More Columns
BENEFITS BEAT
Defined-Contribution Report: Mostly Hits, with a Few Misses
A landmark new study of public defined-contribution plans is the best of its kind so far. But there's still room to dig much deeper.
Questions, success stories or anecdotes about benefit issues in government? Girard Miller wants to hear from you.
E-mail him
An insightful, data-rich new research report by the prolific scholars at the Center for Retirement Research at Boston College takes a good look at the experiences of 12 states that have adopted defined-contribution retirement plans. The report is clearly the best work yet on this topic, and it provides some useful information for policymakers and plan administrators alike.
In the spirit of constructive criticism, and in an effort to advance the dialogue in the ongoing debates over defined-benefit vs. defined-contribution plans, I will provide my observations on the "hits and misses" in this report.
The Hits
First, the authors do a great job of dispelling myths and introducing some reality into the rhetoric that often preceded the adoption of defined-contribution retirement plans at the state level. For example, a statistically dominant factor in their adoption was the presence of a Republican in the governor's office. Looking back on the past two decades, it took a strong state leader — one who was willing to present the case for individual responsibility and ownership as a matter of principle — to push legislation through in support of a defined-contribution plan or option. That is not an insignificant observation. In states where Democrats have political control or veto power, odds strongly favor the status quo and retention of a defined-benefit pension regime. Ideology does play a role here. It's not unlike the debate in Washington, D.C., over personal investment accounts in the Social Security system.
Second, the report points out two inherent weaknesses in defined-contribution plans: fees and investment performance. Fees in DC plans are destined to be higher. That's due to the cost of providing record-keeping for each individual's account, plus the facility to transfer participant investments on a daily basis, plus the inherent cost differential in retail mutual funds vs. the ultra-low "favored nations" money management fees paid by public pension funds. Fees are a drag on individual returns and reduce the ultimate pool of assets available to pay retirement benefits.
The researchers also assess the likelihood of inferior investment returns from DC plans in comparison to those earned by DB plans. They report non-original findings that individuals are likely to make worse decisions than pension fund trustees who are staffed by professional money managers. For many years, individuals in DC plans were too conservative and invested predominantly in "safe" but lower-returning funds such as bond, money market and stable-value funds. Then, after the stock market boomed in the 1990s, they jumped on the equity bandwagon just in time for the bubble to burst in 2000. The tortoises in the DB boardrooms lost money in the last recession, but not as much as individual DC plan participants. Some individuals are more prone to buy high and sell low, succumbing to greed and fear.
The report does a very nice job of dispelling the myth that a DC plan will reduce pension costs with respect to unfunded liabilities for promises previously made. Instituting a new plan for new workers and future service does nothing to wipe out the "pension debt" left from the existing DB plan. So there are no dollar savings to be realized there.
The Misses
What the report fails to mention is that the DC industry is making efforts to improve investor education and. more important, has introduced "target date" retirement funds that provide professional asset allocations akin to defined-benefit plans. These target-date funds are not a panacea, as I noted in my previous column on them, but they do provide a reasonable answer to the critics who charge that DC participants are doomed to be whipsawed by the markets. The report also fails to mention that DB plans have an inherent investment advantage in their access to alternative investments, such as hedge funds and private equity, that expands their scope of diversification and opportunity beyond that available to DC plan participants.
The analysis of portability is worth studying. The researchers note that DC plan advocates tout the attractiveness of portable retirement benefits similar to 401(k) plans in the private sector, and note that in reality a small percentage of public employees have actually opted voluntarily into DC plans despite this advantage. Here, I must question whether their database is sufficient to draw that conclusion, or at least the spin that will be given to this finding by DB plan proponents. When given a choice between a "risky" DC plan and a "safe" DB plan, I have no doubt that traditionally risk-averse incumbent public employees will prefer the latter, all other things being equal.
And the report misses the point totally in asserting that in addition to better health insurance, the mere presence of a retirement plan at all is the most compelling advantage of public employment over private sector work for many people. A more probing question is whether the availability of a portable DC plan (at lower cost to employers, if so designed) would be sufficient to attract workers in a competitive labor market, and this report tells us nothing about that.
