 30 years, if the employer continues to make money and can afford the promises it's made … if there isn't a baby boom that makes it expensive and if there isn't global competition, then defined benefit is a great program," he explained.

"They will always remain an important element for firms that have a very long-tenured work force, whether that's oil or autos, but to be honest, that's a smaller and smaller part of the U.S. labor market," he said. And Mr. Utkus noted that even in those industries, the perception of the work force is changing.

Mr. Oringer, the Clifford Chance attorney who specializes in ERISA, said many employers that have not gotten out of defined benefit programs have tried to convert to cash balance plans, which have some characteristics of defined contribution plans. But even that "has become a dangerous and complex process," largely because of the legal uncertainty that remains about hybrid plans.

"You're left with a situation where employers are still pushing the trend toward DC plans, notwithstanding the fact that there's a broad recognition that, in a perfect world, there would be a role for a guaranteed benefit," Mr. Oringer said.

Indeed, Mr. Ezra said his biggest fear is that if pension reform is not handled with skill and coordination, defined benefit plans might become extinct outside of the public sector.

"If that happens, more and more individuals will have all the risk on themselves, and the social benefits of risk sharing through defined benefit plans might be lost," he said. "Then it might take a generation of that to go by before we realize it again."

While experts bemoan the state of the defined benefit pension system, defined contribution plans are growing — but not without problems and challenges of their own.

According to government figures, only 47% of private companies offer defined contribution plans, and employee participation by workers age 16 or older is in the 45% to 50% range, a level that has held steady for 25 years. Of workers 25 through 64 years old, participation is between 60% and 65%.

Deferral rates for employee contribution into these plans is about 6% of pay, another figure that has remained fairly steady over time.

Mr. Utkus said 10% is typically the contribution experts agree is needed for an employee to ensure a decent balance for retirement, though some actuaries put it as high as 13%.

"Right now, if you look at 401(k) data, the median participant is saving 6% and the typical employer is matching 3%, so you're at nine," he said. "But 20% of employers don't offer matches and half of the population is below that number."

According to data gathered by the Employee Benefit Research Institute, Washington, the average 60-year-old 401(k) participant who has held the same job for 30 years is only $162,042. That amount would return $16,852 a year if invested in a portfolio of stocks that earned the historic stock market return of 10.4%.

Invested in a fixed joint annuity, however, it would provide only $10,500 a year, not enough even with Social Security for a princely life style.

Mr. Utkus added, however, that some workers who are not in 401(k) or defined contribution plans do participate in a defined benefit plan or a profit-sharing plan, "so it's not like everyone who's out is going to be living tightly on Social Security."

Pension experts agree that removing loopholes in tax law and simplifying other rules could accelerate DC coverage and participation rates.

David L. Wray, president of the Profit Sharing/401(k) Council of America, Chicago, said the amount of money in defined contribution plans and individual retirement accounts rose 2.5 times from 1993 to 2003, reaching more than $5 trillion.

"My projection is that by 2013, we'll probably have $15 trillion" in those two types of plans, he said, noting that his organization views defined contribution plans and IRAs as companions.

Defined contribution plans — and 401(k) plans specifically — are healthy when measured by participation as well as by asset growth.

Participation in private-sector defined benefit plans has been declining about 2% a year since 1985, whereas defined contribution plan participation has been rising 3% a year over that time period. Meanwhile, 401(k) plan participation has grown 9% a year.

Still, there are challenges.

"The challenge is we have eligibles (eligible employees) that don't participate and companies that offer no retirement plan whatsoever," Mr. Wray said. "Getting eligibles in and getting companies that don't have a plan — that's our call to arms."

Among retirement experts, the most talked about incentive for addressing increasing enrollment is encouraging employers to automatically enroll all new workers in the company-sponsored retirement plan, counting on inertia by workers to stay in a plan once covered.

