The National Conference on Public Employee Retirement Systems announced a new proposal with much fanfare last week. Their "Secure Choice" initiative is ostensibly a well-intended effort to "spread the wealth" and assure "retirement security" for private-sector workers at the risk and potential expense of state taxpayers. The concept is simple: Let state pension funds or a sister organization offer a pension benefit to private employers and their workers, with the employers and ultimately taxpayers underwriting those risks, just like they do for public pensions.

Employees are guaranteed a minimum rate of return, despite the risks to the plan; eventually they collect all the upside as well. In short, heads the employees win; tails the employers and taxpayers lose. Of course, the proponents don't come out and say that taxpayers will be on the hook if markets chronically underperform as they have in recent years. Instead, they dance around that issue and present a few wishful ideas to mitigate but not eliminate market risks. Their plan relies on actuarial techniques that include a 20 year "open amortization" (similar to credit card amortization) to shift the costs of major investment shortfalls to participating employers — and ultimately to tomorrow's employees whose salaries will suffer just as public employees are now.

In a monopoly-mandatory public pension plan, that kind of intergenerational billing structure can survive, but it will collapse in a voluntary system in the private sector. Why would private employers pay more than the current service cost when they could buy new workers a better benefit outside the plan? What stops them from freezing or exiting the plan and leaving the deficiencies unfunded at taxpayer expense? Unless the plan expressly resolves these issues, this structure will be unmarketable. Even worse, it will fail every time markets plunge and stagnate as much as they have in recent years.

Audacious doubling-down. One has to admire the chutzpah of an organization at the forefront of combined pension deficits exceeding $700 billion nationwide, for suggesting that their underwater systems should now be cloned and exported to the private sector. Never mind that skyrocketing costs of underfunded pension funds are causing many public employers to freeze salaries, lay off workers and cut public services. Let's just "double-down" and lure private employers and taxpayers into similar predicaments in the future!

I confess that in a 2007 column, I floated a similar idea. That was before the devastating impact of the Great Recession showed us all vividly that it's a doomed strategy. Back then, when pension plans had funding ratios of 85 percent heading upward, it seemed like a good idea to offer taxpayers the right to participate in a similar program. To my readers' credit, your feedback back then was universally negative — as was the reaction of experienced and savvy professional colleagues in the pension consulting business. I learned a lot from that blowback. Apparently the NCPERS board has not likewise learned much from its members' investment experience in the past decade. That said, I would support four of the worthwhile concepts in their white paper: improved retirement security, sharing mortality risk, economies of scale and professional management of collective investment trusts.

While the core product in this Secure Choice pitch-book is dead on arrival, it could be salvaged if the promoters focus instead on less ambitious goals that exploit its four advantages without further burdening taxpayers with investment risks that are priced incorrectly for the benefit of plan participants. The public will not tolerate another scheme to game the taxpayers by making guarantees at their expense. As proposed, this is just not a business for state governments to enter. Worse yet, it could backfire on the public pension community and create a backlash against any and all forms of socialized risk in the retirement space.

The "free lunch" mentality. An underlying goal of Secure Choice seems to be to broaden public support for public pensions by "spreading the wealth" to those who buy into this type of program. The unavoidable problem with the idea of selling a public-sector pension to private workers is that pension trustees cannot guarantee the returns of their investments unless they invest solely in risk-free securities. But risk-free securities won't come anywhere close to achieving the discount rates that public pension funds use in structuring their plans and the unrealistic targets of this proposal. To achieve the "virtual" and minimum benefit crediting rates of the Secure Choice plan, the state would have to permit investments in risky securities and then hope they will never earn less than 3 percent compounded plus expenses, let alone ever lose value. Of course, we all know how well that worked in the past decade. To paraphrase a loquacious modern politician, "How's that hopesy-assumptioney thing workin' for ya now?"

The business concept here is that a public-sponsored pension plan would compete with private insurance companies that offer annuities by arbitraging its higher-risk investment authority and a unique unspoken franchise right to bill the taxpayers for any market and underwriting losses left behind by private employers.

A litmus test. There is a simple litmus test for this scheme. We need only back-test what would have happened if this proposal were implemented at the beginning of this century in January 2000. That is not a difficult exercise: A rookie actuary or even a second-year MBA/MPA quant grad student can perform that analysis.

