8 people, 8 separate groups

[Belgarath]

Just noticed the numbers don't match the title of the post. I had already typed this before I saw page 2 of the census, which had 4 more employees. But the question is the same.

We had a question come up from another TPA who does not handle cross tested plans. He has a client who has 12 employees, (including the owner) and wants to have 12 groups.

We've never administered a plan on this basis. I know at one time there was some discomfort that the IRS might consider this a "deemed CODA" and there were also dark hints, rumors, and mumblings about it being a lack of a definitely determinable benefit. Sal Tripodi makes reference to a 2001 ABA conference where the IRS did NOT raise the CODA question, but I don't have a transcript of the question and IRS response.

While such a design doesn't necessarily thrill me, I also cannot find any guidance that specifically prohibits it. If after consulting with his tax/legal counsel, the client still wants to pursue, they can full for a full determination letter and we'll see what the IRS does.

My question is - have any of you ever had a client receive a determination letter on a similar basis? Or had one turned down? I'd rather benefit from someone else's misery than experience it on my own...

[Guest texasactuary]

We do this in some of our plans. the language takes you out of volume Submittor and so you have to point it out in your submission but we have gotten favorable determination letters on plans with "each individual is their own individual group" language.

[Blinky the 3-eyed Fish]

Accudraft documents provide this option within the VS umbrella and are routinely approved by the IRS. The deemed CODA issue appears to be off their radar for the time being. Just make sure you pass coverage on the ratio percentage test each year. There have been lenghty discussions on that topic previously if you aren't clear why I say that.

[Guest merlin]

Also make sure that every year's allocation to each group is spelled out in a resolution of the appropriate governing body: board of directors, management committe, etc. This will help insulate the plan against the "deemed CODA" issue.

[TBob]

While I have not worked on any plans that are using this method, I have heard that there is a high potential for age discrimination with this design. It seems like a no-brainer to give the youngest participant(s) a higher contribution in order to pass. This would seem like age discrimination absent some carefully documented reasoning why they received a higher allocation than others.

My 2 cents worth...

[AndyH]

Do we think the Kerry administration will allow this design? They'll probably bring in George McGovern as Treasury Secretary, unless they can find somebody more to the left. Maybe Howard Dean? Does he support discretionary cross tested plans?

[SoCalActuary]

Interesting political comment.

Tbob - welcome to the much more complex world of 401a4 regulations and plan designs.

If you recognize the math of the cross-tested plans, there are inherent age-discrimination issues to both db & dc plans. Unless or until the EEOC gets authority to regulate or litigate these issues, you should probably ignore the age-discrimination problems arising from a common-sense look at these plans. Instead, look at the possible plan designs and their effect on discrimination testing.

Better yet, consider attending one of the advanced seminars on the subject. My favorite is the Larry Deutsch program.

[AndyH]

We in the Northeast know what is coming!

[mbozek]

SoCal: What are the inherent age discrimination issues in DB and DC plans?

[SoCalActuary]

MBozak - Just political rhetoric. I don't see any.

But those who do say: DC plans generally don't provide the same retirement benefit as you get older. Therefore lower benefits by age.

DB plans provide higher value when you get older, so you earn more benefit value than a younger person getting the same pay. Therefore higher value by age.

If politicians, social engineers (do-gooders?) and regulators could just agree that both db and dc plans are legitimate benefit delivery programs, then we could eliminate such absurdities as age-discrimination lawsuits in cash balance plans.

TBob - if you make value judgements on plan design for discrimination testing, it is an interesting philosophical perspective. Remember that 401a4 testing forces the plan sponsor to give some participants the same high benefits provided to the older & higher paid leaders of the business. It's not "fair" that some get more than others. So what? Following the reg's and keeping the clients out of trouble with the IRS and DOL are enough trouble without trying to be the unappointed union rep for the other employees.

[TBob]

To clarify my previous post on the subject...

SoCal...Cross testing is not a new concept to me although I do not consider myself an expert nor am I an actuary or a lawyer. I have worked with many cross tested plans over the past few years and feel that I have a good grasp of the basics and a fair understanding of the more complex issues, yet I learn more every day from you experts! (not being sarcastic!)

I am also not making value judgements regarding plan design or the amount of the allocation going to the big dogs in the company. My job is not to write the laws that protect the NHCE but to provide the best allocation per my clients wishes that stays within the boundaries of those laws. If my cleint's goal is to maximize the HCE's while minimizing the NHCEs, then I gladly calculate the best way to do that within the boundaries of 401(a)(4). Anyway...off the soap box...

Regarding Age Discrim...I sat in a teleconference with a lawyer whose firm had claimed to have pioneered the idea that you could put every participant into their own class for cross testing. He suggested that a plan sponsor that uses this design should take care to document very thoroughly the criteria they used to determine the allocation to each NHCE. Since a greater allocation to younger, lower paid NHCE's will raise their EBAR more than the same contribution to older higher paid ones, he felt that giving the younger NHCE's more just to pass the test could become age discrimination. Again, I'm not an attorney, just passing along what I've heard in the past.

