Increasing Deferral Rates by Employees

[Guest Tracy]

Our recordkeeper informs me that for Plan years beginning in 1999, we may exclude from testing employees who have less than a year of service or are under age 21 even if the Plan allows such employees to participate.

As a means of helping our ADP test, we are thinking of amending our Plan to allow employees to begin contributions to our 401(k) from date of hire (or after, say 3 months) in the hopes that more employees will begin contributing and keep contributing after completing a year of service. Currently, our Plan requires 1 year of service and 1,000 hours worked.

Does anyone have any thoughts on this topic or legal pitfalls we should be aware of?

------------------

[Guest Dook]

You certainly may do this (permissive disaggregation), as long as each testing group meets the requirements of section 410(B) coverage testing. Your recordkeeper will have to perform those tests.

If the plan uses current year ADP/ACP testing there are no issues however, if prior year testing is employed you will have to determine the NHCE testing percentage to be utilized from year to year depending on whether you continue to disaggregate the groups. (see ques. in this 401(k) thread from Robb Muse 6/4/99 Prior Year ADP Testing)

[MWeddell]

Beginning with your 1999 plan year, if one permissively disaggregates the employees who could have been excluded under Code Section 410(a), i.e. those with < 1 year of service or < age 21, then all nonhighly compensated employees in this otherwise excludable group are ignored for testing and any highly compensated employees (which there usually aren't) are included in the testing with the employees who are age 21 or more and have 1 or more years of service. Code Section 401(k)(3)(F). The short answer, Tracy, is that your recordkeeper is right that shortening the service eligibility requirement doesn't need to affect the ADP test.

Is shortening the eligibility period a good idea for increasing plan participation? Buck's 1997 401(k) plan survey (the question was dropped from their 1998 survey) says with 3 months or less of eligibility, participation is 80%, with 6 months it is 78% and with 12 months it is 74%. I suspect that the cause and effect is much smaller than this correlation implies. Plan sponsors who have high turnover are likely to choose a long eligibility period and likely to have lower participation rates. Nonetheless, there's at least some evidence that the idea is worth pursuing.

Certainly other ways of improving participation are more effective (negative elections, higher match, targeted communications, etc.) and should also be considered if you haven't already.

[Kathy]

I do think that this decision should be based on your own situation and circumstances. If you have high turn-over in employees, you may end up with a lot of small accounts for which you have to pay additional administrative costs. If they're under $5,000 you may have the ability to "cash them out" if your plan document permits this, but if they are over $5,000 only the participant can decide when the money comes out of the plan.

[Tom Poje]

and if the plan happens to be top heavy, you now have to provide top heavy minimums for more employees, those who wouldn't normally be in the plan. Be careful of that.

By the way, disaggregating employees is not new in 1999. you have been able to do this for a number of years. What is new, is that all HCEs, regardless of length of service are treated as having more than 1 year of service. and even that appears to be optional. the new rules did not do away with the old rules of disagregating ees. (that is, you could still count an HCE with less than a year with all other employees with less than a year, but why do an extra test when you don't have to.)