The quick answer is that a decade of paid-out benefits, minimum credited interest and other accrued liabilities would have accumulated from "assumed" and "credited" and "virtual" earnings that never materialized in the past decade. The bill for these wishful promises would then be delivered at the doorstep of employers. When the employers walk away — as the plan document admits they can do — the deficits would pass on to remaining employers and not far behind them, the taxpayers. It's totally naïve to expect private employers in a voluntary system to grin and bear the bills they receive to pay down pension deficits — the way public employers in mandatory, monopoly state and local pension systems have had to do. Corporate CFOs will look at this arrangement the same way they view the federal pension benefit guarantee corporation — they will run fast in the opposite direction.

Industry and employer opposition. The insurance companies have strong lobbyists in state capitals where they fund political campaigns, and it's inconceivable that they will stand quietly on the sidelines while state legislatures consider laws that cut into one of their target annuity markets. I don't mind giving them some healthy public-sector competition, but not with a scheme that burdens jobs-creating businesses and jeopardizes taxpayers. The mutual funds now selling 401(k) plans won't be much happier and the Investment Company Institute will dispatch its lobbyists to the state capitols pronto. The state taxpayer associations will oppose this scheme with a vengeance, and I fear that the backlash could be ballot initiatives to ban defined benefits plans of any kind if they have the potential to create a taxpayer-funded liability. The Chamber of Commerce will go ballistic, as I'll explain below.

The sponsors of this idea apparently have given little thought to marketing. Even if they succeed in slipping this proposal through a legislature, they will quickly learn how it feels to be boycotted or blackballed as few businesses active in the chamber will join this plan for public policy reasons explained below. Private employers simply don't want any plan that burdens their owners with unfunded liabilities each time we hit a recession. Nobody who has survived long in the business world is dumb enough to encumber strategic capital that can be deployed more efficiently at the bottom of a recession to buy capacity at bargain prices, at the very time this plan would burden employers with added pension costs and unfunded liabilities. Obviously none of the proponents has ever owned or run a competitive business, or they'd know this idea is a non-starter strategically. There is absolutely no incentive in this proposal for businesses to take on this burden.

In the spirit of constructive criticism, there are two variations on this theme that I would suggest to the proposal's sponsors and legislative advocates. One is the collective defined contribution trust, which would self-insure the risks of stock market fluctuations, investment performance, mortality and other actuarial assumptions across all participants. The other is a pension-exchange feature I suggested last year in response to the Obama administration's request for suggestions to improve income security for retirees.

Collective defined-contribution trusts. Like a traditional pension plan, a collective defined contribution trust takes advantage of the benefits of risk pooling and economies of scale through a common trust fund. Everybody shares in the risks and returns on a pro-rata basis. Investments are professionally managed and can be overseen by a board similar to a pension fund. Retirees never outlive their money because they receive a lifetime benefit based on actuarial calculations. But this benefit is not fixed or guaranteed in amount — only for its lifetime duration. If you live to 99, your annual payment then could be 70 percent or 250 percent of your payment when you were 67, with or without an inflation adjustment, depending on investment results in the common pool. But most importantly you will get a payment in any event, as long as you live — provided the trustees and actuaries perform their jobs properly. Longevity risk is eliminated, and that's one of the most important retirement security issues of our time. This structure clearly answers the "indigent elderly welfare" justification without burdening either employers or taxpayers.

Unlike a traditional defined benefit pension plan, a collective defined-contribution (CDC) trust eliminates risks to employers and taxpayers. Plan participants share in the underwriting and investment risks, not the sponsors. Still, that leaves them far better off than their current IRA and 401(k) plans where they can outlive their money and speculate with their nest eggs. These CDC systems resemble certain union-sponsored Taft-Hartley plans that regularly adjust benefits actuarially. Under this arrangement, individuals could contribute annually and receive a credited earnings rate shared by all other participants based on the actual investment returns of the pooled investment fund — not some "virtual" number concocted by actuaries. Upon retirement, their accumulation can be converted into a lifetime pension. While receiving benefits, market fluctuations would be reflected by changes in the annual benefit payment. There are many variations on this theme, such as an option for near-retirees to annuitize to enhance their pre-retirement planning, but these are the three most essential features: group underwriting of longevity risk; collective shared investment experience, expertise and economies; and no residual risks to employers or taxpayers.

The trustees could also negotiate from strength in numbers with private insurance companies for group fixed-annuity contracts that would enable retirees to obtain the lowest-cost fixed-income guarantees for those who can't bear the volatility risk of market fluctuations. That lays off the investment risk to the insurance companies on terms most favorable to the retirees without burdening employers and taxpayers needlessly.

As good as this all may sound, there will still be one vociferous opponent of any collective structure: the U.S. Chamber of Commerce, which finds pension funds abhorrent because of their influence in corporate governance. Unless investments are limited to index funds or a governance provision requires "mirror voting" of share proxies held by these trusts (so that they are politically neutral in corporate boardrooms), the Chamber would likely oppose the interference of such funds in business operations — especially by inherently political boards of trustees. Safeguards against such meddling would counter that opposition.

