Life insurance product failures are often the result of small decisions compounding over time rather than a single event, according to a paper presented to the 2026 Actuaries Institute All Actuaries Summit.
In his paper titled The Failure Metric, Managing Director of Retender, Ilan Leas, examined 10 life insurance product case studies from Australia and overseas where he found that failures were generally driven by “pushing boundaries” in an attempt to address genuine consumer needs.
Leas wrote that failures emerge when incentives become misaligned, basic insurance principles are overlooked, distribution realities are ignored or future systems and behaviours are not anticipated.
The paper identified three broad conclusions:
- Product failure tends to be cumulative and predictable
- Failures are often behavioural, structural and human rather than purely technical
- There are no “silver bullets” that can prevent future failures but there are some questions that might change the outcomes
The paper examined failed products from different markets, in particular trying to draw lessons that are relevant to some of Australia’s current designs, namely:
- US post-level term insurance and the relevance to premium reviewability
- UK long-term care insurance and its relevance to post retirement products
- South Africa severity-based trauma products and their relevance to new severity based TPD designs
United States
In the US, Leas reported insurers there offered level premiums for an initial period before premiums increased substantially onto an age-based stepped pattern after the guaranteed term ended.
…affordability today cannot come at the expense of viability Tomorrow…
He said the products were designed to address affordability and lower entry barriers but relied heavily on assumptions about customer behaviour.
“Although sold as term products and guaranteed, insurers underestimated how consumers would react to an eight-times price increase at the end of their level premium period,” he wrote.
The result was that most policyholders left while a smaller group with higher-than-expected mortality remained.
“The experience with tweaking premium shapes highlights that affordability today cannot come at the expense of viability tomorrow,” Leas said.
United Kingdom
In another case study, UK long-term care products were developed to help consumers meet care costs in old age.
Leas said claims emerged more quickly than expected and care costs increased. Some insurers subsequently raised premiums (requiring another single premium) after initial guarantee periods expired.
“Below the surface though, the products weren’t really reviewable,” he said.
“An initial premium guarantee term was incorporated that technically allowed repricing thereafter, but asking an older perhaps more vulnerable cohort to top up a lump sum or lose their cover was not reasonable.”
The paper said complexity made the products harder for consumers to understand and more difficult for advisers to recommend.
…Innovation must be balanced with simplicity…
South Africa
Leas described how severity-based trauma products emerged in South Africa in the late 1990s to address what he described as a twin problem with traditional trauma cover: windfall payments and a lottery effect.
He described how the product launched successfully in South Africa, but argued the design itself was not the primary driver – the insurer was simultaneously a new entrant with a health data advantage, a changed distribution model, a disaggregated risk structure and a rich wellness program.
“The South African culture also encouraged consumers and advisers to try new things,” Leas wrote.
He said the product initially failed in the UK and Australia, where advisers struggled to justify a product that sometimes paid less than competitors, compounded by higher premiums.
“Innovation must be balanced with simplicity,” Leas wrote. “A rational product design, even if demonstrated elsewhere, can falter if it’s too complicated for consumers to understand or for advisers to confidently recommend.”
…insurers thought one thing, while consumers thought another…
Common Threads
Leas said a common theme across the case studies was the difference between what consumers believed they had purchased and what insurers believed they had promised.
“Beneath all the noise sits a very uncomfortable question. Did either of the parties to a failed product, that being the consumer and the insurer, allow themselves to be misled?” he said. “What risks did consumers believe they were transferring?”
He said many consumers believed they were buying long-term promises, while insurers often retained flexibility to respond if experience differed from expectations.
“The case studies often highlighted this perception gap – insurers thought one thing, while consumers thought another.”
…The goal, then, is not to eliminate failure…but to change its shape…
Leas also described how life insurance products often outlast the people responsible for designing them.
“Life insurance is particularly unique in that the CEO approving a product today is, in substance, making decisions for the next CEO – perhaps three or five CEOs ahead,” he said.
Institutions should make greater use of what he described as “failure capital”, referring to lessons accumulated through decades of pricing, claims, regulatory and product experience.
“Life insurers, however, possess something few industries do – failure capital,” Leas said. “The long liabilities should correspond with the asset of long memories.”
The paper discussed how the system’s stewards – boards, auditors, actuaries, management – each face personal incentives that make early action rational to delay. There is a conflict in that at the precise moments discipline matters most, the incentive to exercise could be at its weakest.
The paper concluded with a series of questions for product teams to consider when developing new products, including:
- How volatility has been incorporated into product design
- What risks customers believe they have transferred
- What lessons from previous failures have been embedded in the design
“The goal, then, is not to eliminate failure,” he said, “…but to change its shape. Smaller failures. Earlier failures. More transparent failures. The kind that teaches quickly, still protects customers, and ultimately strengthens the system.”
With our thanks to Ilan Leas, click here to access the full paper: The Failure Metric
