Corbel Cash Balance Document

[AndyH]

Anybody using the Corbel document for new CB plans? I'm wondering how others are dealing with the limitations.

Interest crediting, for example, is limited to (a) a fixed interest rate or (b) GATT rates.

Unless I am missing something, (a) is not allowed by the IRS at present and (b) really doesn't exist any more.

Also, the allocation group options (% or $) seem a bit limited.

These are the options from a 1/28/08 version.

Anybody else having this experience? Any insights as to their plans and timing?

[SoCalActuary]

PPA says that you use a market rate and avoid age discrimination. If you have a fixed rate, you risk second-guessing.

But informal discussion by IRS people says that a fixed rate of 5% is not a big deal. I don't know what the upper limit is on their comfort range. Personally, I am still irritated that 8.5% is used for general testing when it is not a market rate, but who says the IRS has to be consistent?

As to Corbel's other choices, they respond to the market place, and they allow you to use your own custom paragraphs, so design what you want. If you think it should be a standard feature to use your design ideas, tell them.

[AndyH]

There was a-CCA sponsored Webcast a week or two ago and the IRS people seemed to be pretty strongly advising against a fixed rate.

Plus, under the recent guidance decreasing or even changing the rate seems to be problematic, so I would prefer to "get it right" from plan inception; but there seems to be no acceptable choice at present

Regarding plan design, I am unaware of any present guidance on how allocations or allocation groups are allowed to be designed in a CB plan other than by using a vanilla approach so I'd prefer to have some document guidance from a mainstream provider at this point. Otherwise what is the point of using a Corbel?

[SoCalActuary]

The alternative to a Corbel document is to attend the Larry Deutsch seminars on non-discrimination testing.

Then you will know more about the many choices available, sort of like eating stuff that is not on the McDonalds menu.

The fixed interest rate issue is still a problem, though. Some plan designers, and plan sponsors as well, want simplicity.

I wonder if anyone has submitted a 5% interest credit plan to the IRS and received any negative response. Anyone???

[John Feldt ERPA CPC QPA]

We plan to submit one later this year. On all others we've use one of the "safe harbor" Notice 98-6 indexed rates so far.

Perhaps I'm not sure what is meant by "I am unaware of any present guidance on how allocations or allocation groups are allowed to be designed in a CB plan other than by using a vanilla approach". Nearly all of our cash balance plans have separate benefit accrual rate groups (which match up with the DC plan rate groups) and the plan is tested under 401(a)(4) with the DC. I don't think that's too unusual either, is it?

We've use the Corbel language, but heck, it's individually designed, so we change the items that are needed to be changed within the body of the document. Of course we'll always submit for a D-letter on these individually drafted plans.

[AndyH]

Not having seen many CB plans other than through some seminars, webcasts, etc. I don't have a good feel for design limits if they exist. Years ago Sal Tripodi wrote a guidebook on cross tested DC design which I found to be invaluable back then. I'd love to see something similar on CB plan design.

We thought Corbel's checklist would be a useful guide to basic design. It is limited to percent of pay per group or dollar amount per group. I understand it can be modified but I'd like to get some sense of if any boundaries exist. I know there are ADEA issues for example, and I would not use groups that reference age in any way. I wonder about 401(a)(26), CODA issues, individual groups, etc. Maybe they are non-issues; I don't know for sure. I would not want to put in 50 of these this year to find out that the IRS has an issue with a particular design issue once the FDL's application responses start arriving.

What about defining allocation groups based upon service; any issues there? How about an allocation rate of x% of pay but not more than $y. Any problem with that? Why isn't it an option on Corbel's checklist?

Are one person allocation groups a problem? How about 0% groups (other than the issues applicable to dc plans as well and the obvious 401(a)(26) issues)?

I get the sense that the IRS will address this stuff sooner rather than later before the abuse starts. But where are the lines in the interim?

[FAPInJax]

Designing plans is always a question of what you feel comfortable arguing with the IRS. There is really no magic to the design of a cash balance - it is pretty much open to whatever you could do for a DC plan (think of the cash balance plan as a DC plan that just happens to require actuarial certification <G>). Therefore, have a ball and do anything you wish (feel comfortable with) and the nondiscrimination testing will tell whether it works.

[John Feldt ERPA CPC QPA]

I agree with FAPInJax. The questions you pose: if you believe these groupings work in a DC plan, then the same rules apply for the DB plan in that regard. Instead of allocation groups, think of them as benefit accrual rate groups.

So, for example, if you are comfortable using shoe size to determine allocation groups in a DC plan, then by all means, use the same criteria to create differing benefit accrual rate groups in the DB plan!

[JAY21]

Any opinions on whether using a DC like multiple group structure still passes age discrimination until the 'similarly situated' safe-harbor rule ? I believe the proposed regs added the 'under the same formula' to the test for similarly situated, which I took to make it easier to pass (similarly situated under the same formula) but I'd be interested in others comments.

[SoCalActuary]

Don't tell the lawyers, but DC plans also have age discrimination issues. Now that's our little secret, so don't tell.

They just don't have the high visibility since many of them are small plans.