A life insurance illustration can show the premium, death benefit and projected retirement income in considerable detail while leaving a consequential design decision almost unnoticed. Somewhere in the paperwork, the policy is identified as using the Guideline Premium Test, or GPT, or the Cash Value Accumulation Test, or CVAT. Many policies have the option to go either way and a choice has to be made. Few policy owners have any understanding this choice is being made for them, let alone the effects of the choice.
That selection can affect how much premium the policy will accept, how much death benefit it must maintain and the insurance charges assessed as cash value grows. For an advisor evaluating a policy, understanding the selection helps explain why two designs with similar funding can produce different results.
GPT is often a useful starting point for universal life policies designed primarily for accumulation and supplemental retirement income.

CVAT is common in traditional whole life and can offer useful funding flexibility in universal life. Neither is universally superior. The appropriate choice depends on what the client is trying to accomplish, the funding schedule and intended management of the policy. There are seemingly innumerable variables when building a life insurance policy and they affect how the policy performs under each test.
Two ways to qualify as life insurance
Section 7702 of the Internal Revenue Code establishes two routes for a contract to qualify as life insurance for federal tax purposes. One is CVAT. The other combines guideline premium requirements with a cash value corridor, generally referred to together as GPT. Both constrain the relationship between insurance protection and money accumulated inside the contract.
The simplest explanation is that GPT limits premium funding and requires a minimum death benefit relative to cash value. CVAT limits cash value relative to the actuarial value of future policy benefits. The insurer performs the technical calculations; the advisor needs to understand their effect on the proposed strategy.
These tests do not establish whether the policy is suitable, competitively priced or likely to accomplish the client’s objectives. Tax qualification is one requirement in the analysis. A policy can satisfy its tax test while being expensive, poorly designed or dependent on optimistic assumptions.
________________________________________________
The Guideline Premium Test focuses on how much money you put in.
Its basic message is: “For this amount of life insurance, you can pay only so much premium.”
The Cash Value Accumulation Test focuses on how much money builds up inside.
Its basic message is: “For this amount of cash value, you must maintain enough life insurance.”
________________________________________________
GPT controls funding and the required death benefit
Under GPT, the cumulative premium ceiling is generally the greater of the guideline single premium or the accumulated guideline level premiums. These are tax calculations, not a payment schedule the owner must follow. The cash value corridor separately establishes an age-based minimum relationship between death benefit and cash value. As the insured gets older the policy can have a higher ratio of cash value to death benefit, where the cash value and death benefit can be the same at age 95 resulting in no mortality expenses to the policy.
A simplified funding example makes the distinction clear. Assume the insurer confirms that a policy can accept $10,000 in cumulative premiums through the current year. If the owner has paid $90,000, the remaining room is generally $10,000. Whether the policy needs additional premium to remain adequately funded is a different question.
The corridor can also change the actual death benefit. A policy issued with a level death benefit does not necessarily keep that benefit level forever. Sufficient cash value growth can force an increase in death benefit to maintain compliance.
For an accumulation strategy, this creates a practical balance. The advisor wants adequate room for the planned funding while avoiding insurance protection that adds cost without serving the client’s objectives. The lowest initial face amount is only one part of that calculation.
CVAT connects accumulated value to insurance protection
CVAT requires that cash value not exceed the actuarially calculated net single premium needed at that time to fund future benefits. It has no separate GPT cumulative premium ceiling. But the required relationship between cash value and insurance protection still limits the design.
In flexible-premium products, CVAT may allow funding that GPT would restrict. The tradeoff can emerge later: CVAT often requires a higher minimum death benefit as cash value increases. One carrier’s published comparison explains that CVAT may permit more premium flexibility while producing higher death benefits and potentially higher insurance charges. The differences are policy-specific.
Referencing higher premium limits requires context. It does not promise unlimited funding into a fixed amount of insurance. Carrier restrictions and the policy’s tax calculations still apply, including the separate rules governing modified endowment contracts (MEC).
