Jump to content

Unit Accrual vs. Fractional Accrual Formulas


Recommended Posts

I have mostly small plans with one owner that we try to maximize, while minimizing the other participants. It's late in the day and I may be loopy, but I am wondering when a safe harbor flat benefit formula using fractional accrual can be worse for this type of client versus a unit accrual plan.

A benefit formula is designed to get the owner at a certain benefit level. Mathematically, each person who would have less than 25 YOP at NRD would have the same benefit under a flat benefit formula as if the plan were designed with a unit accrual formula. However, those with greater than 25 YOP at NRD would accrue LESS of a benefit under a flat benefit formula.

Simple math but this example might illustrate it better.

Flat benefit: 150% of FAP reduced for YOP at NRD < 25. Assume owner would have 15 YOP at NRD and comp is $100,000.

Proj Benefit = $100,000 * 150% * 15/25 = $90,000

AB 5 years into the plan = $90,000 * 5/15 = $30,000

Unit Accrual = 6% of compensation (designed to produce same benefit at retirement as above)

AB 5 years into the plan = $100,000 * 0.06 * 5 = $30,000

Say a person has 30 YOP at NRD:

Flat:

Proj Benefit = $100,000 * 150% * 25/25 = $150,000

AB 5 years into the plan = $150,000 * 5/30 = $25,000 (thus, less than the 30K produced by the unit accrual formula)

So, again, in a situation with a certain benefit being targeted to a person or persons who would have less than 25 YOP at NRD, it would seem a plan in which there are other employees with potential for YOP > 25 at NRD would be best designed with a fractional accrual formula such as above. Thoughts? Calling Mike Preston....

"What's in the big salad?"

"Big lettuce, big carrots, tomatoes like volleyballs."

Link to comment
Share on other sites

Well, if we start as a premise that the plan will end when the owner retires (presumably prior to 25 years), going with the 25 year flat reduction design will produce a smaller benefit than a unit accrual approach. The "downside" is that the funded projected benefit in this case would obviously be larger for your population resulting in higher costs going in (in fact it would be larger for anyone with more projected accrual service than your owner, assuming that if you took the unit approach you limited credited service to say, the amount of service your owner might have at retirement). However, this may not necessarily be a bad thing as the front loading inherent in unit accrual, especially if there is a real disparity between the owner's service and other participants, may mean that when your owner retires you may be looking at an underfunded plan.

Link to comment
Share on other sites

If using a safe harbor formula, best result is usually with the step-rate unit credit formula with fractional accrual (1.401a4-3b4iC1 safe harbor). This can even yield a better result than a general tested formula in some cases. For reasons I don't get, this is not used very often in small plans, even when it is really a no-brainer that it is the best safe harbor for a max-min solution (e.g., older owner and younger ees).

Link to comment
Share on other sites

David, my sentiments exactly.

Mwyatt,

"The "downside" is that the funded projected benefit in this case would obviously be larger for your population resulting in higher costs going in (in fact it would be larger for anyone with more projected accrual service than your owner, assuming that if you took the unit approach you limited credited service to say, the amount of service your owner might have at retirement)."

Are you speaking from a funding aspect? You could always have limited the unit accrual formula to a maximum of 25 years, so the projected benefit would be equal using both types of formulas. Also, you could still fund the plan using unit credit.

"However, this may not necessarily be a bad thing as the front loading inherent in unit accrual, especially if there is a real disparity between the owner's service and other participants, may mean that when your owner retires you may be looking at an underfunded plan."

I didn't understand this. Why is unit accrual front-loaded?

"What's in the big salad?"

"Big lettuce, big carrots, tomatoes like volleyballs."

Link to comment
Share on other sites

Unit accrual is front-loaded in the sense that it is the same percentage for every participant in each year, while flat benefit accrual for those with more than 25 projected years (or more than X/.7 projected years in the case of the design based safe harbor - which is usually even better than the safe harbor flat benefit accrual using a minimum of 25 years) will have differing accrual rates for typically HCE's, i.e., higher.

David, the problem with mutliple tier benefit formulas is that most actuarial calculation systems don't have the options available to calculate the benefits automatically. I don't know whether you meant to imply this or not, but I usually use 3 tiers, not 2 tiers in the benefit formulas designed under 3b4ic1. Works even better.

Link to comment
Share on other sites

It depends on what type of formula you want to implement (it seems like everybody has their own favorite). But how about 3.35% per year up to 10 years, 1.675% per year over 10 up to 20 and 2.5125% per year over 20 up to 33, all fractionally accrued.