In their technical work, the researchers sought to regress various explanatory factors against the introduction of a statewide defined-contribution plan. One shaky hypothesis in their model was a questionable assumption that the presence of teachers in a plan as a proxy for unionization should discourage DC plan adoption. Demographics and historical facts don’t bear this assumption out. In Washington state, for example, the teachers’ union supported DC plan legislation because many teachers leave service early and hence don’t fare well under traditional pension systems. In Denver, only 15 percent of the teachers receive a full pension because of their personal mobility, according to a recent study which could easily lead to the conclusion that a DC plan would be preferable for many.
Where the report seems weakest in its analysis is the subject of cost-shifting. Here, I think the researchers miss the mark. They start off well by noting that a major reason DC plans now dominate the private sector is that employers are shedding both costs and investment risks — as well as avoiding regulatory and fiduciary hassles, and costly federal pension benefit guarantee premiums. But the report seems to conclude that public employers could not reduce their retirement plan costs by adopting a DC plan, simply because government employees already pay a relatively higher percentage of their pay into DB plans, which makes them bad candidates for a cost-shift. That's not the point — and certainly not the one DC plan advocates would make. Their issue is whether the public employers' contributions are excessive in light of competitive labor-market conditions, regardless of what employees are paying into the system.
For the record, I would not be one to argue that a lower-cost DC plan will provide better retirement replacement income and equal security. But offering a lower-cost-and-benefit DC option may be the eventual solution to the "pension envy" debates that will likely continue in coming years if the disparity of benefits between public employees and their private sector counterparts appears to widen.
Finally, what is completely missed here is what many government finance officers see as the inherent weaknesses of DB plans and the advantage of DC plans: the "ratchet effect" of pension increases and pension benefits retroactivity in a collective bargaining environment. Under a DC plan, the employer's contribution is made at the time services are provided, and that's it. No mas. If the plan is enhanced at a later time, nobody goes back and adds more money to the employee's account in order to win the union's vote for contract approvals.
But in the DB world, this happens all too often, as noted in my previous column on pension retroactivity. Costs of DC plans are not shifted to future taxpayers as they often have been in pension plans, and that's the real cost-shifting argument. The report sweeps this issue under the carpet with a section on "moral hazard" without any data to address the prevalence of this practice or even a mention of the huge increase in employer contribution rates that occurred in the last decade as benefits were imprudently sweetened in the bubble market years, as noted in my previous column on rising taxpayer costs.
The Bottom Line
Here's why all this is so important: It actually has very little to do with pension plan politics. In light of the recent credit collapse and 2008’s stock-market correction on the heels of the 2002 market meltdown, the Baby Boom generation will now seek the comfort of a safe defined-benefit pension even at their children's expense, and the advocates of DC plans have had the wind knocked out of their sails by roller-coaster stock markets. The public-pension DB-DC battle is over, at least for that generation. But there is a bigger elephant on the table nowadays called OPEB ("other post-retirement benefits," mostly retiree medical benefits). That's where we need to focus our analysis of DB-vs.-DC plan designs.
If there is a place where defined-contribution solutions seem appropriate as part of the hybrid plan designs that will be needed to properly fund retiree benefits, the OPEB space is the leading candidate. Just look at the numbers: The estimated total unfunded liabilities of defined-benefit OPEB plans are somewhere in the range of $1 to $1.5 trillion. Even worse, the nature of the benefits is that the costs go up far faster than general inflation and are not "defined" as they are in a pension plan where there is a ceiling on the future cost stream. Using a defined-benefit plan to fund an undefined, inflation-prone benefit is like using a hammer where an adjustable wrench is needed.
Thus, a defined-contribution feature for new and younger workers would seem to be the most practical solution for many public employers who otherwise could eventually bankrupt themselves if they simply continue along the DB path for OPEB benefits without mitigating those costs, establishing realistic ceilings, and sharing more costs and medical-inflation risks with employees. Especially for a benefit that is far less prevalent in the private sector, and is rapidly being eliminated in many companies. My previous column on OPEB funding provided additional perspective on this point.
The Grade
I'll respectfully give Professor Munnell and her colleagues an A- on this report. Hands-down, it's the best work ever done in this corner of the field. I'm hopeful they will produce an ongoing stream of even more insightful and objective work in the future!
Girard Miller, an analyst of benefits and investments with 30 years of experience in the public, private and nonprofit sectors, can be reached at Girardinmalibu@charter.net. His general market observations and institutional investment strategies are his own and should not be construed as investment advice or recommendations concerning specific securities. More biographical information.