Legislation recently introduced by Rep. Rahm Emanuel, D-Ill., should remove the impediments toward automatic enrollment, said J. Mark Iwry, a benefits tax counsel at the Treasury Department in the second Clinton administration and now a non-resident senior fellow at The Brookings Institution, Washington. The bill, the 401(k) Automatic Enrollment Act, would confirm that automatic enrollment doesn't violate state laws barring employers from deducting money from employees' paychecks without their signature.

Mr. Emanuel's bill also clarifies that employers can make the default option something other than traditionally low-risk, low-return options such as money market accounts without fear of being sued for losses, so long as the investments are made in a "prudent manner" such as in diversified, well-balanced lifestyle funds. The bill would also make it easier for employees to opt out of the automatic enrollment after being in the plan for a short time.

Mr. Iwry said he received a formal pronouncement by the Internal Revenue Service in March 2004 that should also promote this practice. He said it clarifies that when a plan switches to automatic enrollment, employers may defer up to the same percentage of an employee's pay the plan permitted before the plan had automatic enrollment.

And, for the first time, the IRS letter clarifies that employers may escalate on a regular basis, or at the time of a raise, the amount of employee contributions they automatically defer.

Mr. Iwry also suggested that a better retirement policy would offer protection for employers under the non-discrimination test as long as a certain percentage of employees were enrolled in the plan. Under the current rule, employers have to make the offer of a matching contribution.

Behavioral economist Richard H. Thaler, a professor at the Graduate School of Business at the University of Chicago, said he agrees with this idea, but he would take it one step further.

He suggests complete waivers from non-discrimination testing for companies that offer automatic enrollment, automatic increases in contributions in line with wage increases and "sensible" default investment strategies such as a low-cost blend of stocks and bonds.

"What I would like to see is any company that offers that package gets a waiver from the non-discrimination rules," he said.

One of the most prevalent trends that continues to gain steam in bolstering defined contribution plan participation is making them look more like defined benefit plans.

But there are risks with this approach, as a case in Canada showed.

Mr. Ezra at Russell explained that a university pension plan in Canada switched to a defined contribution program but with a floor, or defined benefit-type, guarantee if investment returns were poor.

"The first question was, at what level do you place the floor?" he explained. "Second, it didn't take long before the university people covered by the plan realized they had some downside risk underwritten, so they took high equity exposures.

"It isn't easy to mix these kinds of things," he said. "The details will dictate whether they're really defined benefit or defined contribution."

Pension experts agreed that reforming the private pension system is more critical to Americans' retirement security than reforming the Social Security program, but because President Bush has made Social Security reform a priority, and because he has the bully pulpit, he has put it on the nation's front burner.

Raising the issue in a public context is valuable, experts said, even if they don't all agree with the strategy of allowing individuals to use part of their Social Security tax for individual accounts.

"The pluses of the Bush Social Security plan are that he at least put the issue on the table," said Peter G. Peterson, chairman of the Blackstone Group, New York, in an e-mail response to questions. Mr. Peterson wrote "Running on Empty: How the Democratic and Republican Parties Are Bankrupting Our Future and What Americans Can Do About It," which was published last year.

He said while private accounts have "much to say for them" if done properly — which, he said, is "a very big if" — they do not address Social Security's ills.

"The Bush administration talks about the $11 trillion unfunded liability," Mr. Peterson said. "The private retirement accounts do virtually nothing to address that fundamental solvency problem. That will take some combination of benefit cuts and possibly revenue increases."

Robert J. Shiller, economics professor at the Cowles Foundation for Research in Economics at the School of Management at Yale University, New Haven, Conn., said the Bush plan has positive and negative aspects, and the negatives outweigh the positives.

"He wants people to become more involved in the capitalist economy, and the effect might be people will learn more about it and maybe be more productive," he said. "People who own property tend to be more invested in their community and their government. But the other side is that when you give people ownership, you're giving them risks, too."

In a research paper released in late March, Mr. Shiller found that, based on simulations he ran, personal accounts would lose money 71% of the time and yield a median return of 2.6%.