Why GPT often fits accumulation and retirement income
An accumulation-oriented indexed universal life (IUL) or variable universal life (VUL) design seeks to retain as much of each premium as practical after policy expenses, subject to the insurance need and tax requirements. A lower required death benefit can help by limiting the amount of insurance at risk.
In simplified terms, the net amount at risk (NAR) is the difference between the death benefit and the policy value. Contract calculations may include adjustments, but the economic point is straightforward: requiring the insurer to provide more protection above the policy value can increase mortality charges.
Consider a hypothetical policy with $1 million of cash value. Assume one design requires a $1.2 million death benefit and another requires $1.35 million. Those figures illustrate a possible design difference; they are not a universal comparison of GPT and CVAT factors. The second design has approximately $150,000 more insurance at risk. Its additional annual cost depends on the insured’s age, underwriting class and contractual charges.
If the client values that additional protection, it may serve a legitimate purpose. If the primary objective is retirement income, the advisor should determine how carrying it affects the plan. Higher insurance charges can leave less value available to support withdrawals and loans.
This is the practical basis for considering GPT first in many accumulation cases. The conclusion should still come from comparing actual designs. Early funding requirements, planned benefit changes or product features can outweigh a later advantage in insurance costs.
Where CVAT can be useful
Traditional whole life commonly uses CVAT because the test fits the product’s contractual pattern of premiums, cash values and benefits. The Society of Actuaries identifies CVAT as better suited to conventional whole life and GPT as more appropriate for flexible-premium products such as universal life. However, as we’ll get to, there aren’t hard and fast rules. An owner generally is not selecting between the tests in whole life as they might in a universal life application.
That is an important exception to the shorthand that accumulation means GPT. A whole life policy designed for cash accumulation, including a design using paid-up additions, will generally still use CVAT. Product structure comes first.
CVAT can also merit comparison when substantial funding is concentrated early or a client anticipates irregular deposits. It may accommodate a funding pattern that is awkward under GPT. That flexibility must be evaluated alongside the required death benefit and the intended tax treatment of future distributions.
For legacy planning, a higher death benefit as cash value grows may be attractive. But a larger illustrated benefit does not establish that CVAT delivers better value. Compare the death benefit produced by the same premium outlay, the costs of maintaining it and the results under less favorable performance assumptions.

It’s important to understand that many insurance design decisions are a form of horse trading. It’s generally not possible to have your cake and eat it too. There is cash value vs. death benefit. Early values vs. later values. Guaranteed vs. non-guaranteed. An increasing death benefit policy with a higher cash value to death benefit ratio may push the death benefit up per the corridor rules and result in higher mortality charges than another policy design that starts with a higher level death benefit that remains level for some time and results in a small spread and lower mortality charges. It’s dangerous to take anything for granted.
Sometimes the performance of one policy design is better in the early years and later years while the other design projects better performance during the midterm years. At times it can be counterintuitive which test works better for the goal at hand. Sometimes one looks better all around for decades, projecting higher cash value and death benefit. Then, eventually, it flips. Maybe not until after age 80, but when it does it can change significantly. This would be meaningful if, on one hand the insured has a family history where males seldom live into their eighties, or the on the other hand, the insured has a family history where living well into their nineties and occasionally beyond age 100 is the norm.
MEC status is a separate decision
Neither GPT nor CVAT determines, by itself, whether a policy is a modified endowment contract. Section 7702A establishes the separate seven-pay test. Broadly, it compares cumulative premiums during the testing period with calculated cumulative seven-pay limits. Benefit reductions and material changes can affect that testing.
________________________________________________
The MEC Test
“What are the premium and death benefit parameters to qualify distributions as tax free?”
The GPT and CVAT Tests
“What are the premium and death benefit parameters to qualify the policy as life insurance?”
