Cross-Testing Beyond The Basics. Karen Smith, President, Nova 401(k) Associates

Transcription

1 Cross-Testing Beyond The Basics Karen Smith, President, Nova 401(k) Associates

2 Audience Level Advanced Solid experience with coverage testing and 401(a)(4) testing Solid experience with cross testing Time 50 minutes (not much time!)

3 Agenda The best factors to use for calculating your EBARs Component plan testing Current year testing vs. accrued-to-date Rate group banding Using corrective amendments

4 Best factors for EBARs

5 EBAR If testing on a contribution basis, EBAR stands for Employee Benefit Allocation Rate. Don t need assumptions If testing on a benefit basis, EBAR stands for Employee Benefit Accrual Rate. Pre-retirement interest rate Post-retirement interest rate Post-retirement mortality

6 EBAR How to normalize an increase in account balance Choose assumptions Accumulate to Testing Age Determine Equivalent Benefit Divide Equivalent Benefit by Compensation to arrive at the EBAR

7 EBAR Standard assumptions 1.401(a)(4)-12 Interest 7.5% - 8.5% Pre-retirement None Post-retirement mortality UP GAM M or F 83 IAM M or F 71 GAM M or F 71 IAM M or F 417(e)

8 Best factors for EBARs Best factors are usually at one of the extremes 8.5% 1971GAM M typically best 7.5% 1983 IAM F may be best for calculating DB/DC average benefit offset or other very special cases DB/DC testing with gateway, sometimes an intermediate set of assumptions will help manage the gateway requirement.

9 Best factors for EBARs Can you use assumptions other than standard assumptions? Must use reasonable assumptions. Standard assumptions are deemed reasonable. What other assumptions would be reasonable? And would they be better than standard assumptions? Not aware of anyone who uses other than standard assumptions

10 Component Plan Testing Component Plan Testing

11 Component Plan Testing Component plan testing allows us to treat a plan as if it were multiple plans and then test each plan using different methods. Component plan testing is helpful when you have multiple, heterogeneous HCEs. It allows you to test each HCE using that HCE s optimal testing method. Not exotic any more?

12 Component Plan Testing Component plan testing can be a great tool when unexpected HCE changes. It can also be helpful to design plans from the outset with component plan testing in mind.

13 Requirements for Component Plan Testing Component plan is for 401(a)(4) testing not the following Coverage, ADP/ACP, Gateway, Top Heavy. or 401(a)(26). Each participant must be in exactly one component plan.

14 Requirements for Component Plan Testing The plan satisfies coverage before restructuring. Each component plan must satisfy the coverage test. Generally component plans satisfy the 70% ratio percentage test so that you do not have to worry about reasonable classifications. It is not possible to restructure some small plans using the 70% ratio percentage test. The numbers don t work. Example: 2 HCEs and 3 NHCEs No way to get both rate groups over 70%.

15 Requirements for Component Plan Testing Each participant must be in exactly one component plan. ABPT is satisfied on plan/controlled group basis. Gateway requirement applies if any component plan requires the gateway.

16 Common Component Plan Testing Example A 60 year old owner with $300,000 compensation and his 30 year old wife with $30,000 compensation Owner will test well on an accrual basis Wife typically will not test well on an accrual basis because she is too young. So, testing her on a contribution basis with either a percentage a pay or flat dollar amount typically will work better.

17 Other Component Plan Testing Examples Mix and match Net pay vs gross pay Average pay vs current pay Imputing disparity vs. not imputing disparity Accrual basis vs. contribution basis

18 Constructing Component Plans Put each targeted HCE in her own component plan Choose method most likely to pass 401(a)(4) for that HCE Review NHCEs and put the NHCEs who will help an HCE pass into that HCE s component plan Allocate remaining NHCEs and HCEs so that component plans pass coverage

19 Component Plan Mistake The only component plan mistake I regularly see is not using component plan testing enough! Lots of missed opportunities.

20 Current year testing vs. accrued-to-date

21 Current vs. ATD for Cross-Testing Profit Sharing Allocations Current What do you accumulate? Current PS Allocation Current PS account ATD To what age do you accumulate? Testing Age Testing Age What assumption do you use to accumulate? Standard interest rate What is the benefit accrual? Accumulated amount / (APR * 1) Standard interest rate Accumulated amount / (APR * Testing service) What is Testing Compensation? Current year or average Must be average What is EBAR? Benefit Accrual/Testing Compensation Benefit Accrual/Testing Compensation

22 ATD vs Average Compensation Often using average compensation with current accrual will capture most of the difference between current accrual method and ATD method

23 ATD Opportunities Current or recent plan increase for HCEs relative to NHCEs Using ATD will capture that the HCE s prior year accruals were lower. Eg. The plan had a pro-rata allocation and is being amended to a new comparability formula. Using ATD, will capture the HCE s lower allocations in prior years.

24 ATD Opportunities Fresh start events with the accrued to date method, you can choose a fresh start date and disregard benefits before that date. If there has been a change in the plan or demographics, sometimes choosing an fresh start date when that changed occurred will provide testing opportunities.

