Moreover, close attention must be paid to the consequences of decisions, and corrective action must be taken if those decisions are not serving the investor's interests. Sometimes limited fine-tuning is all that is needed; at other times, no amount of tinkering seems to achieve the desired results, and a complete reassessment of the approach is required.
It appears to me that participant direction of retirement plan assets fits in the latter category.
Aiken goes on to note that even the (overrated, virtually unattainable) rules of section 404(c) "did relatively little to prevent participants from making poor investment selections." As longtime readers know, we too see problems with the conventional model of participant direction (see this and this, for example).
"Because a participant holds a separate account, rather than an interest in a commingled aggregate portfolio," Aiken writes, "the costs of participant-directed plans can be higher than they are for sponsor-directed portfolios." No doubt. But as we'll explain in a moment, the big picture turns on the fact that participants hold separate accounts at the individual level.
Aiken suggests that it might be "time to ask whether the whole concept of participant direction is a losing proposition for investors." That's a timely, important question. Here's Aiken's exploratory answer:
Centrally administered portfolios, like those of foundations, endowments and defined benefit plans, are less complicated. Investment fiduciaries have direct control of the process and are accountable from design to execution. The investment committee or chief investment officer, usually with the support of professional advisers and money managers, makes asset allocation decisions, directs the investments, and can change the portfolio composition as needed.
Plan sponsors and their advisers should consider whether a fundamental change in their approach for 401(k) and other defined contribution plans from participant direction to investment committee or trustee direction of the assets would better serve the best interests of the investors they work for.
Again, all of that sounds perfectly reasonable to us, but for one fact: Centrally administered portfolios would create a disconnect between plan assets (which would be managed at the system level) and participant liabilities (which would remain at the individual level).
Generally, the economic needs (i.e., risk and return objectives) of defined-benefit pension funds are relatively easy to specify, such that central decision-makers can manage assets at the system level (and participants and their beneficiaries can plan the liability side of the ledger with relative certainty as to the future payments they'll receive). The economic needs of individual account-holders, on the other hand, vary widely and unpredictably, such that central control of plan investments could/would create a disconnect between group-level assets and individual-level liabilities.
Put differently, as long as accounts balances remain individually-specified (which, after all, is pretty much the definition of defined-contribution plans), most 401(k) participants will need help with their investment decisions, but they also need the opportunity to tailor their investment programs to their idiosyncratic circumstances.
In our view, the appropriate middle ground looks like this: Participants confront a dramatically simplified choice of five or six globally diversified, risk-adjusted, pension-style portfolios (rather than a long menu of mutual funds written in Wall Street language that might as well be Sanskrit). To work outside those boundaries, they must secure the consent of their primary beneficiary (or beneficiaries), as ERISA requires fiduciaries to protect the interests of beneficiaries as well as participants.
The objectives here are (a) to make the process easy and (b) to reduce the likelihood of big mistakes. Done prudently, it can also produce meaningful reductions in plan expenses. We think this type of program solves most of the problems associated with the traditional model of participant direction while leaving room for participants, beneficiaries, and their advisors to connect assets and liabilities.
There's more individual-level risk in this type of program than in traditional pensions, but the defined-contribution regime is here to stay. The question now is how to make such plans work as well as they can.
Blaine F. Aiken, "It's Time to Rethink Retirement Plans," InvestmentNews, July 12, 2009