________________________________________________
A policy can qualify as life insurance under CVAT while becoming a MEC because it was funded too rapidly. A large single payment is particularly important to evaluate. CVAT’s funding flexibility does not exempt the policy from MEC testing.
A MEC generally retains life insurance treatment for death benefits, subject to the usual rules, but its lifetime distributions are treated differently. Distributions generally come from gain first, and loans are generally treated as distributions. Taxable amounts may also incur a 10% additional tax before age 59½ unless an exception applies. It’s important to understand this also applies when the policy cash value is used as collateral.
For a client whose primary goal is leaving a death benefit, accepting MEC status may be deliberate. For a client expecting substantial lifetime access through loans, it can materially change the strategy. “Maximum premium” should always be accompanied by an explanation of which limit is being discussed.
The seven-pay-test is a cumulative calculation, if the test tells us $700,000 is the limit, that means $100,000 per year. If the premium in the first year is $90,000, the premium in the second year can be $110,000. However, if the plan is to pay $90,000 in the second year, $110,000 cannot go in the first year. If you wanted to put in $700,000 over two years rather than 7 years, the death benefit necessary to accept it would be increased substantially. Interestingly, it would be increased the exact same and stay the same indefinitely whether based on GPT or CVAT.
However, staying with the $100,000 per year example, under GPT, whether we calculated for $100,000 for 7 years, 10 years or 50 years, the initial death benefit is the same. Under CVAT the same scenarios would result in three different death benefits.
Policy changes can alter the comparison
A plan to fund aggressively and later reduce the death benefit should be evaluated before the policy is issued. A face-amount decrease under GPT may require a return of premium or distribution of accumulated value, while CVAT may avoid that particular funding-limit problem. That doesn’t mean every CVAT reduction is permissible or free of tax consequences.
Benefit changes require carrier analysis of the applicable tax tests. The desired transaction should be modeled explicitly rather than assumed from an illustration that never implements it.
Product features also matter. Some policies restrict particular riders to one test. For example, one carrier publishes that certain policy series offer an over-loan protection rider only with GPT. Availability differs by product and series. A test that appears attractive on funding alone may exclude a feature central to the intended income strategy.
Owners should also understand the permanence of the selection. It should therefore receive attention before the application is signed, not after the client wants to make a transaction the original design cannot accommodate.
What advisors should compare
A useful comparison begins with the same client objective and premium schedule. When both tests are available, request designs that implement the intended funding, death-benefit options, later reductions and distributions. Keep performance assumptions consistent so the test selection’s effect is visible.
Two comparisons may be necessary. Holding the initial face amount constant helps isolate differences between the tests. Separately optimizing each design for the client’s funding and insurance needs shows which complete strategy works better. Confusing those comparisons can produce a misleading recommendation.
For an income-oriented case, examine annual charges, cash value before distributions, required death benefits and sustainability under reduced crediting or investment returns. For a protection-oriented case, evaluate premiums needed to maintain the intended benefit and the strength of contractual guarantees. Compare guaranteed and non-guaranteed results separately.
Ask the carrier to confirm remaining premium capacity and MEC status, particularly before extra funding or a benefit reduction. Retain the rationale for the original selection and revisit the policy when the client’s objectives change.
GPT and CVAT are technical tax tests with practical consequences. An advisor does not need to reproduce the actuarial calculations to ask the essential question: How does this selection affect the funding, costs and future transactions this client actually intends? The answer belongs in the policy recommendation.
Playing with the numbers
When running some numbers we can see where the differences show up. We’ll assume a 40-year-old health male. The policy is variable universal life for $1 million with the premium paid every year for the policy to target cash value equal to death benefit at age 100. Whether we use the GPT or the CVAT, the premium and cash value is the same.
On the other hand, if I keep everything the identical but increase the premium modestly, the values continue to stay the same for decades until one test starts pushing up the death benefit slightly. The other test starts pushing it up one year later. The difference is negligible, but it’s there.