25 ATD Opportunities HCEs have lower investment return than the NHCEs ATD captures investment return, so if the HCEs are earning less investment return than NHCEs capturing this in the accrual rates may be helpful. This could happen when HCEs have made poor or risky investment choices or have invested conservatively on purpose.

26 ATD Mistakes Checking the box to run ATD without loading the necessary account balance information Checking the box to run ATD without loading the necessary compensation information for the compensation average Including participants who do not have a current year accrual in the test

27 Rate Group Banding

28 Rate Groups A rate group is an HCE and all participants who have an EBAR that is higher or the same than an HCE. Sort the participants by EBAR from highest to lowest. Draw a line under each HCE and the HCE and everyone above her make up a rate group

29 Rate Groups The maximum number of rate groups is the number of HCEs. Participants (including HCEs) are often in multiple rate groups.

30 Rate Group Banding With banding, treat all employees as having the same EBAR if they are all in the same corridor: If testing on an allocation basis, 5% above/below midpoint For normal accrual rates, 5% above/below midpoint For most valuable accrual rates, 15% above/below midpoint

31 Rate Group Banding Special rule when determining EBARs as % of compensation: If testing on an allocation basis,.25 above or below midpoint If testing normal accrual rates,.05 above or below midpoint If testing most valuable accrual rates,.05 above or below midpoint

32 Rate Group Banding No requirement corridors have the same width. No requirement that all rate groups be banded.

33 Rate Group Banding However, there can be no overlapping rate groups. And, the HCEs can not have a significantly higher EBAR than NHCEs in band.

34 Banding How useful is rate group banding in practice?

35 Banding The requirement that the HCEs in a band not have significantly higher EBARs than the NHCEs limits its usefulness. Be careful when checking a box in your software that does the banding for you.

36 Corrective Amendments

37 Corrective Amendments A plan has failed its 401(4) test or coverage test. You cannot see anyway to get it to pass. What do you?

38 Corrective Amendments A corrective amendment under 1.401(a)(4)-11(g)!!!!

39 Basic 11g requirements A testing failure What is a testing failure? 410(b), 401(a)(4), 401(a)(26), BRF testing What is not a testing failure? Operational failures such as allowing loans when the plan document does not provide for loans, early entry into the plan, failure to distribute safe harbor notices Does there really have be a testing failure?

40 Basic 11g requirements No failsafe language Some documents have language that automatically corrects certain testing failures. Failsafe language may specify who to add to the plan if there is a coverage failure Failsafe language does not allow for flexibility in correcting testing failures, but it does allow for certainty.

41 Basic 11g requirements Time for adoption of the amendment Generally 9 ½ months after the end of the plan year Sometimes longer if the remedial amendment period is longer

42 General 11g requirements The 11g amendment may not reduce any participant s benefit (a)(4) -11(g)(3)(ii) Can not take money out of an HCE or key employee s account. Can not reduce a benefit that is hard-coded in the plan. Can not kick an HCE out of the plan

43 General 11g requirements The 11g amendment cannot be discriminatory. The amendment itself must be non-discriminatory. If you bring in an HCE to pass 401(a)(26), you have to be able to show that the amendment is non-discriminatory so you will have to increase benefits for NHCEs or bring in NHCEs. The end result must be non-discriminatory. If you bring people in for 401(a)(4), remember you still have to pass coverage and your testing group may have changed. You must get to a pass.

44 General 11g requirements The 11g amendment must have economic substance. Giving an additional amount to a participant who has terminated not vested in the testing year does not have economic substance. How vested does a participant need to be?

45 Special considerations If correcting coverage for 401(k) or 401(m) portion of a plan by adding participants, a QNEC will be involved for the participants who are added. Special rules to correct BRFs and rate of match failure

46 Special considerations Document issues Lose reliance on opinion letter or determination letter? Single amendment or separate amendment? Safe harbor 401(k) plans

47 Special considerations Perfect Logistics Allocate money to the plan Failing test Amend plan Allocate additional amount

48 Examples of 11g amendments Takeover profit sharing plan had an allocation condition of last day of the year employment. The plan fails 401(a)(4) testing. By 11g amendment, all NHCE participants who terminated prior to the last day of the year are provided an allocation and vested if necessary. A cash balance plan does not pass IRC 401(a)(26) testing. By 11g amendment, the cash balance allocation is increased for only the year in question.

49 11g mistakes Correcting the testing failure in the obvious way instead of the optimal way. If going to do a 11g amendment, might as well, get a really good result. Example Correcting a coverage testing failure by bringing more NHCEs into the plan instead of looking at increasing allocation to pass ABPT. Example Correcting a 401(a)(4) testing failure by giving a broad group of participants an additional allocation instead of giving an additional allocation to a select group of participants.

50 11g mistakes Forgetting to show that the amendment itself is nondiscriminatory. Example bringing in an HCE to pass 401(a)(26) and not testing to show the amendment itself is not discriminatory. Think about how to test the amendment in your software.

51 11g mistakes In a DC plan, the plan amendment not being carefully drafted to be clear on the money types the person is being brought in for. i.e. inadvertently admitting to plan for deferrals in addition to profit sharing In a DB/CB plan, amending the formula to provide a benefit for an individual but not removing the exclusion language.

52 Questions?