Mr. Shiller simulated the future returns of the personal accounts using historical U.S. stock, bond and money market data from 1871-2004. He used a hypothetical worker born in 1990, who starts participating in the accounts in 2011, retires in 2055 at age 65, and contributes 4% of earnings into the personal account. He assumed the money was invested in lifecycle funds.

But Martin S. Feldstein, an economics professor at Harvard University, Cambridge, Mass., and president of the National Bureau of Economic Research Inc., also in Cambridge, said research has found that the mean real return of a portfolio with 60% stocks and 40% bonds was 6.9% for the period 1945 through 2003.

In an NBER working paper released in February and titled "Structural Reform of Social Security," Mr. Feldstein offered suggestions for reducing the investment risk of personal savings accounts, such as including long-term government inflation-indexed bonds in the portfolio.

"For example, a personal retirement account manager (an insurance firm, bank or mutual fund) could guarantee a 30-year-old that the funds that he saved that year would at least retain their real value to age 67 by investing about 60% of the funds in such bonds and the rest in equities," he wrote. "The expected performance of that investment (based on past experience) would be a real rate of return of 4.7%, substantially higher than the implicit yield on Social Security."

In addition, he said simulations using that type of investment at every age, in a mixed plan that is part private savings account and part Social Security, show that the resulting "annuities" from the plan would provide benefits "substantially" greater than the benchmark Social Security benefits of today. Also, the probability that the annuity would be less than 90% of the benchmark benefit is less than 1%.

As the debate over Social Security rages, economists and others have focused on ways to boost the anemic savings rate in the United States.

"What this country desperately needs now is more net national savings and, as part of that, more personal savings," Blackstone's Mr. Peterson said, adding that the Bush administration's Social Security reform proposal "does nothing to address those fundamental problems."

Mr. Peterson's two-step plan for solving the Social Security funding problem includes a savings component: a mandatory retirement account of 2% to 3% of pay, with some subsidies for the poor.

"This would be a totally funded plan, which increases savings, which we very much need," he said. "I would put this money in index funds composed of global fixed-income and equity instruments. The administrative costs would be far lower here than if we turned over these accounts to individual brokers."

Peter A. Diamond, an economics professor at the Massachusetts Institute of Technology, Cambridge, agreed that the focus must be not only on personal savings, but also on national savings, a topic on which he was blunt. "The surest way to increase national savings would be to repeal the tax cuts and have the government borrow that much less. You can manipulate incentives to private savings, but there's nothing that has the level of impact and the level of assurance as the government doing it right," he said.

Beyond reform of the retirement system in the United States, a small but growing number of experts are suggesting the whole notion of retirement must be reformed.

Boston College's Ms. Munnell is among that group, which believes America's work force will have to face the inevitable: Instead of working until age 62 and collecting early retirement benefits, or retiring at the more traditional age 65, people are going to have to work longer and effectively redefine the period of their lives after they stop working.

To characterize this new definition of retirement, demographers and retirement experts have coined a new phrase, the "old old," to describe those people in their 80s and older.

"Anyone calls me elderly when I turn 65 and I'm going to punch him in the nose," said Zvi Bodie, a professor of finance at Boston University's school of management, who is 62 and doesn't plan to retire for at least two more decades.

Mr. Bodie anticipates a return to the late-19th century concept of full retirement — a short span of time at the end of life. That span, however, is more likely to be when people are in their 80s or 90s, rather than in their 50s or 60s, and it will cost money.

"Financing the baby boom retirement is going to be expensive throughout the system — Social Security, defined benefit, defined contribution, personal savings," said the Vanguard Center's Mr. Utkus.

"That means people need to put more resources into retirement or work longer. I'm betting on a considerable number of people working longer as part of the solution," he said. "Work earnings (during retirement) are going to be much more significant than they were for prior generations."

A survey by Merrill Lynch & Co. Inc., released earlier this year, found that baby boomers are already moving in that direction.

Fifty-eight percent of middle-aged Americans surveyed expect to start their own business, work full time or alternate between work and leisure after their traditional career ends, according to the survey.

