CHAPTER 8

Designing Benefit Offerings

Introduction

Talk to any prospective new employee and one of the first considerations on their mind is how much will they be compensated for their employment services. Compensation comes in various forms: wages, bonuses and commissions, overtime, and various other soft forms of compensation. For retirement planning purposes, what comprises “compensation” is very important for discrimination testing purposes.

What would you think if I tell you that there are ways to legally skew benefits in favor of either those who are considered highly compensated employees (HCEs) or those who are older employees? Nuts ... not entirely. You will learn about these options and more in this chapter.

Learning Goals

• Explain why accrued benefits are important in a defined-benefit (DB) plan.

• Explain why the definition of “compensation” is so important.

• Understand how an employer can use integration with social security, cross-testing, or age-weighting to provide either highly compensated or older employees an additional benefit allocation.

• Discuss the role of voluntary employee after-tax contributions and the rules that apply to them.

Nondiscrimination Requirements

In practice, executives get a lot of additional benefits. The key is to figure out a way to properly incentivize the executives and other exceptional employees while not minimizing the vital contributions from the pool of nonhighly compensated employees (NHCEs).

Retirement plans have great flexibility in designing contribution formulae, but they must adhere to nondiscrimination testing. One of the great challenges of plan design is to give the HCEs as much benefit as possible, while still passing the relevant nondiscrimination testing. Internal Revenue Code (IRC) 401(a)(4) is the guardian of NHCEs. It tests nondiscrimination with either benefits or contributions. If benefits are involved, then the benefits provided in a given year are tested and not the ultimate benefit to be received. Remember that it is important to distinguish between nondiscrimination testing and contribution limits [415(b) and (c)].

It is also important to distinguish between an “accrued” benefit and a projected benefit. An accrued benefit is the amount earned up to a given point in time. In a defined contribution (DC) plan, this would be the plan balance. In a DB plan, it would be the benefit earned based on the salary and tenure until the current day. A projected benefit is the estimated future value if the current trend continues and various assumptions prove true (actuarial assumptions, asset returns, and so on).

Another question that must be grappled with in considering non-discrimination testing is: when must a year of service be credited for an employee? For a DC plan, the answer is straightforward. An employee covered by a DC plan will be credited with one year of service after completing 1,000 hours of service in any calendar year. The answer is much more customizable for a DB plan. An employer offering a DB plan can select their own definition of a “year of service” as long as it meets three criteria. The first criterion is that the standard must be both reasonable and applied consistently to all employees. The second criterion is that the standard not be set so high that a typical worker in that industry might not qualify. The third criterion is that anyone who worked at least 1,000 hours must be given partial credit. A typical DB definition of one year of service could be 2,000 hours in one calendar year. That translates into 50 weeks of 40-hour work weeks. If this were the standard chosen, then a company has created a window from 1,000 to 2,000 hours where employees will need to be given a pro-rata partial benefit.

Companies can change their benefit formulae whenever they like with the caveat that only future benefits will be altered. Any benefits earned under the prior benefit formula must remain in place. This is known as the anticutback rule, because prior benefits earned cannot be taken away.

What Is Compensation?

Another component of employer-sponsored retirement planning that needs a consistent internal definition is “compensation.” Will the company only include wages? What about stock options, bonuses, and overtime? If they include bonuses, but exclude overtime, then this would be considered discriminatory because the HCEs are more prone to receive bonuses and stock options while the NHCEs typically receive overtime. Whatever their definition, compensation is capped at $265,000 (2016 limit) for benefit calculation. This cap enables the HCEs to receive more benefit than the NHCEs, but not the extreme HCEs.

Consider an employee who earned $500,000 in a given year. Their employer contributes 10 percent of compensation for all employees into a money-purchase pension plan (MPPP). This employee notices that they only received an allocation of $26,500 for this calendar year. They are frustrated that they did not receive $50,000 ($500,000 [compensation] × 10%). How could this be explained to them? The reason for their allocation being capped at $26,500 is that they have crossed the income cap of $265,000 (2016 limit). Their employer simply applied the 10 percent figure fixed for all employees to the income cap to factor the retirement plan allocation.

You have heard the term “safe harbor” in other chapters, but we will now begin to apply it to multiple plan types. In general, a safe harbor plan is an allocation scheme that if applied will avoid the need for nondiscrimination testing. A safe harbor plan will require a uniform retirement age applied to all employees and a consistent vesting schedule. The employer cannot have different standards for the HCEs and NHCEs.

