The impact of the COVID-19 has been felt across the insurance industry.
© iStock/ piranka

Examining the Impact of COVID on Group Life Underwriting

    alt txt

    properties.trackTitle

    properties.trackSubtitle

    September 2021

    The impact of the COVID-19 has been felt across the insurance industry.  Underwriters, in particular, are being challenged to understand the many ways COVID-19 is impacting the opportunities they underwrite. Historically, Group Life underwriters used previous claims experience to project future results. However, the effect COVID-19 has had on mortality over the last year and a half may not be indicative of future experience and therefore requires special examination.

    2020 was challenging year for group carriers with most reporting Group Life losses. As a result, many companies devised their own approach to incorporate COVID-19 claims experience when rating future business. Varied methods are expected while the disease evolves, as do the strategies we use to combat it. Underwriters should also consider incorporating data from a reliable source such as the Society of Actuaries (SOA) when considering the pandemic’s impact on the Group Life market.

    2020 was challenging year for group carriers with most reporting Group Life losses. As a result, many companies devised their own approach to incorporate COVID-19 claims experience when rating future business.

    Beginning in early 2020, the SOA Group Life Experience Committee conducted a continuous survey of group carriers to track and assess the impact of COVID-19 claims in the Group Life market.1 Participants included 20 carriers and approximately 90% of the group market, making the results representative of the U.S. Group Life industry. The SOA developed charts depicting results by month as well as exhibits for the 12-month period from April 2020 to March 2021.

    In this report, we’ve identified key findings in the SOA report that are most relevant for underwriters. The charts included are all derived from the same data set, yet each provides a different view, giving underwriters holistic insights into the pandemic’s impact. In general, carriers started receiving COVID-19 death claims in March 2020; but the industry only saw a material number of claims in April 2020, making the 12-month incurred analysis from April 2020 to March 2021 especially relevant for Group Life underwriters. For this reason, most of the charts included in our report feature a column of data specific to this 12-month “peak” period, in addition to quarterly data.

    Method and Definition of Terms

    It’s important to fully understand the charts and methodology the SOA used to create the analysis in the report. The tables and charts found in the SOA report and referenced in this paper are based on claims data compared to the “baseline”, which is defined as the average claims for the previous three-year period of 2017-2019. The baseline is essentially the expected claims for the period and is represented by 100%. The actual claims are presented as a ratio percentage to the expected claims. So when a table shows actual incurred claims as a percentage above 100%, this is considered excess mortality. 

    Claims are analyzed on a seasonally adjusted basis because mortality historically fluctuates by month. This method provides a more precise measure than considering mortality based solely on the full time period. The SOA analyzed claims using two categories - claims coded COVID and claims coded Non-COVID. Claims coded COVID are deaths where the cause of death was considered COVID-19. Roughly half of the carriers in the survey classify claims as COVID-19 only if it is listed as either primary or secondary cause of death on the death certificate. Many of the other carriers do additional research to identify COVID-19 claims, such as reviewing obituaries or communicating with medical examiners/ funeral homes. Non-COVID claims are the remaining excess claims, where COVID-19 was not coded as the cause of death.  Claims that exceed the expected claim levels, but were not coded as COVID-19, are classified Non-COVID. The SOA recognized the limitations in capturing the full impact of the pandemic based solely on claims coded COVID, which is why Non-COVID excess claims are also added.

    Non-COVID claims can include deaths where the individual may have had COVID-19, but it was not considered the cause of death. Another possible explanation for these excess Non-COVID deaths is due to deferred health care. There is considerable evidence that a significant number of heart surgeries, medical procedures, and cancer screenings were deferred during the pandemic. The possible result is that we experienced higher mortality due to these delays in treatment and preventive care.2

    Key Finding: Excess death rate for peak COVID months was 119.6%, nearly 20% above baseline; 77% of these excess claims were coded COVID.

    A key measure from the SOA report is incurred claims by count, shown in the table below. The total/baseline row is the excess mortality above 100%, including both COVID and Non-COVID claims. The primary finding from this table is that for the 12-months from April 2020 to March 2021, the excess death rate compared to the seasonally adjusted baseline of the previous three years was 119.6%. This excess mortality of 119.6% is based on 15.1% COVID claims combined with the Non-Covid claims of 4.5%.  Claims coded COVID were 15.1% of the total, making up approximately 77% of the total excess claims for the period. In comparison, the remaining 23% or 104.5% claims above baseline were considered Non-COVID or claims where COVID-19 was not coded as the cause of death. The table also shows seasonally adjusted claims by quarter and notes that the highest excess mortality was in the fourth quarter of 2020 at 126.9%. 

    Table 1 | Count-Based Incurred Incidence Results Relative to 2017-2019 Baseline Period (SOA Table 2.1)
    Count-Based 1Q20 2Q20 3Q20 4Q20 1Q21 4/20-3/21
    Total / Baseline 98.50% 115.20% 114.70% 126.90% 121.90% 119.60%
    COVID-19 Claims 1,152 12,862 9,690 22,195 20,248 64,995
    COVID / Baseline 1.00% 12.30% 9.50% 20.60% 18.10% 15.10%
    Non-COVID / Baseline 97.50% 102.90% 105.20% 106.30% 103.80% 104.50%
    Source: Society of Actuaries, Group Life COVID-19 Mortality Survey (2021, July)

    Key Finding: Significant differences in excess death rate by month.

