Andy Steiner Sees the (DC) Future

“I’ve always been concerned about determining the appropriate default portfolio for our participants—after all, we had about 27,000 individuals in our DC plan. So, to start with, we concentrated on giving them an appropriate number of basic investment building blocks—all passive—that would serve as a foundation for their portfolios. Once we had that in place, we tackled the question of how we could help them to better construct intelligent portfolios.

By 2000, we had nine investment options in the plan. I felt strongly that the best way forward was to find a way to get advice to our participants and that online advice would be the way to go. We looked at the field of providers and chose Financial Engines. Many plan sponsors were then and still are uncomfortable with the fiduciary implications of providing advice, but our analysis convinced us that the value that we were providing to our participants far outweighed the litigation risk.

(Art by Naftali Beder)

The plan paid for the advice; it really wasn’t a large expense. About 25% of our participants used it. But, by 2006, we were faced with deciding what default option to use: We had soft-frozen our defined benefit plan in 2005, so the DC plan had become the primary retirement plan for new participants. We had become very comfortable with Financial Engines and its executives at this stage, and liked their approach to constructing default asset-allocated portfolios. So we chose their managed account approach, which we much preferred to the various target-date offerings we were then being shown.

It’s an offering that’s been embraced by participants—about one-third of all participants use it, and about 25% of all the plan’s assets are invested in these managed accounts. There is a participant fee to use this service, but even with that included the cost of the service to participants is less than that of most balanced funds available to DC plans. It’s a very personalized service: Participants interact with a Financial Engines advisor who works out with you the level of risk you’re comfortable with and allocates your portfolio across the plan’s nine options.

It’s not clear to me why other plan sponsors haven’t followed suit. To be fair, we always had an unbundled plan, so we avoided the issue of a recordkeeper recommending its own target-date solution. We liked simplicity, and the managed account structure allowed us exactly that. I personally think the default space is up for grabs—an asset-allocated solution of some sort, you may call it a target-date fund if you like, is going to be the correct outcome, but the devil is in the detail. In my view, the key is the glide path. We hear a lot about the underlying assets in the glide path, and particularly right now about alternatives, and that is obviously a topic for lengthy discussion. But I think a much bigger issue is the magnitude of retirement income provided by DC plans. No one wants to buy annuities—the pricing is just too excessive and unclear. But the next generation of these asset allocated funds has to have some sort of outcome that gives individuals income—effectively LDI for DC plans. Whoever first solves this problem, and embeds it into a managed account or target-date fund, will find a lot of interest.

There is an information issue here, too. Regulators are already focused on showing some retirement income metric based on account balances, but the math does not fully show just how unfunded participants are. We are facing a huge funding gap that is going to loom into view the moment that we are able to better define it. That might well blow the roof off our industry, so it’s perhaps understandable that we are not moving particularly quickly to define it, but sooner or later we are going to have to address personal funding status and required contribution rates.”