[SoCalActuary]

TBob - thanks for your comments, I understand your concern a little better now.

The attorney offering the comments was reaping doubt, in my opinion. But then again, maybe that person sees a litigation risk I don't see.

When the plan sponsor selects the individual rates for one-person one-rate testing, they can give their guidelines on who should receive contributions. If the TPA or actuary advises a set of benefit rates that works for 401a4, the plan sponsor can still insist on minimums or other changes in treatment.

This actuary does not feel comfortable saying when age-discrimination is real or perceived, because I have no feel for the legal arguments made successfully or not in court. Therefore, if I offer a design solution that passes a4 and 410b, the final decision is still with the plan sponsor. If they are concerned about age discrimination, sex discrimination, racial discrimination, or nepotism, then I need to hear it from them.

Your point was interesting.

[AndyH]

TBob, it sounds to me like the lawyer you mention lacks an understanding of the mathematics of cross testing.

As you know, giving a young NHCE more to pass the test is the exact opposite of the goal. The goal of course is to give the young NHCE less and justify that because he is younger. If anything, that is reverse age discrimination.

[mbozek]

There is no such thing as reverse age discrimination under federal law, in the sense that benefits cannot discriminate against younger employees. In a DB plan age discrimination prevents a lower level of benefits for older employees on account of age, e.g., IBM cash balance plan discriminates against any employee because employee at same salary level who is one year younger would get a larger accrued benefit at NRA for each years accrual due to the additonal interest credited to his account in the CB plan which results in a greater annuity payable at NRA. Age discrimination under ERISA covers all employees, not just those over 40.

[SoCalActuary]

Cash balance age discrimination only makes sense if you look at the monthly retirement benefit, not at the cost of employment. For two employees getting the same cash balance allocation, their cost of employment is essentially identical. I personally don't call that discrimination.

[mbozek]

But the Federal ct held that ERISA age discrimination is determined with reference to the monthly retirement benefit at normal retirment age which always results in a reduction of the rate of benefit accrual under ERISA 204(b)(1)(H) on account of age because a younger employee will receive a higher allocation of interest from the year of allocation to NRA than an older employee earning the same compensation. As the judge noted this formula would be permissble in a DC plan.

[SoCalActuary]

That's why there's a Supreme Court.

Other cash balance litigation said that the concept described is absurd on its face.

Who will win? Those who believe cash balance is a great way to deliver benefits or those who can't make it work with existing regulation issued before these plans became popular?

Cash balance enthusiasts say - it's a money purchase plan with a guaranteed interest rate and db benefit limits. Employees get two guaranteed amounts of 1-contribution and 2-interest credit. This does not sound like some evil thing to be destroyed.

[mbozek]

What case are you talking about? Xerox recently settled a case involving valuation of CB benefits of teminated employees for 240 million after losing an appeal in the same circuit ct where IBM is appealing its decision.

From a structural viewpoint CB plans are regarded as DB plans in which the interest component accruing on each years allocation creates a lower relative accrued benefit at NRA for any employee when compared to a younger employee earning the same salary. While some can say the difference in benefit accrual results from the time value of money, the lower benefit accrual at any relataive age when compared to an employee 1 year younger can be construed to be age related which violates ERISA. Previous Sup Ct decisions have held that age discrimination in benefit plans are decided under ERISA and not the ADEA, e.g., ERISA protects all employees in a retirement plan not just employees over 40. Under ERISA age discrimination will be juged by the result, e.g., a lower benefit accrual for older employees, not the intent of the employer to increase benefits through use of time value of money in the formula. Given the strict constructionalist majority on the Sup ct and the hostility in Congress to recognizing an exception in the ERISA prohibition against age discrimination for CB plans I dont think IBM would prevail if the case was decided with the present justices in the Sup Ct. IBM has an additional problem in defending the CB formula because it is also defending a complex PEP formula previously used under the plan which reduced the benefit accruals of older employees which was found to be discriminatory by the trial Ct.

[SoCalActuary]

Let's move this discussion to the DB-Cash balance forum.

Cash balance plans cover millions of employees, including a number of large employers. The state of California covers their part-timers with such a plan.

ATT, Bank of America, and a number of others would be guilty of age-discrimination if the old 411 regs are the law of the land.

I believe this would make all such plans impossible to maintain.

I am looking for the cites from the old cash-balance cases. Georgia-Pacific and Bank of Boston were two of the litigants. I even gave a speech on this a few years back. The judge in one of these cases heard your argument and ruled against it.

From my review, the IBM and Xerox cases had other issues that were bad cases for setting precedents. In some respects, they "deserved to lose" on some of their issues. This did not justify the old DB only accrual rules being applied to cash balance designs.

For that matter, any level cost accrual method in a db plan would meet the same failure as cash balance plans. This would include traditional db plans that use the individual reserve method for calculating accrued benefits and the fully insured 412i plan. It is easy to demonstrate that the level contributions put in at an earlier age produce a higher rate of accrual than the same amount of contribution at a later age, thus violating the old obsolete db accrual rules.