"They want to stay engaged and they want to keep working," but on their own terms, said Cynthia Hayes, first vice president at Merrill in Plainsboro, N.J.

The latest participation data for older workers from the Bureau of Labor Statistics reinforces this notion.

Today, 68.9% of men aged 55 to 64 are still working, up from 65.5% in 1994. Among those 65 or older, 18.8% are still working, compared with 15.6% in 1993.

The statistics for women are more dramatic: 57% of the 55-64 age group are still working today, up from 41.3% in 1980. Among women 65 or older, 11.6% are still in the work force, up from 8.1% in 1980.

The trend toward working longer is a welcome one, Ms. Munnell said. She noted that older workers in the labor force might be able to stash away additional money for their retirement, collect higher Social Security benefits when they do retire, and draw down their savings for a shorter period of time.

Delaying retirement also helps the finances of the Social Security program. For each year workers delay retirement, the system accrues an additional 2 to 2.5 percentage points in revenue, without corresponding increases in benefit payments, Watson Wyatt's Mr. Schieber said.

Employers, anticipating a labor shortage as the bulge in the population heads toward retirement, are also devising programs to appeal to older workers who might not be ready to fully retire and want to earn some additional money.

"They're talking about changing the work paradigm," Merrill Lynch's Ms. Hayes said.

For example, Indianapolis-based YourEncore, the brainchild of executives at Procter & Gamble Co., Cincinnati, and Eli Lilly & Co., Indianapolis, allows companies to hire, on a project basis, from a pool of retired engineers and scientists.

YourEncore, which launched in October 2003, expects to add three large companies as clients over the next few months, according to Brad Lawson, YourEncore president and chief executive officer. Participating companies include Boeing Co., Chicago, and National Starch & Chemical Co., Bridgewater, N.J.

"People come to do this not necessarily because of the money … but what really drives individuals is that they want to stay connected and remain intellectually challenged and continue to contribute," Mr. Lawson said.

Using a different model, some firms — like Watson Wyatt — don't have a mandatory retirement age.

Regulators, too, are recognizing the changing definition of work and retirement and are incorporating flexibility into existing pension rules to accommodate "phased retirement" arrangements.

Last fall, the IRS proposed regulations that would permit older employees to begin collecting retirement benefits proportionate to the reduced workload they carry.

Even as employees and employers redefine retirement, another problem looms: so-called leakage — retirees outliving their retirement earnings, either through mismanagement of their retirement assets or investment losses.

Academics like Olivia S. Mitchell, director of the Pension Research Council at the University of Pennsylvania's Wharton School, point to policies in other countries that can prevent leakage.

In Great Britain, for example, monthly pension payments are mandatory for people aged 75 and older, said Ms. Mitchell.

"Mixed models (part annuities, part lump sum) are going to become increasingly popular," she predicted.

Some corporations, such as International Paper Co. in Stamford, Conn., have created innovative programs to protect workers against spending all their retirement benefits at once.

In January 2004, IP executives set up a program to encourage retired workers to keep their savings in the company's low-cost 401(k) plan and withdraw the money in equal installments over 20 years. In addition, retired workers can borrow as much as 50% — up to $50,000 — from their 401(k) accounts and replace their borrowings over as much as 10 years.

"It's one small effort we made to make our savings plan more attractive," said Robert M. Hunkeler, vice president of investments at International Paper. As a result, two-thirds of retired IP workers surveyed by the company last fall said they preferred keeping their money in the $4.1 billion plan instead of rolling it over into individual retirement accounts.

Other efforts to protect retirees from outliving their savings are being floated by financial service companies.

These include:

• allowing people with hundreds of thousands of dollars locked up in home equity to get a reverse mortgage, enabling them to take up to a specified percentage of the value of the home as an annuity over a period of time;

• increasing the punitive tax on early withdrawals from retirement plans and IRAs to 15% from 10%;

• making it easy for employers that offer only a retirement plan to let participants buy a group annuity; and

• reducing the tax rate on annuities from pension plans to the capital gains rate, currently 15%, instead of the higher income tax rates.