Safe harbor designs can offer one of the three basic templates. The first template is a level percentage of compensation. The second template is a level dollar amount. You are about to learn about the third template, which is a plan that is integrated with social security.

Integration With Social Security

As previously mentioned, the third safe harbor template is a plan integrated with social security. These are sometimes simply called “integrated” plans. The overshadowing concept is that the employers can contribute an additional amount for employees whose income is higher than an internally selected “integration level.” The integration level is usually set at or near the taxable wage base (TWB), which is established by the Social Security Administration. Essentially, the Social Security Administration says that they will only tax (social security tax) an employee’s wages up to $118,500 (2016 limit). Any money that an employee earns above that number does not result in any social security taxes being withheld from their gross pay or the resultant employer match. This limits the amount of money being paid into the government’s retirement savings program for all workers. In this sense, the social security system discriminates against the HCEs.

Both MPPPs and profit-sharing retirement plans (PSRPs) can use an integrated formula. The caveat is that if the employer offers both an MPPP and a PSRP, then they can only integrate one of the two plans. The other must remain nonintegrated.

Let’s assume that a company establishes an integrated PSRP for its employees with the integration level equal to the TWB ($118,500 for 2016). The way that the integration process works is that any employee (typically only the HCEs) who has earnings above the integration level can receive an additional contribution by the employer.

An employee subject to an integrated plan can receive an additional contribution equal to as much as 5.7 percent of their compensation! This is huge for the HCEs! There are always caveats .... First, the maximum additional compensation to which the 5.7 percent could be applied is capped at $265,000 (2016 limit). The maximum additional contribution that an HCE could receive in 2015 is $8,350.50 ([$265,000 - $118,500] × 5.7%). Another caveat is that the figure of 5.7 percent only applies if the employer already contributes at least 5.7 percent for all other employees. If the employer only contributed 5 percent in an MPPP, then their additional contribution under an integrated MPPP would be limited to 5 percent. The other caveat relates to the integration level chosen by the company. If the company chooses an integration level equal to the TWB ($118,500 for 2016), then the full 5.7 percent can apply subject to the previous caveats mentioned. If, however, they set their internal integration level below the TWB, then Table 8.1 provides the maximum additional employer contribution at various levels of integration levels.

Table 8.1 Schedule of the integration levels and the maximum additional contributions allowed

You will notice the sliding scale of available maximum additional contributions for various integration levels shown in Table 8.1. You might be wondering why at the lowest integration level the percentage jumps back up to 5.7 percent. The reason is because at this low value of an integration level, the NHCEs will likely be included as well and therefore, discrimination is not a concern.

Consider two different examples. First is Susan, who is 47 years old and earns $280,000 per year. She contributes 5 percent of her salary with a 2.5 percent employer matching contribution. Her company uses an integration level of $115,000. What is her total contribution for 2016? She will receive a total contribution $24,375.00 into her account. She is able to contribute $14,000 (5% × $280,000). The employer match is 2.5 percent, but it is limited by the compensation cap of $265,000. So, her employer contribution is $6,625 (2.5% × $265,000). The final component is the employer’s integration contribution, which is $3,750. The additional integration contribution is capped at 2.5 percent even though the integration level is 97 percent of the TWB. It is limited because the employer’s matching contribution is lower than the integration level table would otherwise suggest. The 2.5 percent integration contribution is applied to $150,000 ($265,000 - $115,000).

Another example is the PSRP used by Steve’s employer. Steve is 49 years old and he also earns $280,000 per year. Based on their allocation formula, his employer will contribute 12 percent of his salary into his profit-sharing plan. Nice deal for Steve! The company has an integration level of $110,000. How much contribution will Steve receive into his account? Steve’s PSRP contribution is $31,800 (12% × $265,000 [compensation cap for 2016]). The integration level is 92.8 percent (110,000/118,500) of the TWB, so the employer can contribute another 5.4 percent as an integration contribution. This 5.4 percent is applied to $155,000 ($265,000 - $110,000). That means an extra $8,370 as an integration contribution for Steve.

Cross-Testing

If an employer does not use a safe harbor plan, then they still have two other legitimate ways that benefits might be skewed to the older employees who are also often HCEs. The first method is called cross-testing, which favors older workers through an annuity simulation process.

