Auditors & Actuaries: Life Insurance
Rethinking Mortality

• Terrence Dopp
• December 2020
• 

Key Points

• Taking Stock: Companies including RGA and Brighthouse revised their initial projections at the close of the second quarter as a more nuanced picture of the pandemic came into view.
• Issue: The industry saw something of a payout reprieve to date, with four-fifths of pandemic deaths among seniors and many others in populations with historically low insurance penetration.
• Hanging Sword: With a resurgence in the fall and lowered interest rates likely to persist for an extended period, companies could still face economic difficulties that outweigh mortality costs.

At the beginning of the year, as the novel coronavirus moved from a theoretical malady to a pandemic, life insurers did what they always do. They planned and drew up worst-case estimates of their exposures.

For the Reinsurance Group of America, that meant issuing an initial estimate that 100,000 U.S. deaths from the disease would mean $400 million to $500 million in extra pretax mortality costs. At the close of the second quarter, the company revised that figure downward by about $200 million. The revision came with the caveat that any gains could be partially offset by a higher possible range of general population deaths. “Early on in the pandemic when we were putting together models for the specific event, there wasn't a whole lot of data to go on,” Timothy Rozar, an executive vice president at RGA, said.

Missouri-based RGA has $3.5 trillion of life reinsurance in force and $76.7 billion in assets, according to its latest annual report. It touts itself as the only global reinsurer focused on life and health. In September, AM Best reaffirmed its A+ (Superior) financial strength rating of the company and its subsidiaries, reflecting a strong balance sheet, operating performance, favorable business profile and enterprise risk management.

“The data was just early and emerging, and to be honest is still emerging in many ways,” Rozar said. “So, I would say we made a best-estimate assumption based on what we knew at the time. As data comes in, both research data from the medical community, as well as population data from health organizations in the U.S. and around the world and of course our own data, that will inform revisions.”

In some respects, the event forced the life insurance industry—which traditionally has an event horizon of decades—to follow a path typically taken by other lines, such as property/casualty: Release the first estimate of exposure to a catastrophe event, followed by revisions as more accurate data becomes available. Life policies routinely have a life span of 20 or more years compared to P/C's annual renewals, and the impact of COVID-19 is uncharted territory compared to the annual hurricane season.

The pandemic has many unknowns, including how many cases will develop in winter, and the timing and efficacy of a potential vaccine. Add to that the economic issues: the size and impact of coronavirus relief packages, along with continually lowered interest rates that harm balance sheets.

Still, mortality numbers that were derived in the heat of the moment have changed as time allowed insurers to gather new evidence about the economic status of people who succumbed to the disease, whether they had life insurance, and if so, how much.

Michael Porcelli, an AM Best director, summed up the life insurance industry this way: “We don't really live in a catastrophe world.” He said life insurers tend to be conservative by nature, hence the goal to calculate a worst-case scenario around the problem and make downward revisions to estimates that were made with caution. But, he said, life insurance isn't like other lines.

The downward mortality risk revisions are “almost universal” across the industry, he said.

“We have to take those with a grain of salt. I think companies realized that, and I don't think there was an attempt for excess precision,” said Bruno Caron, a senior financial analyst with AM Best. “It's very normal to see changes with those types of estimates. That's just part of the conversation.”

Life insurance has seen its share of changes in 2020.

An increase in direct-to-consumer sales drove an increase in the total number of U.S. individual life insurance policies being sold in the second quarter, though new annualized life insurance premium dropped 3% as customers increased their focus on smaller, simplified- and guaranteed-issue policies that cover final expenses, according to LIMRA's Second Quarter 2020 U.S. Individual Life Insurance Sales Survey.

Lockdowns forced insurers to up their digital game or sacrifice sales. Turbulence early in the year took a toll on their investment portfolios. Driving it all was the specter of COVID-19.

Rozar, of RGA, cited the Sept. 11, 2001, terror attacks in New York as a precedent. He said the company based its initial models on pandemic studies, historical events including the 1918 Spanish flu outbreak and its best outlook. “Life is such a longer-term event that having discrete disruptions is pretty rare,” he said. “Both of those situations were what we would classify as a financial event, and manageable relative to the broader environment.”