    There has been considerable variability of Group Life death claims based on the incurred month, illustrated in the chart below. The dotted line shows incurred deaths per 1,000 for each month in 2020 compared to the same period in 2017-2019. Note that the dotted line showing 2020 results indicates that every month starting with April 2020 to the last available month of March 2021 indicates excess mortality. However, there is considerable variability in excess mortality among months. For instance, April, November, December, and January had very high incurred mortality, while the lowest months were June and September. COVID-19 deaths typically occur in the range of two months after exposure,3 so a potential explanation for lower mortality from May to June might be explained by the initial lockdown, which started approximately in mid-March 2020. Looking at the latest available months of February and March 2021, there is an improving trend from the previous months, though the SOA notes that expected claims reporting lag was added to complete these two months. 

    Figure 1 | Aggregate Incurred Claim Incidence Per 1000 Live By Calendar Year and Month (SOA Figure 6.1)
    AGGREGATE INCURRED CLAIM INCIDENCE PER 1000 LIVES BY CALENDAR YEAR AND MONTH
    Source: Society of Actuaries, Group Life COVID-19 Mortality Survey (2021, July)

    Key Finding: Excess death rate was 127.3% of face amount vs. 119.6% by claim count for the peak COVID months.

    Excess death claims based on the face amount were considerably higher than claims based on count. Overall, compared to the baseline, seasonally adjusted incurred Group Life claims by amount for the period April 2020 through March 2021 were 127.3% vs.119.6% by claims count. The SOA estimated that roughly half the difference is due to changes in age and gender mix. The remainder is likely due to salary and face amount inflation over the experience period. One potential explanation for what the SOA characterizes as “changes in age and gender mix” could refer to larger face amounts for higher wage-earning males. Typically, males in this age band are higher wage earners electing larger face amount coverage than females or younger ages of both genders. The excess death rate has been consistently 5-10% higher for males than females and the combined 45-64 age band has experienced higher mortality compared to the baseline than younger ages. This difference could partially explain why the amount-based claims saw a more significant impact than count-based claims. 

    Table 2 | Amount-Based Incurred Incidence Results Relative to 2017-2019 Baseline Period (SOA Table 2.2)
    Amount-Based 1Q20 2Q20 3Q20 4Q20 1Q21 4/20-3/21
    Total / Baseline 102.70% 119.90% 124.70% 132.90% 131.60% 127.30%
    COVID-19 Claims 57,699 K 463,995K 397,884K 813,876 K 842,909 K 2,518,664 K
    COVID / Baseline 1.40% 12.10% 10.70% 20.70% 20.70% 16.00%
    Non-COVID / Baseline 101.30% 107.80% 114.00% 112.20% 110.90% 111.30%
    Source: Society of Actuaries, Group Life COVID-19 Mortality Survey (2021, July)

    Key Finding: Blue-collar workers had the lowest excess death rate (118%) during the peak COVID months.

    When measuring excess mortality by industry collar, the blue-collar group saw the smallest increase in excess mortality for the 12-month period from April 2020 to March 2021 at 118%. The white-collar saw the most increase at 126% or 26% excess mortality.  We saw a pronounced difference between blue and white-collar impacts early in the pandemic, where Q2 and Q3 of 2020 showed a much higher rate of COVID deaths for white-collar workers. This difference became less pronounced late in 2020, but the gap again widened in Q1 2021. These early results were surprising since we know that most blue-collar workers could not work from home and limit exposure like many white-collar jobs. It’s possible that early in the pandemic, urban centers with high white-collar content caused this early spike, but as the disease spread to different areas of the country, mortality among the three collars became more consistent. Though there was seasonal variability, if we look at the 12-months from April 2020 to March 2021, the rate of COVID deaths compared to the baseline was consistent among the three industry collars. The table shows that the rate of COVID deaths ranged from 115.4% to 116% compared to the baseline as measured at the end of the period.

    Table 3 | Excess Mortality by Industry collar (SOA Table 7.1)
    EXCESS MORTALITY BY INDUSTRY COLLAR
    Source: Society of Actuaries, Group Life COVID-19 Mortality Survey (2021, July); All Collars2

    Key Finding: The 45-64 age band had the highest excess death rate of 129% for the peak COVID months.

    The SOA combined claims coded COVID and Non-COVID to evaluate deaths by age to equal excess mortality. For instance, at the 65-99 age band for April 2020 to March 2021, the excess mortality is 116%, or COVID 14.8% combined with the Non-COVID 1.2%. Note that the highest excess mortality for any quarter was 143% for the 45-64 age band for incurred claims during the first quarter of 2021. We know that the majority of COVID deaths come from the older population; however, the rate of increased mortality compared to the baseline was actually higher on a relative basis at the younger age bands for the Group Life market. In fact, if we look at the excess mortality for the first quarter of 2021, the 45-64 age band saw a much higher excess mortality of 143% than the 116% observed in the 65-99 age band.  One possible explanation for this difference might be the higher vaccination rate for ages 65 and older during the first quarter of 2021. Finally, note that during the peak COVID months of March 2020 to April 2021, the Non-COVID rate was much lower for the 65-99 age band than the other age bands at 1.2% for this period.

