- The General Concept
- The Ground Rules
- Practical Uses for Cross-Testing
As the national economy
continues its recovery, more and more businesses are beginning to see their
financial situations improve to near pre-recession levels. Companies that
have not thought about making profit sharing contributions for years are
starting to consider their options.
Just as the economy as a whole or the circumstances of a particular
company change over time, companies should review their retirement plans to
make sure the design changes with them.
Companies that find themselves on solid footing may find themselves
thinking of making employer contributions to their retirement plans. Whether
the goal is to maximize benefits to the owners, reward employees, reduce tax
liability or some combination of all of these, the cross-tested plan design
is one worth considering.
Although there are a number of ways a company may choose to divide a
profit sharing contribution among the employees, there are three methods
that are commonly used.
Salary Proportional (a/k/a Pro Rata):
This method divides the contribution based on the proportion that each
individual participant’s compensation bears to the total compensation of all
eligible participants. It results in each person receiving a uniform
percentage of his or her pay.
Integrated (a/k/a Permitted Disparity):
This method considers that employees whose pay exceeds the taxable wage base
do not receive social security benefits on their total compensation and
allows those people to receive a larger profit sharing contribution to help
equalize the benefit.
Cross-Tested (a/k/a New Comparability):
This method allows employees to be divided into groups based on valid
business classifications, i.e., owners and employees, and provides different
levels of contribution to each group.
The first two methods are relatively straightforward and are considered
to be "safe harbor" allocation methods, meaning that they automatically
satisfy certain nondiscrimination requirements. However, with ease and safe
harbor status often comes limited flexibility.
The cross-tested method, on the other hand, provides a great deal of
flexibility but also comes with a few more rules to follow and must undergo
additional testing to ensure it complies with the nondiscrimination rules.
For companies that are willing to accept a little more complexity, new
comparability plans can be used to meet a number of business goals.
Cross-tested designs generally rely on the time value of money to allow
companies to maximize benefits to the owners who may have spent the earlier
parts of their careers reinvesting everything into growing the business.
Since they are closer to retirement, it takes a larger contribution to fund
an equivalent benefit than it does for someone who is just entering the
A simple example may help to illustrate. A company has two participants
in its plan; the owner (age 55) and an employee (age 35)—and it wants to
provide a retirement benefit of $100,000 to each one at age 65. Assuming
there are no investment gains, the owner would need a contribution of
$10,000 per year for 10 years to reach the target benefit, while an annual
contribution of $3,333 would get the employee to the goal.
Once you factor in an assumed interest rate, the spread gets even
greater. The actual calculations and tests are much more involved, but this
is the general concept.
Unlike a defined benefit plan in which the company would have to commit
to making those contributions each and every year, in a cross-tested profit
sharing plan, the company has the discretion to contribute more or less or
nothing at all each year.
There are several additional rules that apply to cross-tested plans.
As noted above, the plan must define the employee groups that are used to
allocate contributions. In the early days of this design, many plans would
specify groups based on company ownership, officer status, division, office
location, etc. Some often-seen combinations were owners and employees;
partners, associates and non-lawyers; doctors, nurses and staff; etc.
More recently it has become common for plans to specify that each
participant makes up his or her own group, providing maximum flexibility in
making contributions. While a law firm could still decide to contribute the
same amount for all non-lawyers, it could decide to contribute more or less
for certain employees as long as all of the other testing requirements are
To ensure that rank-and-file employees receive enough of a benefit
relative to the highly compensated employees or HCEs (generally the owners
and those earning more than $115,000 per year), the company must provide a
minimum gateway contribution to the non-HCEs. This is kind of like the cover
charge to get into the cross-testing club. In other words, it does not
guarantee the plan will pass the other nondiscrimination tests.
The amount of the gateway contribution is the lower of 5% of compensation
or one-third of the highest percentage allocated to any HCE. For example, if
the highest HCE allocation is 9% of pay, the gateway contribution to the
non-HCEs is 3%. Once the highest HCE contribution reaches 15%, however, the
gateway is capped at 5%.
For 401(k) plans that make a flat 3% of pay contribution to meet the safe
harbor rules, that safe harbor contribution actually counts toward the
gateway requirement if the company also decides to make a cross-tested
profit sharing contribution. In other words, assuming all other tests are
met, it may be possible for the sponsor of a safe harbor 401(k) plan to
contribute an additional 6% of pay on behalf of the owners (bringing the
total to 9%) without having to contribute anything more for the employees.
This is another nondiscrimination test the plan must pass. Essentially,
all of the contributions made on behalf of each employee (in some cases,
including 401(k) deferrals) are added together and converted to a benefit at
the plan’s retirement age using several factors taken from IRS tables. The
average benefit of the non-HCEs is then compared to the average benefit of
the HCEs to make sure they are within the appropriate range of each other.
Some plans will pass the test giving only the gateway contribution to the
employees and providing the maximum to the owners. Other plans will need to
provide additional contributions to some or all of their non-HCE
participants in order to increase the average benefit to a passing level.
Since this test is based on the demographics of the workforce, the
results are likely to change each year depending on the degree to which the
demographic composition fluctuates. Using a small medical practice as an
example, the addition of a new physician who is much younger than the other
doctors and maybe some of the longer-term staff could cause a plan that was
once passing with ease to fail.
Another common cause for extreme demographic shifts is when the child of
an owner comes to work for the company. Since children are generally
attributed their parent’s ownership, they will be considered HCEs even
though their actual pay might be very low. Companies anticipating such
changes should speak to their TPAs ahead of time to determine the impact to
the average benefits test and consider any design modifications that might
avoid a problem.
We have already discussed using this design as a means of maximizing the
benefits for owners or certain key individuals; however, there are other
situations when cross-testing can come in handy.
With the improving economy, some companies are also beginning to pay
employee bonuses again. But, along with the cost of the bonus itself comes
additional payroll taxes. By using a cross-tested plan design, a company
could make individualized profit sharing contributions to certain employees
without incurring the cost of the payroll taxes.
Not only does this option eliminate the extra payroll cost, it also helps
to address increasing concerns of employee retirement readiness that are
becoming more prevalent among companies. Recognizing that a bonus is meant
to be a reward, and many employees appreciate cash in hand more than a
contribution, some companies will split the "bonus" amount, contributing
half to the plan and paying out the other half in cash.
From time-to-time when a company removes certain investment options from
the menu, that change can trigger a surrender charge to all those invested
in the option being eliminated. This most often occurs in conjunction with a
change in service providers. Some companies facing this situation do not
want their participants to be harmed as a result of the change and would
like to "reimburse" them by contributing to the plan.
The challenge is that these types of charges are usually assessed
proportionately based on account balance; however, the money the company
deposits as a reimbursement must be allocated as a contribution. Plans that
provide for pro rata or integrated allocations would have to allocate the
reimbursement accordingly. By amending the plan to provide for a
cross-tested allocation with each participant in his or her group, the
company could target the contribution to those impacted by the surrender
It might not be possible to make everyone whole, but this option can
sometimes get very close. For example, to the extent any HCEs share in the
allocation, it could trigger the gateway requirement for all non-HCEs
(including those not affected), so it may be necessary to find another way
to compensate HCEs. In addition, some people who share in the surrender
charge will be former employees, and contributions can only be allocated to
those who are participants during the year of the contribution. While not a
perfect solution, it can be a step in the right direction.
A well-designed cross-tested plan can be a very effective tool for
satisfying a variety of company objectives, but it also comes with a few
more moving parts. As a result, it is even more important to work with a
knowledgeable TPA or consultant who will ask the right questions to
understand your goals and design a plan tailored to meet them.