From virtually a standing start in the early 1980s, 401(k)s and other defined contribution plans grew to enroll well over 62 million working Americans by 2006, roughly half of America’s total workforce. Over that time, defined contribution plans accumulated a total of over $4.2 trillion. The workplace system had also begun “rolling” multiple billions of dollars into Individual Retirement Accounts and become the largest source of IRA flows.
This is a doubly astonishing story because 401(k)s and other DC plans were initially seen mainly as supplemental savings vehicles, not as the primary sources of retirement income. Policymakers in Washington saw them as “add-ons,” if you will, to a retirement system grounded on traditional defined benefit pensions and Social Security.
Clearly, some companies were quick to recognize the potential of their 401(k)s. Many new firms, such as Microsoft, used defined contribution plans as their only retirement option all along, never even establishing a traditional DB pension.
Partly because they were seen as supplemental, this first generation of DC plans was purely voluntary. Both plan sponsors and participants had to opt in by choosing to offer plans or by deciding to take part.
Over time, 401(k) plans offered an increasingly wide array of choices. By its very nature, this voluntary, multichoice plan design required multiple decisions from participants — to actually enroll, to choose a deferral rate, and to structure their own retirement investment portfolios from a growing array of options.
These choices, in turn, were heavily dependent on quite expensive communication and educational programs. The underlying assumption was that we could somehow turn nearly every working American into a skilled pension manager.
And while the long bull market lasted, from 1982 to 2000, many people thought that was working out pretty well. As workplace savings grew to hundreds of billions, the 1980s and 1990s brought a wave of competition from mutual fund companies, insurers, and other providers, and that drove continuous improvement in features and services, including online access and daily valuations.
Competition also restrained the fees charged by the mutual funds used in 401(k) plans, whether those were stock, bond, or money market funds — keeping fees well below asset weighted averages for the industry and enabling workplace fund investors to secure discounts not available outside their plans.
With markets booming, assets surging, choices proliferating, and fees well under control, it is hardly surprising that by the turn of the new millennium, the defined contribution concept had become America’s retirement plan of choice, passing traditional defined benefit programs in assets and numbers of participants. And DC plans have continued to widen that lead.
But as the century turned, the first generation of DC plan design had clearly begun to hit the ceiling on several fronts. Participation rates had stalled in the mid-to-high sixtieth percentile range. Roughly one third of eligible workers were not signing on, no matter how much we begged them — or incented them — with matching funds. And despite intense, expensive efforts at communications and education, savings rates, too, had stalled at just over 7% system-wide.
As for asset allocation, it was all over the map. Far too many young workers were huddled in low-risk, low-return stable value funds, while some of their colleagues on the cusp of retiring were 100% exposed to equities.
The purely voluntary, multichoice model, in short, had reached the point of diminishing returns. Why? Because it required participants to make the right call several times, first to participate, then to save the maximum allowed by law, then to diversify well, and then to rebalance as markets moved. It therefore offered multiple points of failure and required several correct decisions to succeed.
The good news was that fresh thinking from behavioral finance research in academia and live experience in the workplace had evolved to address those flaws. All of the key elements needed to create a better model were emerging. These solution elements included automatic enrollment and savings escalation programs — which pilot studies showed to have a dramatic, positive effect — and dynamic life-cycle funds, which solved for diversification, rebalancing and risk-mitigation automatically. Life-cycle funds began gaining acceptance by the late 1990s and really took off following the dot.com bust early in this decade.
The single most important insight of the new thinking was to recognize that the main variable in workplace savings is plan design itself. What behavioral economists call “framing” or “the architecture of choice” heavily influences the choices that people actually make.
The dominant force in participants’ behavior is inertia, which keeps most people on the course they initially choose. That is why getting people on the right track automatically is so vital. Once people are enrolled, and their savings are set to rise automatically, and they are guided to smart asset allocation strategies, inertia alone keeps most of them on the right track.
Perhaps the most striking example I could cite to you of the sheer power of inertia is a now classic study of the differing levels of organ donor rates we see among quite similar countries in Europe. In Germany, roughly 12% of the people have signed up as organ donors voluntarily after a massive education effort that has gone on for years.
Right next door, in Austria, the donor rate is 99.98%. These people are next-door neighbors. They all speak German. So why this huge difference? Because the Austrian system automatically makes you an organ donor. You can opt out freely, but if you don’t, you are in.
Excerpted from a speech given by Robert L. Reynolds President and CEO Putnam Investments, at the 401KWire.Com Influencers’ Summit 2009: DC-IO Partnership Washington, DC May 6, 2009. The full speech is embedded below.
