The math of life insurance pt. 1

The best life insurance advisors in Canada know this. They know that lapse-supported pricing is built into every permanent product they sell. They know that the consumer who understands this mechanism and acts on it is positioned to capture a structural advantage that exists nowhere else in the financial landscape. And they know that most of their clients have never been told about it.

That gap is a design failure. The compliance documentation stack was built to protect advisors and companies. The illustration was built to compare products. Neither was built to explain to a client how the pricing works — and what that means for every decision they make, from the day they sign to the day the claim is filed.

Every level-premium life insurance policy is front-loaded by design. The 45-year-old buying permanent coverage pays the same premium at 45 as at 75, even though the actual cost of insuring a 75-year-old is dramatically higher. The insurer collects excess premium in the early years to fund the shortfall in later years. The pricing logic is straightforward. What is less visible is the second variable built into that calculation.

Before any policy is priced, the actuary estimates what percentage of policyholders will exit before the death benefit is ever paid. This is called the lapse assumption. And it goes directly into the premium calculation. A higher expected lapse rate produces a lower premium. The anticipated profits from early exits effectively subsidize the cost of coverage for those who stay.

This is not a minor calibration. A Society of Actuaries analysis demonstrated that a 30-year term policy projected a profit of $103,000 under normal lapse assumptions — and a loss of $942,000 if nobody lapsed. That’s a million-dollar swing on a single policy, driven entirely by whether policyholders quit or stay.

Academic research from Wharton School economists Daniel Gottlieb and Kent Smetters documented the mechanism plainly: insurers earn substantial profits on policies that lapse and lose money on those that persist. Lapsing policyholders cross-subsidize households who keep their coverage.

The research also confirms this applies explicitly in Canada. In the early 2000s, A. David Pelletier, then executive vice-president of RGA Life Reinsurance Company, noted that Canadian insurers were accounting for higher projected future lapses to discount premiums and arrive at more competitive pricing — with the profits expected to come later when policyholders exited.

The consumer paying that competitive premium has no idea they’re doing so.

A lapse by any other name

Voluntary surrender is the most visible form of exit. It is not the only one. The pool is enhanced by every category of policy where premiums are collected and a death benefit is not paid.

That includes claims never filed because the beneficiary did not know the policy existed — the policy in the drawer in the basement that nobody thought to mention. It includes claims denied for misrepresentation. It includes reduced paid-up elections that gave the policyholder something, but far less than the full benefit they had been building toward.

The family that never finds the policy is economically identical to the family that cancelled. Every premium ever paid contributed to the pool. Nothing was collected in return. That contribution flows to the policyholders who stayed and whose beneficiaries knew exactly who to call.

This is where the advisor’s role extends beyond product selection and into something more fundamental. Telling a client which policy to buy is one conversation. Making sure their executor knows the policy exists, where it is and who to contact — that is a different conversation entirely, and it is one our industry does not have consistently enough.

Some advisors and financial planners argue that lapses hurt the pool — that early exits mean the insurer never recovers its upfront costs and a shrinking premium base earns less investment income for remaining policyholders.

That argument is true for early lapses. In roughly the first five years, an exit costs the pool more than it contributes. Acquisition costs — commission, underwriting, administration — have been paid out before the policy has generated any offsetting investment return. An early lapse generates a loss for both the insurance company and your client, not a windfall.

But the lapse assumption embedded in product pricing was never primarily about early exits. It is about the distribution of exits across the entire policy lifetime, and the financial weight of that distribution sits in the middle and later years. That’s when the insurer has been earning investment returns on accumulated premiums for a decade or more, when reserve releases on lapsed policies are largest and when the gap between what was assumed and what actually happened is most consequential.

The consumer who held for 15 years and then surrendered, the family that never found the policy, the policyholder who took a reduced benefit at year 20 — those exits fund the advantage. Staying through the early years is the price of admission. The benefit accumulates after that.

In 2020 I corresponded with Steve Krupicz, a fellow of the Canadian Institute of Actuaries with whom I worked closely for many years. His observation reframed how I think about this mechanism entirely.

Using the CIA 86-92 mortality table — the standard Canadian table used for calculating adjusted cost basis — a newly underwritten 45-year-old male non-smoker has a mortality rate in year one of 0.06%. Six one-hundredths of 1%. Current industry lapse assumptions run approximately 1–2% annually. The lapse rate is roughly 17 times higher than the death rate at that age.

Krupicz noted that this 45-year-old does not reach a 1% mortality rate under that table until age 64. Under more current mortality experience, the crossover happens at approximately 70.

So, for the first 20–25 years of a permanent policy, the insurer is not primarily managing mortality risk. They are primarily managing lapse risk. The premium the consumer pays reflects both assumptions simultaneously — and the consumer has no idea that for most of their policy’s active life, the dominant pricing variable is not whether they will die, but whether they will quit.

A regulatory shift

The Canadian term to 100 experience is the clearest historical proof of what happens when lapse assumptions fail. These products were priced assuming roughly 5–6% of policyholders would exit annually. Actual lapse rates proved to be closer to 1–2%, with rates below 0.3% at longer durations.

The pricing assumed a pool that would be largely empty by year 40. Instead, it was nearly full. The financial consequences were severe — a repricing crisis from 2010 onward that effectively made term to 100 a less competitive life insurance product option.

The regulatory response transformed the landscape. The Office of the Superintendent of Financial Institutions’ life insurance capital adequacy test (LICAT), which replaced the minimum continuing capital and surplus requirements framework on Jan. 1, 2018, introduced for the first time an explicit capital charge for lapse-supported products.

Under LICAT, every product must be designated as either lapse-supported or lapse-sensitive, and the insurer must hold capital against adverse lapse scenarios — including a permanent 30% decrease in best-estimate lapse rates and a catastrophe scenario of zero lapses in the first year.

For a product priced assuming 5% lapses, the insurer must now hold capital against a world where only 3.5% lapses, and against a scenario where nobody lapses at all. The capital cost of relying on lapse assumptions effectively consumed the pricing benefit those assumptions provided.

The 2017 exempt test reform under Bill C-43 simultaneously altered the product design landscape. The shift to participating whole life as the dominant permanent product — now representing approximately 90% of whole life new premium in Canada — was not accidental. Par whole life’s cash surrender values, adjustable dividend scale and participating account structure make its economics work whether policyholders stay or go, rather than depending on exits to generate profit.

The lapse advantage still exists across the product portfolio — most directly and certainly in term and non-participating products where the guaranteed cost was locked before regulatory changes, and indirectly through the participating account in par products where lapse experience flows through the annual dividend scale.

Tomorrow: How to capture an advantage from lapse-supported pricing and why the opportunity to do so is declining.