Pension-exchanges for IRAs and 401(k) plans. The second variation to consider is one I proposed in early 2010. I dubbed this feature a pension-exchange plan. It shares some of the features of the NCPERS program but limits the employer and taxpayer risks by making limited guarantees to retirees only.

There is no inherent reason that taxpayers should take on the underwriting risks of investments during a private-sector employee's career. Employers certainly have no interest in that function, which has nothing to do with their core business. Further, there is no justification for today's younger workers to subsidize retirees through the implicit subsidy of sharing investment income derived from investment horizons that exceed the retirees' lifetimes. For example, we can't fairly give 80-year-old retirees the income derived from 30-year bonds that will outlive them. Similarly, we shouldn't give retirees risk-free income from risky investments and then burden the youth with all the downside costs. It's bad enough that those younger workers will pay for their elders' Social Security; why should the plan burden them further?

Under a pension-exchange plan, the retirees can elect to trade all or part of their accumulated IRA, 401(k) or similar individual defined contribution account balances for state-operated pensions of equivalent actuarial value. So it's essentially what you might call "Secure Choice Lite." Besides limiting the scope to retirees only, where I part company from NCPERS is the restrictions and protections that must be incorporated in this endeavor. In my prior column I outlined several restrictions that must accompany such a plan, which I have since refined, below:

  • Actuarial assumptions and earnings expectations must be much more conservative than a typical public pension plan — and definitely less than the "standard" 7 percent assumption in the NCPERS proposal. Although this trust fund may be perpetual in its charter, its liabilities run for the average remaining life expectancy of retirees, which would be about 10 years on average. Thus, a secure bond portfolio must dominate the asset allocation and that will force the actuarial discount rate to be lower than NCPERS dreams to be feasible. I originally quoted a number like 6 percent and now find that to be too optimistic in light of recent 2 percent yields on 10-year bonds. (See my companion column on lessons from the Great Depression.) Likewise, mortality assumptions must be conservative and updated frequently so that longevity breakthroughs don't deplete the fund.
  • An individual's maximum pension purchase should be limited to that state's average household income. Nationally, that number is about $50,000. We don't need to socialize risk for the affluent at the expense of lower-income taxpayers.
  • The trust funds must establish a reserve against future market downturns that is permanently off limits to beneficiaries.
  • Plan trustees must have a clearly defined fiduciary obligation to taxpayers as well as plan participants with a specified, actionable obligation to maintain sustainable and prudent actuarial funding.
  • Cost of living allowances should be capped and permitted only when the fund assets exceed actuarial liabilities by a specified surplus level.
  • If the plan experiences an actuarial shortfall of more than 10 percent, benefits must be trimmed to limit the potential cost to employers and protect taxpayers. In this regard, the plan would morph into a collective defined contribution trust with a stop-loss for the taxpayers.

That's a lot of bells and whistles, but failure to include them is a sure path to a financial fiasco in the future. And as explained in my prior column, this exchange concept would have to be approved by federal tax and labor-law authorities, which will happen only if the labor groups supporting their concept are willing to take it to their friends in Washington. I would not rule out that possibility, as the Obama administration is seriously searching for ways to better secure retirees' income.

That said, it will still be a tall order to get any of these ideas through a state legislature, unless the risks to taxpayers are eliminated or carefully contained along the lines I have suggested. Then the advocates and administrators still have to sell their plan in the competitive marketplace. That's not a job for people who are accustomed to monopoly marketing to an audience that includes public managers who often get the same deal on terms favoring them more than the rank-and-file. The idea that private employers will flock to the NCPERS scheme is pure naiveté — in which case any expected economies of scale will collapse under the new bureaucracy's overhead costs. Their board would be wise to confer with a focus group of private employers, especially the gatekeeper CFOs, to market-test their ideas before their members pitch anything to a legislature.

At the end of the day, the collective defined contribution trust emerges as the most cost-effective model with the fewest warts and least risk to employer and taxpayers. From my anecdotal experience, it's the only structure that many private employers might find attractive and viable — especially if the portfolio proxy voting issue is solved or the fund invests predominantly in index funds like the federal employees' thrift plan. If states are ever going to enter this kind of business, that's the model I would support. But I won't hold my breath waiting for lawmakers to buy in, until the scars of the Great Recession have healed, labor markets tighten up, and employers have a compelling business reason to promote retirement benefits in order to attract workers.

The National Conference on Public Employee Retirement Systems defends their Secure Choice proposal.