Link to comment
Share on other sites

I believe the accrual rules allow you to use Participation or Service. Further, the only restriction on past service would be that imposted by 401(a)(4). I can see situations where past service might be acceptable. I can see situations where past service might not be a good idea.

Link to comment
Share on other sites

Actually, with a small plan that obviously has a limited life span, granting past service for accrual/benefit purposes may not be the way to go. Creating an underfunded plan from the get go isn't a great way to start out, unless the only meaningful past service would be granted to the owner. Just have to look at the situation and consider the projected scenario when your owner will be winding the plan down, not necessarily at the retirement of all parties involved.

Link to comment
Share on other sites

I would avoid past service with a 1-person plan - it can prevent zeroing the contribution if benefits are frozen, and if ees are added later and the plan becomes insured by the PBGC, the premiums can be enormous. I like to use past service only as leverage for the common max/min poblem with owners/employees.

Link to comment
Share on other sites

I don't disagree with the results of what you are protecting against, David, but I have to say that there are a number of situations where increasing the savings rate outweighs those concerns.

Link to comment
Share on other sites

  • 2 weeks later...

David, I agree with you about not granting the PS and creating PS liability and problem down the road. As to the OBRA FFL, I can't remember ever having to grant more than 1 yr PS to overcome the FFL limitation.

I come across plans where PS credits, althogh limited to 5 yrs, are as much as the future projected svc credits and wham, too much accrued and can't get zero cost even with benefit accruals frozen!

Link to comment
Share on other sites

The message which initiated this thread prompts an observation/question:

Consider the flat formula: 150% of FAP reduced for YOP at NRD < 25. Consider an owner with comp of $100k and 25 YOP at NRD.

Proj Ben = $100k * 150% * 25/25 = $150k and AB 5 years into the plan = $150k * 5/25 = $30k.

Question:

Generally, producing high(er) “accrued” benefit for an owner is of no importance since the owner generally gets what’s left in the plan after non-owners are paid out.

Given this, and the fact that one cannot fund for more than 100% of Hi 3 (ignore for the moment the option of using salary scale to increase the projected for funding) wouldn’t it be better to apply the accrual fraction after applying S415 limit to the proj ben to reduce the accrued for the owner and the non-owners?

In the above example, owner’s accrued after 5 yop would be: 5/25 * (150% * 100k, limited to 100k by S415’s 100% Hi 3) = $20k – no loss in reality

Consider an employee with proj YOP = 25 and Comp of $30k.

Applying the accrual fraction before applying S415 to the Proj ben, AB after 5 YOP would be: 5/25 *1.5 *30k = $9k.

But applying the accrual fraction after applying S415 to the proj ben, AB after 5 YOP would be: 5/25 * (1.5*30k, limited to $30k by S415’s 100% Hi 3 limit) = $6k, a saving in real cost of providing benefits, namely, lump sums payable to participants.

A follow up question.

For the first time, just came across a “standardized” adoption agreement, which allows the option of applying the accrual fraction before or after applying S415 to the projected. But it states that if the latter method (apply accrual fraction after applying S415 to projected) is selected then the plan will be discriminatory under section 401(a)(4)!!?

a) Is this true? If so, where does it say in the Code, Regs, IRS opinion letters…?

b) I thought, in a “standardized” adoption agreement, all options were de facto non-discriminatory (as long as they were not inconsistent with other selected options)?

Link to comment
Share on other sites

It depends on what type of formula you want to implement (it seems like everybody has their own favorite).  But how about 3.35% per year up to 10 years, 1.675% per year over 10 up to 20 and 2.5125% per year over 20 up to 33, all fractionally accrued.

I don't mean to ignore fosfur's question (I don't know the answer), but I started the past service discussion, which we all seem to agree on. But I was trying to understand Mike's formula and how it might be advantageous to a client. I didn't see any way except if past service were being granted. Mike, care to elaborate?

Link to comment
Share on other sites

flosfur,it is really hard to comment on the second question, because all of the relevant language isn't available. However,for the first question, is there any reason why you want to limit the benefit for the owner but you selected a formula that was 150% in the first place? Wouldn't it have been easier to just use 100% reduced for YOP < 26 so that the issue doesn't come up?

With respect to the second question, if you implement the limits of 415 before applying the accrual fraction aren't you just defining the plan as providing a projected benefit of 100% of pay, accrued ratably? How would that satisfy the 401(a)(4) rules on using a 25 year period? Maybe I'm just not understanding what you are getting at.