    Table 4 | Excess Mortality by Age Band (SOA Table 7.5)
    EXCESS MORTALITY BY AGE BAND
    Source: Society of Actuaries, Group Life COVID-19 Mortality Survey (2021, July); All3

    Projecting Future Mortality

    As we look forward, several scenarios could have a positive effect or a tailwind on future mortality. The most apparent positive impact on future mortality would occur following herd immunity from the combined effect of the vaccines, along with people who have built antibodies from contracting COVID-19. The virus would not have sufficient hosts to spread the disease and would theoretically cease to be a pandemic by reaching herd immunity. One less noticeable tailwind is the effect of advanced deaths incurred during the pandemic. In other words, advanced deaths are defined as unhealthy people who possibly died due to the effects of COVID-19, but otherwise might have lived longer. The possible result is mortality could improve if we consider that unhealthy people died during the pandemic making the remaining population healthier.

    On the other hand, there are conditions that we might consider possible headwinds or factors that offset positive mortality. One potential headwind to improved mortality is deferred health care during the pandemic. Previously, we mentioned that deferred health care could explain some of the excess deaths that were not coded COVID. Deferred health care could also have a negative impact on future mortality if visits to healthcare providers continue to be limited. Vaccine hesitation is another headwind that hinders our efforts to achieve herd immunity. Vaccination rates have plateaued in recent weeks and are lower than the vaccination activity earlier in the year. The Centers for Disease Control and Prevention (CDC) estimated as of September 1st that 74.4% of people in the U.S. over age 18 have received at least one dose and 63.6% are fully vaccinated.4 However, recent efforts to encourage vaccinations as well as the FDA approval of the Pfizer vaccine could reverse the downward vaccination trends. Overall, several factors could positively and negatively affect future mortality, and there is considerable uncertainty regarding the future. 

    Group Life Underwriting Going Forward

    Underwriters rely on historical experience to predict future results, and COVID-19 has added complexity to that task. In recent years, carriers have extended standard rate guarantee periods for Group Life clients from two, three, four, or even up to five years. Part of the reason for this trend of extended Group Life rate guarantees is that while U.S. mortality has fluctuated on an annual basis, there has been a positive mortality trend for several years, resulting in relatively stable Group Life experience.5 This stability limits the risk of offering longer rate guarantees. If we compare Group Life to other ancillary products such as Long-Term Disability, dental, and Short-Term Disability, life experience has tended to be much flatter and more predictable. However, this all changed in 2020 with the unexpected results from the pandemic.

    The key question for underwriters is how do we consider COVID-19 deaths and 2020 claims experience when rating cases? Thanks to the SOA Group Life survey, we have reliable data regarding the historical effect of COVID-19 and excess mortality on the Group Life market. Also, many carriers are tracking COVID-19 deaths within their own block of business, potentially giving carriers a leg up with renewal actions on in-force business. We’ve seen varied approaches to considering COVID claims. Some carriers are removing 2020 experience, some are not making any adjustments, still others are considering modified adjustments to the claims experience with the expectation that excess mortality will continue for at least the next few months. While each approach may be different, it’s more important than ever for underwriters to make sound underwriting decisions based on the most current and best available information.

    As Group carriers are working to devise strategies to consider the effect of COVID-19, the fundamentals of sound underwriting remain unchanged. For example, diversifying risk by packaging multiple lines of business, remains a solid underwriting practice. Insuring multiple lines of coverage gives carriers the ability to diversify earnings and offset losses from one product with the potential gain of another.  COVID-19 has added more unknowns to the list of underwriting considerations, but perhaps it has added variables that make underwriting more challenging and even more interesting. On a more human note, we hope this pandemic resolves ending the terrible suffering and allowing the full quality of life we all once enjoyed. 

    Contact the Author:
    Richard Beckel
    Group Underwriting Consultant, Group & Living Benefits
    References 1Society of Actuaries. (2021, July).  Mortality and Longevity: Group Life COVID-19 Mortality Survey. https://www.soa.org/globalassets/assets/files/resources/experience-studies/2021/group-life-covid-19-mortality.pdf 2 Vose, J.M. (2020, September 15).  National Library of Medicine: Delay in Cancer Screening and Diagnosis During the COVID-19 Pandemic: What Is the Cost? https://pubmed.ncbi.nlm.nih.gov/32965661/ 3 PLOS ONE (2020, May 12). Time-to-Death approach in revealing Chronicity and Severity of COVID-19 across the World. https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0233074   4 Centers for Disease Control and Prevention (2021, August). CDC Data Tracker.  https://covid.cdc.gov/covid-data-tracker/#vaccinations_vacc-total-admin-rate-total 5 Society of Actuaries (2021, January). U.S. Population Mortality Observations – Updated with 2019 Experience. https://www.soa.org/globalassets/assets/files/resources/research-report/2021/population-mortality-observations.pdf