The Easy Benefits Answer May be Compliance

I ran into an interesting situation the other day. I can't disclose enough of the details for reasons of confidentiality to use that as a case study. The point of the exercise, though, was that a client was looking for a loophole to escape the claws of what it was told is a horrible provision of the Internal Revenue Code. After analysis, however, I was able to determine that the loophole, as they put it, just doesn't work for them.

There, was a better and simpler solution, though. Just follow the law (without regard to the loophole).

While I can't use this week's example, it harkens back to the late 80s and early 90s (yes, I was doing similar work even then). Readers in their 50s or older may recall that the first Tax Reform Act of 1986 (TRA86) regulations to come out were the proposed 401(l) regulations that graced the pages of the Federal Register in late 1988. Lawyers, consultants, and actuaries (me among them) raced to digest those regulations and to be the first to tell their clients and prospects how to comply.

I digress for a moment to explain why this could have been so important. 401(l) explained how to integrate certain qualified retirement plans with Social Security in order for those plans to be exempt from the general test for nondiscrimination under Code Section 401(a)(4). Prior to TRA86, permissible integration was determined under Revenue Ruling 71-446, and whether or not a group of plans was discriminatory in nature was determined under the comparability rules of Revenue Ruling 81-202. What was different, though, was that if you had an integrated plan and you couldn't prove that it was properly integrated per RR 71-446, you risked your qualified status. Under TRA86, proper integration was more of a convenience.

Returning to the main plot, we all rushed out to our clients and many with defined benefit (DB) plans redesigned them to be properly integrated making them safe harbor. Those clients then needed to deal with transition rules under Notice 88-131. Some with good memories will remember racking our brains over choices between Options II and III and Alternative II-D.

Other sponsors made a different decision. They chose to not redesign their plans. That meant that they had to comply with the general test for nondiscrimination.

Well, for most, that turned out not to be such a big deal. It meant that every year or two years or three years, they had to engage their actuary to perform a test that was usually pretty easy to pass. What they got in return was a lack of disruption. For them, the easiest answer was not to find a way out of testing, but simply to pass a test that they could easily pass.

Returning finally to my initial point from this post, the client in question was trying to get out of a rule that counsel had told them was pretty horrible. And, it's true, it can be a horrible rule. But, in this specific case, the compliance burden associated with this rule will be minimal.

And, that means that what is perhaps the optimal solution will be simple.

Sometimes, the right answer has nothing to do with loopholes. Sometimes, the easiest (and best) answer is compliance.

Another Argument for Defined Benefit

I know, defined benefit (DB) plans are dead. Actually, while there aren't as many as there used to be, I'm going to give you one more argument why they make more sense as a retirement vehicle.

Yesterday, I wrote about managing the risk in active pension liabilities. Way back in 2010, I wrote about generally managing risks and noted that plan sponsors tend not to manage defined contribution plan risks. Most of those risks that I have considered have been financial risk. Today, I am going to focus on the intersection of financial risk and compliance risk and make a case to have a DB plan as your primary retirement vehicle rather than a 401(k) plan.

In the world of 2015, we see consolidation in many industries. We also see companies, often private, being gobbled up by private equity firms. Either of these actions will usually create a larger controlled group. And, people who focus on retirement plan compliance know that most retirement plan compliance testing must be done on a controlled group basis.

Before working to point out a solution, let me give you an example to help focus on the problem.

BPE is a big private equity firm. Their general approach to retirement plans (and other benefits) has been to ignore them and let each company do what it wants. But, as BPE get bigger, its controlled group gets more complex. Having multiple industries represented in its portfolio, BPE is ultimately the sponsor of all kinds of 401(k) plans. Their engineering company (EC) has an extremely generous 401(k) plan that matches 150% on the first 8% of pay that an employee defers. Their pork rinds company (PRC) has a 401(k) plan that matches 10 cents on the dollar on the first 2% of pay that an employee defers.

BPE never saw this as a problem. But, then one day, an inquisitive Principal (IP) at BPE was reading my blog (of all things) and came across this. He saw that BPE might have a compliance problem in its controlled group because of the disparate nature of its 401(k) plans.

Ring ring ring -- that's my phone as IP calls me. He wants to know how to fix the problem. He says that surely this problem can't be real. After an hour on the phone, we have inventoried all the plans at BPE and found that they have failed to satisfy various compliance tests (coverage under Code Section 410(b), for example) for several years.

IP has a solution though. He tells me that BPE will force some of its companies to retroactively cut the employer match in some of these more generous plans.

Bzzz!

You can't do that. In fact, if BPE were to choose to fall on its sword and approach the IRS for a negotiated retroactive solution, we would suspect that the IRS would only be receptive to increasing benefits for nonhighly compensated employees (NHCEs) in the less generous plans.

IP is not happy about this. PRC runs on very low margins, but because they make more pork rinds than any company in the world, they do throw off a lot of cash. However, increasing benefits would eliminate most of that free cash that is being generated. There is stunned silence on the other end of my phone.

While the story is fictitious, the gist of the scenario is not. I've seen this happen. By being a serial acquirer, companies run into compliance problems and with 401(k) plans not being the easiest to prove nondiscriminatory, either costs escalate or they get cut at the portfolio companies that tend to employ more higher paid individuals.

What sort of plan tests better? A few weeks ago, I wrote about some. Suppose BPE had a defined contribution looking cash balance plan. One of the nice things about these plans is that they test well. Designed properly, and proper design truly is a key, financial risk is manageable. And, with that as their primary plan, the secondary 401(k)s can be managed so that compliance there will no longer be an issue.

