In Parts I and II of this series, I discussed Teresa Ghilarducci’s proposal for a government-run retirement program and my recommendations for improving 401(k) plans. This column explains why some of my proposals are less revolutionary than they might appear.

Automatic Enrollment
I suggested that automatic enrollment should be mandated for all 401(k) plans. While that might seem like a big change, it is not. Based on various reports, more than half of the larges plans in America already are enrolling automatically—and the results are good.

My proposal for automatic enrollment would, at first, require that only newly eligible employees be enrolled. Then, after five years, all employees would be enrolled automatically. The reason for the difference is that there can be a significant additional cost for matching contributions if all employees were required to be enrolled immediately. On the other hand, once five years has gone by, and automatic enrollment has increased participation a little bit each year, there would only be a relatively small number of nonparticipating employees who pre-date the beginning of the automatic enrollment program.

Automatic Deferral Increases
Plans should be required to provide for annual 1% automatic deferral increases up to at least 12% of pay. That is based on studies that show that employees need to be adding at least 15% per year (counting both deferrals and company contributions) to have a reasonable chance of ending up with adequate benefits.

My sense is that Vanguard has done the best job of encouraging plan sponsors to use automatic deferral increases in conjunction with automatic enrollment, and their data show the results: A significant number of their plan sponsors are using automatic deferral increases. For a sense of the benefits of automatic deferral increase, look at the research done by Shlomo Benartzi and Richard Thaler.

Qualified Default Investment Alternatives or QDIAs
Plans should automatically put employees in well-diversified portfolios, like the types of investments that are QDIAs—target-date funds, lifestyle and balanced funds, and managed accounts. The employees could then re-direct their investments if they want. There are ample data that show that most participants do not know how to build portfolios that properly balance risk and reward.

There is some concern that target-date QDIAs may be structured to be too aggressive—particularly for employees within 10 years of retirement. In looking at the investment results from 2008, most 2010 target-date funs were down 25% to 30%. In other words, a participant within a year of retirement at age 65 lost 25% to 30% of his 401(k) account. While investment experts might believe that those results are inherent in investing, many retirement experts would say that those losses are unacceptable. As a suggestion, Congress could mandate that, in order for an investment to qualify as a QDIA, the fiduciaries must determine—or the investment provider must certify—that material consideration has been given to safety during the 10 years before the fund’s maturity date.

On the other hand, where a plan offers distribution guarantees (like guaranteed minimum without withdrawal benefits), the target-date funds could be invested to focus more on growth because the downside is protected.

Conclusion
To paraphrase Einstein, it is insane to keep doing the same things and expecting difference results. We need to change our 401(k) perspective from voluntary savings plans to broad-based retirement plans. The first step is to increase participation; the second is to increase deferrals; the third is to improve the quality of participant investing in a way that supports both the return on capital and the return of capital (to paraphrase Will Rogers).

Disclaimer Required by IRS Rules of Practice:
Any discussion of tax matters contained herein is not intended or written to be used, and cannot be used, for the purpose of avoiding any penalties that may be imposed under Federal tax laws.

Source: PLANSPONSOR magazine