Cross-testing is a means of testing benefits within a DC plan. An actuary will convert all employee plan balances into equivalent life annuities and then test the hypothetical annuity payments for nondiscrimination. You will not need to perform this test, but you do need to conceptually understand it.

This process will benefit all older workers because any underfunding revealed through cross-testing can be retroactively fixed by the company making an additional contribution. Typically, the older workers will be the HCEs, but that is not always the case. Theoretically, there could be some NHCEs who also fall in the pool of older workers.

Plan costs for a cross-tested plan will be much higher than other plan types because of the extensive actuarial efforts required and annual testing that must be done to ensure compliance.

A cross-tested plan also has a 5 percent gateway, which means that at least 5 percent must be contributed for all NHCEs. In general, cross-testing is a good and legal way to skew contributions in favor of older employees, but it does come with higher administrative costs.

Age-Weighting

A much more complicated cousin to cross-testing is called age-weighting. In an age-weighted plan, the benefits must be structured so that everyone receives the same rate of benefit accruals. You do not need to calculate an age-weighted benefit on your own, but you will be required to understand how this works conceptually. It will be the easiest to explain the concept of age-weighting with an example.

Consider a plan with three participants. Jen, age 50, earns $150,000. John, age 35, earns $30,000. Stephanie, age 28, also earns $30,000. Jen is a business owner, and she wants to contribute a total of $75,000 into an age-weighted plan in 2016. This just happens to be the amount that she had available to contribute. It does not relate to the 415(c) contribution limit.

The first step is to convert all three employees to the same rate of benefit accruals. We will use a rate of 1 percent of each employee’s monthly pay. So, Jen’s 1 percent benefit accrual equals $125 ([$150,000/12 months] × 1%). As both John and Stephanie have the same gross compensation, they will both have the same 1 percent benefit accrual of $25 ([$30,000/12 months] × 1%).

The second step is for an actuary to use a government mortality table to find the current contribution equal to $1 of benefit. Since you will not need to calculate this yourself, the table value of $95.38 will be a given. We then apply this table’s value to each person’s benefit accrual to determine their future dollar need. Jen will need $11,922 ($95.38 × $125), and both John and Stephanie will each need $2,384 ($95.38 × $25).

If we assume that plan assets will grow near the historical average of 8.5 percent, then we can calculate the current dollar amount required to provide the needed future dollar amount. Susan will need $3,506 of the current contribution to yield $11,922 at 8.5 percent interest over 15 years (her time until retirement). This is a simple time value of money application. John will need $206 of the current contribution to yield $2,384 at 8.5 percent after his 30-year expected remaining working life. Stephanie will only need $117 of the current contribution because she has 37 years until reaching the normal retirement age of 65.

Now, we will calculate each participant’s age-weighted required percentage allocation of the planned $75,000 total contribution. Jen should receive 91.57 percent of the contribution because her current contribution needed of $3,506 is 91.57 percent of the total $3,829 ($3,506 + $206 + $117) required. John should receive 5.39 percent, while Stephanie should receive 3.04 percent. One caveat with this process (there are always caveats!) is that the NHCEs must receive at least 5 percent allocation. So, we would adjust Jen’s percentage down to 89.61 percent and Stephanie’s percentage up to 5.0 percent. Jen will receive a dollar contribution of $67,207.50 ($75,000 × 89.61%), while John will receive $4,042.50 and Stephanie will receive $3,750.00.

That was a lot of work! Because of the level of work required to calculate such a simple three-person age-weight plan and because someone in Stephanie’s situation might be frustrated that she earns exactly the same amount as John, but gets a smaller retirement plan allocation, most employers simply go with a cross-tested plan if they want to skew benefits in favor of the older workers.

Common Choices for Various Plan Types

You have now learned about safe harbor options, integrated options, cross-testing, and age-weighting. Which combinations are most widely used in the various plan types?

Profit-sharing plans and their various offshoots typically stick with a safe harbor design. They might choose a level percentage or a plan integrated with social security. They are able to use either a cross-tested or age-weighted model, but the safe harbor plan is the most common in practice.

401(k)s all must apply either the actual deferral percentage (ADP) test or the actual contribution percentage (ACP) test depending on whether or not the employer offers matching contributions. Because they are a profit-sharing-like plan, they can also offer the same design choices that apply to the profit-sharing plans. Safe harbor is also the most common design type in practice.