Unfortunately, the benefits world has been resistant to this whole concept. But, you have an open mind, don't you?

We need to talk.

Compliance or Policy?

This is going to be a fairly short post, I think. I want to talk a bit about what should be the greater influence on benefit program design -- compliance or policy?

Right now, and especially as PPACA (health care reform or ObamaCare, if you prefer) is starting to exert its influence, it seems that most benefit program design is structured to facilitate compliance with the myriad of laws that Congress has passed since ERISA was signed 38 years ago. Just think, you provide a big health care benefit, you pay a Cadillac Tax. Your NHCEs can't afford to defer to your 401(k), your HCEs don't get to benefit. You would like to provide your employees with retirement income, but you cannot stand the volatility in corporate cash flow and P&L of a defined benefit plan.

Seems wrong, doesn't it?

So, even to the extent that you have a policy that is governed by things like true long-term cost and what is right for your employees and your business, you are unable to implement that policy while being in compliance.

Seems wrong, doesn't it?

Of course, it's wrong, but the people who make the laws don't seem to get it. They don't believe in benefits policy. They don't believe in retirement policy. They believe in tax policy and gerrymandering the tax code to make it work, often at the expense of corporations (large and small) and their employees.

Seems wrong, doesn't it?

Private Equity Groups and Retirement Plan Compliance

They buy companies. They sell companies. That means, among other things, that at any given point in time, they own lots of companies.

What are they? They're private equity firms. And, in their normal mode of operation, many of them constitute a controlled group of corporations within the meaning of Internal Revenue Code Section 1563(a)(1).

So? Bully for them, right. Actually, if we look more closely, it may be far more complicated than that when it comes to retirement plans. In the most typical situation, each portfolio company will maintain one or more qualified retirement plans. So, the controlled group maintains lots of qualified retirement plans. And, because private equity firms usually allow each portfolio company to operate autonomously, the proverbial left hand may not know what the proverbial right hand is doing.

Usually, in my experience, they are mostly or entirely defined contribution (DC) plans. That's good, but let's think about the compliance issues. Each DC plan needs to satisfy the coverage rules of Code Section 410(b) on a controlled group basis. Each plan needs to satisfy the benefits, rights, and features rules of Treasury Regulation 1.401(a)(4)-4 on a controlled group basis. Each plan that is not a 401(k)/(m) plan (or 403(b)) needs to satisfy amounts testing under Treasury Regulation 1.401(a)(4)-2 (or -3 for a DB plan) on a controlled group basis. Each plan that is subject to 401(k)/(m) may wind up needing to be aggregated with another like plan in the controlled group for testing purposes.

I'm going to simplify this a little bit, but here is the way that we might look at the compliance situation for a group of DC plans in a controlled group.

• Count up the number of nonexcludible (generally all that are at least age 21, have at least 1 year of service, are not covered by a collective bargaining agreement, and are not nonresident aliens) highly compensated employees (HCEs) in the controlled group. Let's call this number CGH.
• Do the same thing for the nonhighly compensated employees (NHCEs) in the controlled group. Let's call this number CGNH.
• For each plan, determine the number of HCEs and NHCEs, separately who are covered by that plan. For plan 1, we will call those numbers P1H and P1NH. For plan 2, let's call them, P2H and P2NH. You get the picture.
• Now, take the ratio (P1NH/CGNH)/(P1H/CGH). Is it at least 70%? Are the similar ratios for all of the other plans also at least 70%? In my experience, the answer is often no. If the answer is yes, you are in good shape. If it's no, you may have problems.
• At this point, you'll need to consider other issues such as plan aggregation, current year versus prior year testing methods, safe harbor designs, qualified separate lines of business (QSLOBs) and other equally wonderful concepts. You probably need an expert.

We could go on for pages and pages here. These situations often get ugly. The private equity company probably never thought about this stuff. Suddenly, the plans maintained by the companies that they own may have compliance problems.

In reality, when dealing with this type of organization, I've seen this issue more often than not. It's not part of the buying decision, it's not part of ongoing process, and frankly, the people completing (and signing) Forms 5500 for the various plans probably don't even realize the problem exists.

I've seen it. It's usually ugly. I can help.

Beware! Is Your Nonqualified Deferred Compensation Plan Linked to a Qualified Plan

The regulations under Code Section 409A have some strange rules related to "linking" nonqualified and qualified retirement plans, or linking multiple nonqualified retirement plans together. Once upon a time, a fairly typical retirement benefit structure include these plans for executives:

• A qualified defined benefit plan
• An "excess plan" with the same formula as the qualified plan, but designed to provide for benefits above the 415 limit and on pay above the pay cap
• A SERP with a different, usually more generous formula, offset by benefits under the other 2 plans

Often, the SERP had different early retirement benefits and different or more subsidized optional forms of benefit available.

This structure can be particularly problematic under 409A (you can read about it in my article here: http://www.aon.com/attachments/jpm_409a.pdf ).

If you have such a structure, the IRS and Treasury is saying that you cannot fix the problem under the documentary correction programs in Notice 2010-6. Currently, the regulatory position is that linked nonqualified plans can be fixed, but nonqualified plans impermissibly linked to qualified plans cannot be fixed penalty free.

Any advice I would give would vary from situation to situation, but if you do have problematic plans, you will want to fix them sooner rather than later.

This blog does not provider legal, tax, or accounting advice which can only come from a qualified provider.