Link to comment
Share on other sites

Andy, I think past service is irrelevant to the discussion. It is simply a matter of how many years a principal has until retirement relative to how many years the other employees have until normal retirement. The principal is typically older, with fewer years until retirement. The first tier is intended to get the principal to the targeted benefit. The second tier reduces required benefits for those that would reach retirement age within the period defined by the second period of service. This second tier participant is typically higher paid than other non-owners, is in the stage of their life where they are more concerned with current compensation than deferred compensation and appreciates the fact that if the employer has an aggregate salary-benefits package more is directed towards the salary side of things than the benefits side of things. The third tier are all the rest.

Link to comment
Share on other sites

But if the plan is to be terminated upon the attainment of the retirement age of the owner, the second and third tiers are irrelevant if accruals are based upon participation, correct? But that might not be true if past service were granted.

Link to comment
Share on other sites

The second and third tiers are extremely relevant. Remember that these formulas are used in connection with flat accruals. The limitations on the percentages are based on the 133% rule.

Link to comment
Share on other sites

  • 8 months later...

I'm hoping to resurrect this discussion because I must be too dense to get it.

If I understand Mike's formula correctly, I might get the following hypothetical accrual rates depending upon age at plan inception. Assume accruals are based on participation.

Age 55 3.35% per year (Maybe this is the owner)

Age 50 2.79%

Age 45-Age 32 2.5125%

Why instead wouldn't the plan be designed to provide for a flat benefit (or a unit benefit, e.g. 3.35% x YOP, max 10) accruing on years of participation, maximum 14. Under that scenario, assuming a 33.50% flat benefit, the owner would accrue 3.35% per year and everyone else would accrue 2.3929%.

Wouldn't that be more efficient? And wouldn't it meet the alternative flat benefit safe harbor since the average NHCE accrual rate is 71.4% (more than 70%) of the average HCE accrual rate. Or am I missing something.

Link to comment
Share on other sites

It is a trade off between perfection and complexity. Remember that a safe-harbor is, well, safe. While a design based safe-harbor is only safe in those years where the 70% test is met.

I don't disagree with your math, so it looks like your method would always be a bit more efficient.

But maybe I'm missing something, so if I come up with a better answer later, I'll post it.

Link to comment
Share on other sites

OK, I think I thought of it. Maybe. ;-)

Does your formula satisfy the rules for a design based safe harbor? There seems to be a disconnect between the number of years of accrual and the number of years over which it can be accrued. Didn't the IRS come out with something about 10 years ago that put that issue on the table? I used to refer to it as the "Norman Levinrad" issue, because he was the one most familiar with it (and had pointed it out to those who were interested).

What we are left with, if that formula doesn't satisfy the rules, is the 133% rule and there is where I think your formula fails.

Link to comment
Share on other sites

Thanks for the comments, Mike. If there is a problem with my approach then I'm not aware of it but would like to be. I have some stuff that he wrote so I will certainly look at that but if anyone knows or finds anything concrete I'd like to hear it. Thanks again.

Link to comment
Share on other sites

Mike, I pulled out some material composed by Norman Levinrad and Joan Gucciardi which in 1999 (if not still now) was part of the ASPA C-4 required reading. It is a reprint of a session apparently given in Newark 5/11/99, Dallas: 5/24/99, and San Francisco 7/10/99, "Section I Review of the Basics: Safe Harbor and General Testing for Defined Benefit Plans"

And under the sub heading "Flat Benefit Safe Harbors", then subheading "Alternate Flat Benefit Safe Harbor" it gives an example:

"Example. Data Corporation sponsors a db plan that provides a benefit formula of 70% of pay, reduced for less than 7 years of service. The plan uses the fractional method of accrual, considering all years of service. Consider the following two employees and their effective rates of accrual:

Employee Bob Boss (HCE) Ella Employee (NHCE)

Years of Service 7 10

Total Percentage Accrual 70% 70%

Rates of Accrual 10% 7%

Assuming that the company had no other eligible participants, this flat benefit safe harbor will be satisfied because the average accrual rate for NHCEs is at least 70% of the average accrual rate for HCEs. If Bob hired an employee with 20 years of projected service, this flat benefit safe harbor would fail.

Alternatively, the plan formula could be designed to limit accrual years to 10, thereby automatically satisfying the 70% threshhold."

Now that I think about it, this is where I came up with this approach to design, from this example. Do you think that subsequently they may have changed their position on the acceptability of doing this, or maybe you are just recalling Norman as advocating this approach?

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...