Fixing Small-Company 401(k)s

Two Worlds Practically speaking, there are two U.S. employer-based retirement systems, one in relatively good shape and one not. Nearly every large company sponsors a low-cost 401(k) plan, generally equipped with auto-enrollment and auto-escalation programs. In contrast, half of small companies have no defined-contribution plan, and those that do generally carry pricey funds and limited features.

(Pension plans were once a third system, but no longer. Currently, only 5% of Fortune 500 companies offer a traditional defined-benefit plan to new employees.)

The big firms can assign human-relations and finance specialists to establish and monitor their 401(k) plans. Meanwhile, at the small firms, that task typically falls to somebody who works with the firm's main business. Often, that side project falls between the cracks, so that no 401(k) plan is formed. If a plan is created, it will have few assets--and thus will likely be stocked with expensive funds, as the fund company attempts to compensate for the plan's diseconomy of scale.

In short, small-company plans face a structural handicap, one which the invisible hand has not grasped. To be sure, some providers, Vanguard most notably, now offer significantly cheaper small-company plans. But most small 401(k)s remain pricey. In addition, cost improvements do not help those employees who work at the 50% of small firms that lack a plan.

Two simple legislative changes would lead to great improvement. They are advanced in a new Morningstar white paper, "Small Employers, Big Responsibilities," authored by me, Jake Spiegel, and Aron Szapiro. (My name is the first listed because of alphabetization--not a contest that I normally win--but Jake and Aron, being Morningstar's Washington-based policy experts, did the thinking. I prettified the language.)

Open Multiple-Employer Plans Today, small companies are permitted to band together to form a "multiple-employer plan" (MEP). The MEP aggregates assets from each of the affiliated companies into a single plan, thereby enabling those small firms to enjoy something of the large-company advantage. They can pool their resources when running the plan, so that the research, monitoring, and documentation efforts are shared. They also will have a bigger asset base, which enables them to negotiate lower plan costs.

So far, so good. However, as with many 401(k) regulations, the laws governing multiple-employer plans are capriciously complicated, because they were formulated decades ago under different circumstances, and never revisited. Which means that for no good reason, they often prevent that which they seek to permit.

For example, the IRS distinguishes between "closed" and "open" MEPs. Closed MEPs are useful, because the IRS treats them consistently as single entities. However, for a MEP to qualify as closed, all companies must operate in the same industry. The restriction is purposeless; prohibiting companies from different industries from sharing a retirement plan creates no public good.

If such companies do wish to belong to the same MEP, then that MEP is defined as open. An open MEP, as currently defined, is not an attractive option. Each member of an open MEP must file an annual 5500 form; each must meet its own nondiscrimination standards; and if one company is noncompliant, then the entire MEP is treated as noncompliant. Thus, belonging to an open MEP means doing much of the work of a stand-alone plan, while being tethered to somebody who might leap off a cliff. Some arrangement that.

Morningstar's suggestion: Simplify and rationalize the rules. Make all MEPs open by removing the single-industry requirement. Then, make them usefully open, by reversing the current logic. Instead of mandating that companies within a MEP continue to do their own work, while being legally shackled to other firms, permit the work to be conducted within a single, central location, and relax the individual legal responsibilities. Make open MEPs easy to join and stuffed with benefits.

To do so would cost the government, companies, and employees nothing. The task is as simple as rewriting a couple of statutes.

Automatic IRA Enrollment Morningstar is not alone in advocating changes to the existing MEP regulations. The problems are obvious and the solutions straightforward. Most who study the 401(k) industry make a similar recommendation.

Improving MEPs, however, only takes small firms so far. Upgrading the rules will encourage some companies that lack a retirement plan to join a network, and it will help to lower costs for firms that already have created plans. Those are good things. However, this legislative fix will not push all parties over the line. Many small businesses will continue to do what they have always done, through inertia, which means that their workers will have no access to a tax-sheltered, employer-based retirement plan.

Which brings us to our second suggestion: Enroll such employees into an IRA account.

Seven states have already established such programs. The details differ somewhat among each, but the essentials are intact. Within those states, new employees at companies that lack retirement plans are enrolled automatically into an IRA plan. As with all automatic-enrollment features, the employee is not obligated to remain in the plan; the enrollment is effectively a suggestion, not a requirement. In practice, though, most people stay.

The effort is praiseworthy, but the fragmentation is not. There are already seven systems in place, with 43 states remaining. There is no more reason to have 50 different IRA-enrollment programs than there is to have 50 different currencies. Establish one national program where the rules and details are unchanged, no matter where the employee moves or a company's headquarters are located, and be done with the matter.

This proposal, unlike the MEP item, would require that companies give some effort. Qualifying businesses would need to establish a technological link to the IRA program. However, as Morningstar would only mandate IRA enrollment for companies that have electronic payroll systems, the required effort would be minimal indeed. Arranging IRA withholdings would be no more difficult than would enacting any other form of payroll deductions.

These IRA plans would be floors, not ceilings. Companies would neither be required nor expected to provide any contribution matches. The maximum contribution level would be set lower than those of 401(k) plans, and certain other features would be less desirable, so as to encourage the companies to upgrade from the IRA enrollment to offering a full-fledged 401(k) plan. National IRAs would fill a gap--but oh, what a gap.

Wrapping Up These two proposals would be easy to implement, and force no mandates onto employees. They would retain their economic freedom. The benefit, should the proposals be passed, would be to give tens of millions of Americans access to a company-based retirement plan. In addition, small-company 401(k) costs would decline. That sounds like a great deal to me.

If Congress served the best interests of American workers, without regard to scoring points in political debates or satisfying lobbyists, these proposals would quickly be passed and implemented. They cannot be reasonably opposed. Whether such an action will occur is, of course, another question altogether. There isn't much room for optimism there.

John Rekenthaler has been researching the fund industry since 1988. He is now a columnist for Morningstar.com and a member of Morningstar's investment research department. John is quick to point out that while Morningstar typically agrees with the views of the Rekenthaler Report, his views are his own.