Same Boat

RGA wasn't alone in finding mortality costs lower than initial projections.

Brighthouse Financial Inc. also said in its second-quarter earnings presentation that it experienced about $25 million in life insurance costs associated with the pandemic in the United States after 120,000 deaths; earlier estimates called for $70 million for each 100,000 lives lost. A Brighthouse spokesman declined to expound on the results beyond Chief Financial Officer Ed Spehar's comments in an earnings call.

Steven Weisbart, senior vice president and chief economist at the Insurance Information Institute, said life insurance companies frequently update their understanding of the mortality risk that they are assuming, and it isn't shocking that a data-driven industry updates its models based on severity and disease outcomes.

He said the economic impacts of the pandemic may outweigh shorter-term patterns. Insurers also will face questions about whether and how to alter rates to react to this interplay between the investment climate and mortality, according to Weisbart.

“Most of the investments that life insurance companies make are in long-term bonds, and long-term bond yields have been plunging. So that puts even more pressure on the rates charged for transferring the risk because you don't have the investment income to sort of bail you out if you make a mistake,” he said.

For whole life policies owned by older people who die from COVID-19, the cash value of the policy and the length of time premiums were paid will lower the costs compared to paying the death benefits, he said. And the disease has been much harder on the uninsured, Weisbart said.

Disparate Impacts

Statistics bear out COVID-19's impact on groups that historically face a protection gap.

Between Feb. 12 and Oct. 15, the U.S. saw 7.9 million infections resulting in about 216,000 deaths, according to the Morbidity and Mortality Weekly Report compiled by the Centers for Disease Control and Prevention. Of those pandemic fatalities, 78.2% were people over age 65. Broken down even further, 21.7% were in the 65-74 age group, 26% in the 75-84 range, and the largest share, 30.4%, were over 85.

The CDC also found that Hispanic and Black Americans accounted for 24.2% and 18.7% of deaths, respectively, though they comprise 18.5% and 12.5% of the U.S. population.

“The impact from life insurance claims from COVID-19 deaths has been lower than expected, and there are a few reasons for that—some of which, frankly, should be troubling to the industry,” Kweilin Ellingrud, a senior partner at McKinsey & Co., said. “COVID-19 deaths, as we've all seen, have been concentrated among the elderly and disproportionately in diverse groups.”

She said Black Americans and Hispanic people are more likely to be front-line workers and have a higher death rate if they contract COVID-19. They are also less likely to have health insurance, and the elderly are less likely than working-age adults to have a life policy that covers more than funeral expenses.

“This is more of an earnings event; it impacts earnings, not balance sheets,” Ellingrud said. “You've seen some of the carriers start to repurchase shares or take actions that suggest they think it's either not a huge impact to their balance sheet or not as significant as we thought it could be at the very beginning.”

On the flip side is long-term care coverage, according to Ellingrud and Weisbart.

Because some states transferred early COVID-19 patients to long-term care facilities, which have been the site of many pandemic-related deaths, Ellingrud said, people are doing all they can to avoid those spaces. In other words, they've been carrying long-term care policies for years, but when it comes time to use them, they decide against it, she said.

For insurers the pandemic has two salient figures: general-population deaths and insured losses. The tragedy of the disease's toll notwithstanding, companies must look at how many of the broader total of deaths were among those they insured.

Rather than try to formulate an estimate of how many lives the pandemic would claim and base losses on that, life insurers developed estimates of excess mortality costs per 100,000 to maintain flexibility. Doing so allowed them to create a scalable template that was based on medical developments and that could be altered as the disease ebbed and flowed.

The CDC's mid-October figure of more than 216,000 fatalities fell below the 300,000 in excess mortality for the same time frame in typical years, raising questions as to whether the actual number of deaths attributable to the virus was in fact larger.

Weisbart called the industry data-driven and said companies are continually reevaluating every shred that comes in to determine economic conditions and whether rate changes are necessary. “The real danger, if you will, is the low interest rate environment and the continued need for investment income to cover the interest rate guarantees in those long-term policies,” Weisbart said. “That's, I think, the more serious consideration for the industry. Not the pure mortality risk.”