Let’s change it again. When I tell the system to target the cash value and death benefit to be the same at age 90 rather than age 100, the two tests result in a difference right out of the gate with the cash values different under the two tests from the earliest years. It’s still a small difference over time, but we see the testing kicking in much sooner.
Now for a different angle that focuses on cash value accumulation rather than death benefit. In this scenario we’re telling the system to calculate for the lowest death benefit for $10,000 of annual premium. For the GPT the death benefit is $808,000 and for CVAT, it’s $494,000. That’s meaningful. However, I left the death benefit option at level. This results in the $494,000 death benefit to project an increasing death benefit in year 18 with the $808,000 death benefit increasing in year 28. In year 10 already the lower death benefit has a 15% greater cash value.
Next, I’ll change it to increasing death benefit. GPT has a $238,000 initial death benefit with CVAT at $381,000, the opposite of the prior example regarding level death benefits.

This would be a more typical cash value accumulation model and we see GPT is more efficient. But as soon as we determine this we get another scenario where we’re doing a 1035 tax free exchange from an old policy with $250,000 into a new policy with no additional premium. We again calculate for minimum death benefit with each test.
Here we have GPT death benefit of $1,052,775 and a CVAT death benefit of $302,419, a remarkable difference. Since most of us are conditioned that more is more, the larger death benefit looks attractive. But if we engage the zen-like philosophy of “less is more”, notice what we get in exchange.

This is pretty dramatic. The spreads may not look huge at first glance, but if you took the percentage differences or the spread in returns and applied it to a money manager, you have the difference between a hero and a dog.
As you might guess, there are innumerable scenarios to run, but let’s look at another one. When I take the $10,000/year minimum death benefit example and assume that after year 10 the policy owner wants to ramp up premium and I increase the $10,000 to $50,000, the CVAT policy accepts it all and the cash value and death benefit project to grow dramatically. However, the GPT policy only accepts $13,500 per year, limiting flexibility. So, on one hand we may look to the GPT policy for more effective cash value growth, but we’re giving up future optionality.
Here’s one more curveball. A policy can be built to change from increasing death benefit to level death benefit in a given year (but not the other way around). A policy that is not changed from increasing to level eventually has a larger spread between cash value and death benefit than necessary resulting in excess mortality charges. It won’t be able to have cash value and death benefit be equal at age 95. This matters materially at the high mortality charges at older ages. Furthermore, there’s going to be a specific year when it’s the most efficient to change from one to the other.
It’s also a cliché in the market that a policy initially illustrated and sold based on a change from increasing to level doesn’t actually get changed 10 or 20 years down the road when it should because no one is paying attention. It’s not an overstatement that this can result in disaster. It’s one more reason to demand ongoing and meaningful policy management.
It can be confusing
This may have thoroughly confused you rather than provide any clarity. However, that’s kind of the point. We need to understand this can be complicated and needs to be better understood and more attention regarding the subtle details. In reality, essentially no policy owner knows there even are two tests, let along what they mean. Too many agents are in the same boat. The illustration systems default to one test or the other and don’t automatically change based on the policy build.
If we now see that different choices have consequential outcomes, shouldn’t much more attention be paid to policy construction? This entire article is about something a vast majority of policy owners don’t even know exits but dictates what they get. This can be a big deal.
Bill Boersma is a CLU, AEP and LIC and the founder and owner of OC Consulting Group. More information can be found at www.OC-LIC.com, www.BillBoersmaOnLifeInsurance.info or email at bill@oc-lic.com or call at 616-456-1000.
Securities offered through Valmark Securities, Inc. Member FINRA, SIPC. Investment Advisory Services offered through Valmark Advisers, Inc. a SEC Registered InvestmentAdvisor.
OC Consulting Group and its affiliates are separate entities from Valmark Securities, Inc. and Valmark Advisers, Inc.











