New analysis of 2019-2020 data confirms long-standing patterns on mortality rates based on underwriting and policy size
Recently, the Canadian Institute of Actuaries (CIA) published its latest mortality report on the Canadian individual life experience for the policy year 2019–2020.
It covers the one-year period beginning with the policy anniversary in 2019 on an age-nearest-birthday basis for data submitted by seven companies. While it focuses on individual life insurance policies and riders issued in Canada that require full underwriting, it also analyses other individual life insurance segments.
“This is a mortality study that's been done for 70 years, so it’s quite a long-term study,” says Bob Howard, FCIA, a member of the CIA and author of the paper. “From one year to the next, we can see some trends emerging. It still doesn't tell you everything you want to know, and there's statistical fluctuation involved. But over the long haul, it does give us good information.”
In the latest wave of the study, the Institute found mortality rates decreased in 2020, the year the COVID-19 pandemic hit. While that might raise some eyebrows, Howard says it’s too soon to make any conclusions.
“For the most recently published year, we saw an overall mortality ratio of actual to expected of 90% for males, and the previous year was 91%. But the standard deviation and those mortality rates is about 2%, so that difference is well within the range where you really cannot draw any inferences,” he says. “For females, the two were essentially the same in the last two years, so the difference is not significant.”
There was limited overlap between the study period and the first year of the pandemic; Howard estimates only a quarter of the deaths analysed fell within the pandemic window. To fully capture and assess the impact of the pandemic, therefore, will require another snapshot, which he assumes will be available next year.
The data also showed an inverse relationship between policy size and the rate of mortality – the larger the policy size, the smaller the rate of mortality. That’s neither disturbing nor surprising, Howard says, as policy size has been long established as arguably the most significant factor not taken into account in mortality tables prepared by actuaries.
“People have spoken about this qualitatively for many years, and I’m encouraging actuaries to take that into account,” he says. “Look at it quantitatively. How much does it go down for a particular variation in size? The pattern has become quite regular … it’s getting to the point where people can now make inferences.”
Finally, the mortality data shows lower mortality rates for standard-issue policies than simplified-issue policies, where individuals are asked fewer questions during the application and aren’t subject to any further investigation like paramedical exams.
“The company knows that they're not going to get as good information. And people who can qualify for standard insurance probably are going to buy it going the normal route. So simplified issue is mainly for people who have an issue that they know about, or just want to get it done really quickly,” Howard says. “Because of that, the companies, and I would think probably the applicants as well, would expect that the mortality rates are going to be higher.”
While most people might have thought that mortality rates between policies granted through normal and simplified-issue underwriting would start to converge after several years, the CIA data suggests mortality rates for simplified policies will be materially higher for an indefinite period of time.
“There's nothing in the report that would suggest upcoming major changes in insurance rates,” Howard says. “But a lot of that depends on the progress of the pandemic.”