Boston University's Mr. Bodie, who is also managing director of New York-based Integrated Finance Ltd., is working on a program that would allow workers to purchase inflation-indexed annuities with their savings in a defined contribution plan. The annuities would be purchased through his company, which would, in turn, act as conduit with insurance companies.

"IFL will pay out a product, but the stream of income will ultimately be paid out by the insurance company. For the consumer it will be seamless," Mr. Bodie said.

Whatever reform is enacted, either on corporate pensions or Social Security, it will likely have some ripple effect in the country's capital markets, most notably stocks and bonds.

In fact, some impact has already been felt.

After the Bush administration released its pension fund reform proposal, Treasury prices rose amid speculation that the proposal would require companies to sell stocks and buy bonds to better match their liabilities.

No similar reaction was seen in the stock market, but some economists expect that with or without reform, pension executives likely will lower their plans' equity allocations and raise their bond allocations.

Morgan Stanley's Mr. Berner said his firm has counseled corporate financial executives to do just that.

"Both of those would reduce demand for equities and increase demand for fixed-income securities at the longer end of the yield curve," he said.

"We believe the impact of that — especially if state and local governments were to follow — would be a temporary but appreciable impact on stock and bond prices."

Mr. Berner and his colleague Trevor Harris, managing director, have estimated that pension funds reallocating their assets could move $650 billion from stocks to bonds.

That $650 billion would come from equities globally, but Mr. Berner said the bulk of that would be from U.S. equities because U.S. pension plans typically have more exposure to U.S. stocks.

He added the market impact of such a move was difficult to quantify "because pension funds haven't ever so massively reallocated portfolios."

Others are not so sure that so much money will shift to bonds from stocks.

"Certainly U.S. plans are looking at ways to address (asset) volatility, whether by going to long bonds or using other derivative types of vehicles to hedge their equity risk," said Gordon Latter, vice president, pension and endowment strategy at Merrill Lynch Global Securities Research & Economics Group, New York. "But a lot of them feel that the price of bonds is expensive, and (moving to bonds) is actually going to raise the long-term cost of the program. That causes me to ponder whether we'll see those types of flows" to bonds from equities.

Mr. Ezra, Russell's director of strategic advice, said because capital markets are viewed globally and few of them exist in isolation, the impact of pension funds selling stocks and buying bonds could be muted.

"As the relative pricing of equities and bonds changes, that may make them look more attractive to other people" both in the United States and in other countries, he said. "We don't know the order of magnitude of these things, but it's not a closed system."

The continued shift to defined contribution plans from defined benefit plans is not likely to have a big market impact, according to Roger G. Ibbotson, a finance professor at Yale and chairman of Chicago-based Ibbotson Associates.

It isn't clear that the shift affects how much is invested — it's just a change in how the assets are managed, he said. "I'm not sure it increases the total amount" or influences the stock or bond market.

Social Security reform as currently proposed by President Bush, however, could have an impact, Mr. Ibbotson added.

"If the Bush plan goes through and we have independent accounts, that would definitely shift more money to the stock market, because now that money has been used to buy back government debt," he explained. "In some sense, I think that's almost like buying on margin on aggregate.

"I think of it as a leveraged position, and it's more risky for the whole country to leverage ourselves to the stock market," he said. "In the short run, though, it will have a positive flow into the stock market."

Clearly, the driving forces behind the retirement crisis — an aging population, underfunded pension plans, low individual savings rates and a steadily weakening Social Security program — are not likely to change in time to avert the crisis. But employers, legislators and retirement experts are working to at least minimize the damage.

Movement on several fronts, such as improving the rules governing funding and accounting of defined benefit plans, implementing strategies to increase defined contribution participation among both employees and employers, and preventing politics from getting in the way of real Social Security reform, could go a long way toward ensuring baby boomers do not find themselves destitute in retirement.

"Pensions aren't all that old," University of Chicago's Mr. Thaler said. "Defined contribution plans, in fact, are pretty new.

"I think we're figuring out how to do this."