MPPPs, like profit-sharing plans, have access to everything, but employers usually just stay with a safe harbor plan. If they wanted the complexity of the other plan design types, then they would instead offer a profit-sharing plan to capture the benefit of contribution flexibility.

403(b) plans need to apply the ACP test if employer matching is involved. However, they avoid ADP testing. They are also profit-sharing-like plans and so, they can access safe harbors, integrated plans, cross-tested plans, or age-weighted plans. They also typically stick with safe harbors.

The only option for a simplified employee pension (SEP) is a safe harbor. Cross-testing and age-weighting are not permitted. A savings incentive match plans for employees (SIMPLE) also has specific rules, which were discussed in Chapter 6.

Target-benefit plans are not widely used in general. Because they are so close to a straight DB plan, it is usually the DB plan that is chosen if a company wants this type of plan. DB-type plans are being used less and less because of the unfavorable risk profile for the employer. If a target-benefit plan were being chosen, then they would not want to use age-weighting if there are any older NHCEs. This would not have the desired effect.

A straight DB plan has its own set of choices. A DB plan can use a safe harbor design that offers either the same dollar benefit to all employees or the same percentage of the final average compensation (FAC) for all employees. These two options both assume that the employer offers a uniform retirement age. It would not be appropriate for executives to be permitted to retire at 60 while the rank-and-file employees are required to work until age 65. DB plans are permitted to integrate their benefits with social security, but they have a complex calculation that involves a different calculation for “covered compensation” instead of the TWB.

Voluntary After-Tax Contributions

Up until now, we have focused on employees deferring a portion of their gross salary and saving it for retirement. These deferrals are known as pretax savings. Employees are also permitted to make voluntary after-tax contributions within all qualified plans. These after-tax contributions are similar to taking money out of their savings account and investing in a mutual fund or exchange-traded fund (ETF). An employee might do this through their employer because it is easy, it creates natural discipline, and they like the investment choices offered within their retirement plan. After-tax contributions are tracked in a separate “account” within their retirement account although the funds are often comingled with the pretax funds as well. This means that the retirement account custodian will keep a track of any voluntary after-tax contributions and provide a running total on all account statements.

It is the most common to see after-tax contributions in a 401(k) plan. Employees might elect to save this way because after-tax contributions can be accessed anytime without the restrictions imposed on the pretax contributions. The ACP test for nondiscrimination compliance will include all employee pretax deferrals, all employer contributions, and all employee after-tax contributions.

Discussion Questions

1. A friend from your college days has just been told by their employer that the economic conditions necessitate that they will be losing their access to an employer-sponsored retirement plan. Your friend tells you that he remembers reading somewhere that the employer cannot alter any projected benefits within their DB plan. Your friend recalls that you took a class on retirement planning and asks if you know anything about this. What would you tell them?

2. A different friend tells you that their employer offers an MPPP, which is integrated with social security. They tell you that they are planning on not participating in the integration portion of the plan because they do not want to sacrifice any money from their take-home pay as a contribution. What advice would you give them?

3. An employer contacts you to provide advice about what to do with their employer-sponsored retirement plans. They have an MPPP, and they also have a profit-sharing plan to allow for contribution flexibility. They want to give their HCEs an extra incentive to stay with the company and so, they plan to integrate both plans with social security. What would you need to communicate to them?

4. An employer contributes 4.25 percent of the total compensation for each employee into an MPPP that is integrated with social security. They use the TWB as the integration level. A certain HCE who grosses $500,000 noticed that they had an excess contribution of $6,226.25 into their account. They were expecting a much larger number. How would you explain this number to them?

5. If an employer contributes 6.0 percent of the total compensation for each employee, then how much more could be contributed for the employees who earn more than the integration level, which is set at 75 percent of the TWB?

6. How can cross-testing be used to skew the employer’s contributions in favor of the older employees?

7. Does the inclusion of older NHCEs limit the ability of an age-weighted contribution formula to skew the benefits in favor of the HCEs?

8. A small business owner approaches you with an interest to integrate their SIMPLE plan with social security so that the company’s HCEs are realizing a greater value. What comments should you make?

9. A company comes to you for advice after learning that they have not passed the mandatory coverage testing in their 401(k) plan. Upon inspection of their plan, you learn that the employer offers matching contributions and that many of the executives also make voluntary after-tax contributions. What would you say to this employer?