Age Weighted Plan

[Guest rmwright]

Age weighted plan has 3 NHCEs (age 56, 60 and 52) and 1 HCE (age 70). NRA is age 65. 401a4 testing is bombing.

What options are there to pass the nondiscrimination testing?

Thanks!

[rcline46]

Why is non-discrimination bombing? If you mean a cross tested plan with groups that is one option, but a traditional age weighted plan should not fail.

[Guest rmwright]

I'll be honest - I don't have much experience with age weighted plans. From what I've researched, the testing should not be a problem as long as you don't have a HCE participant who is older than the NRA in the plan. That's the situation that I have here with this client.

I contacted Relius and they tell me Relius is calculating the EBARs based on the participant's age 65 and not 70. As a result, the system is calculating a higher EBAR than all of the other participants.

[rcline46]

Set NRA to 65 and 10 participation - yeah, you cant do that in a document.

Set the person's nRA to age 70 if Relius does not have that. An assumption.

I will not vouch for the legality of these items, but it should push the NRA over 65 to get a better EBAR. Check with your advisor/actuary to see if they approve it. I agree the testing is somehow 'wrong'. Actually, the 'points' for this person should be determined the same way the testing is done - from 65 - and not current age.

[Tom Poje]

years ago I wrote an example for the Coverage and Nondiscrimination Answer Book so I'll provide that info, its from Q 11:25 and example 11-9. this 'problem' occurs because the APR for someone who works past NRA is lower than that for NRA.

therefore, consider 2 people, both making the same amount of money, one age 65 and one age 70. the age 70 individual would have a smaller amount of point through the age weighted formula if you used current age. this would be a penalty on the older person - he receives a smaller contribution, so usually age weighted plans use the APR at 65 for person who continue to work past age 65. but as is noted in the thread, the end result is it creates a larger E-BAR for the individual and appears to cause the plan to fail testing. The key reg cite is Treas. Reg. § 1.401(a)(4)-8(b)(1)(ii)]which basically says dont sweat it if alloactions are made at the same rate for someone after testing age as for someone at age 65. (hmmm, I must have the user friendly version of the regs)

since I edit the book, I believe I have permission to give a direct quote

...........................

Two employees, one age 65 (Employee 1) and one age 70 (Employee 2), each earn $100,000 per year. Normal retirement age is 65. The plan uses the UP 1984 mortality table, the preretirement interest rate is 8.5 percent, and the postretirement interest rate is 8.5 percent. Contributions for the year total $50,000. If the plan used APR at actual age if one works past normal retirement (assuming an APR of 84.0341 at age 70), the result would be as follows:

Employee 1: $100,000 × 95.3828 × .01 = 95,382.80 (53.16 percent of total)

Employee 2: $100,000 × 84.0341 × .01 = 84,034.10 (46.84 percent of total)

Total Points

179,416.90

Employee 1 will receive $26,581.33 ((95,382.80÷179,416.90)×$50,000)

Employee 2 will receive $23,418.67 ((84,034.10÷179,416.90)×$50,000)

The end result would seem to be discriminatory toward an individual who works past normal retirement age.

To get around the problem of discrimination, most plans are written to use the APR at normal retirement age when determining the number of points. Since both employees make the same amount of money, using the same APR would result in the same contribution for both employees—in Example 11-9, $25,000 a piece. But what happens when you perform nondiscrimination testing (in which case the APR used is for the actual age of an individual who works past normal retirement age)? [Treas. Reg. § 1.401(a)(4)-12, definition of testing age] A review of the E-Bars would result in the following expression:

Contribution×(1.085)years to Retirement÷APR×12÷Compensation

Note.

Neither employee has any years left to retirement.

Employee 1: $25,000 × (1.0850) ÷ 95.3828 × 12 ÷ $100,000 = 3.145

Employee 2: $25,000 × (1.0850) ÷ 84.0341 × 12 ÷ $100,000 = 3.570

This would seem to raise the possibility of failing nondiscrimination testing. However, a plan will not fail discrimination testing merely because allocations are made at the same rate for employees who are older than their testing age (determined without regard to the current age rule in paragraph (4) of the definition of testing age in Treasury Regulations Section 1.401(a)(4)-12) as they are made for employees who are at their testing age. [Treas. Reg. § 1.401(a)(4)-8(b)(3)(ii)]

........................

now as to getting relius to generate a report that would indicate this, well I suppose you could go into census temporarily change the date of birth (do not rerun eligibility or anything, just change the date of birth) so it appears the person is age 65. then rerpint the nondiscrim report. after all for calculating the persons contribution you treated the person as if they were age 65. then after you have printed the report change the date of birth back to what it should be

[Guest rmwright]

Thank you! Thank you! This works!

Have a great day!

[Mike Preston]

The key reg cite is Treas. Reg. § 1.401(a)(4)-8(b)(1)(ii)]

This I agree with.

[Treas. Reg. § 1.401(a)(4)-8(b)(3)(ii)]

The (1) magically changed to a (3). Shouldn't this also be "(1)"?

Also, note to OP: Tom's method only works if you satisfy the requirement of that particular regulation: the document must call for the contribution to be determined as if the affected individual was age 65. If it doesn't do that, then you don't satisfy the regulation. In other words, an age-weighted document can be drafted such that it takes advantage of this rule, but it doesn't need to do so.

[Tom Poje]

Mike: all for 1 and 1 for all. but my typing ability is something else.