Update from America – Hank George

Greetings from Milwaukee, where the last vestiges of our protracted winter (i.e., snow on the ground) finally disappeared in April.

If I digressed to characterize the wretchedness of January and February here on Wisconsin’s famed frozen tundra, you would likely judge me so disposed to hyperbole that you would not stop reading this report!

This update is based largely upon findings from our 2018 New Business Critical Issues Survey. It was conducted over a span of six months stating in October 2017. Chief underwriters from 93 insurers offering individually underwritten life insurance participated in this survey.

This project was funded by cosponsorship contributions from reinsurers and underwriting-related service providers. Nearly all reinsurers doing business in our individual life market stepped up to assist. Part of their motivation is that distribution of the copyrighted survey report is limited solely to cosponsors and survey takers.

This survey addressed a broad range of subjects. The one garnering the greatest attention is what we call “accelerated underwriting”.

Accelerated Underwriting (AU)

AU is defined as offering life insurance cover up to some maximum age and sum insured without mandatory paramedicals and laboratory tests despite charging same premium rates as those for fully underwritten business.

Based on this definition, 65% of companies offer accelerated underwriting for at least one individual life product. In a substantial portion of companies, AU is permitted on most or all life plans.

Only one insurer reported rejecting AU outright. Most of the rest were sorting out the matter at the time they took the survey. The “good money” (in race track parlance!) says that nearly all will follow suit and roll out AU in the near future.

Why?

Because those not offering AU will see virtually all of the best new business go to competitors. This will lead to substantial attrition in the ranks of their (increasingly disgruntled) advisors. In today’s tight market, these two outcomes carry a dismal corporate prognosis.

The three most significant drivers of AU development were, in this order, actuaries, chief underwriters and reinsurers. Sales/marketing executives and CMOs had notably lesser roles and barely 10% cited “outside consultants” as significant assets.

Such is the nature of the barren American marketplace for underwriting-related consultancy services nowadays!

Many larger insurers with robust retention limits and/or favorable reinsurance arrangements offer $1 million or more of cover on an accelerated basis. Some go as high as $2.5 million.

When asked what they consider the maximum age at which AU is practical based on the number of applicants likely to qualify, 42% said age 60 and 32% favored age 55.

Nevertheless, many of these same companies offer to underwrite on an accelerated basis through age 65.

What does it take to qualify for accelerated underwriting?

Most insurers require a handful of rapid-access evidence, typically consisting of:

An application, with the risk history taken by teleinterview or via online questioning
An MIB report
A motor vehicle record (MVR) noting any violations within a period of years
A pharmacy record listing all prescription medication taken over some interval, currently Rx records can be accessed on 7775%-80% of insurance seekers.
An electronic data report addressing a potpourri of financial details, court records, etc. et al

If the proposed insured qualifies for preferred risk cover based on the foregoing evidence, approval is “accelerated” by virtue of no need for a paramedical or lab tests.

The Achilles heel of AU is that insurers must rely on what the proposed insured tells them regarding such essentials as height/weight, tobacco use, HIV status and so on.

A recent paper in a major actuarial publication has fueled concern for rather generous understatement of weight and this is reflected in the survey as the domain conferring the greatest worry over potential material misrepresentation.

A majority of survey takers (61%) believe that nondisclosure of tobacco use – likely the most impactful of these issues – will exert a “moderate” adverse impact AU mortality.

Some of us old gits reckon the magnitude of “smoker’s amnesia” may be a tad more impactful! Time will tell.

Direct-to-Consumer (DTC) Testing

Americans have unfettered access to relatively-inexpensive DTC tests ranging from coronary artery calcium (CAC) scans and abdominal aortic aneurysm ultrasound to laboratory profiles so comprehensive they contain every routine blood test we use in underwriting except NT-proBNP.

The US Food and Drug Administration (FDA) has approved the 10-component 23andMe genetic test profile for DTC testing and also created a fast track to accommodate sanctioning of other genetic profiles for this indication. DTC BRCA1-2 breast cancer gene mutation screening is also available at relatively low cost.

To make matters worse, a consensus agreement affecting the substantial majority of US states prohibits asking applicants whether they have had any medical tests not ordered by healthcare professionals (which, of course, defines DTC tests!).

When asked about the extent of their concern for nondisclosure regarding DTC testing, 44% said it was “moderate” and 36% owned up to just “minimal” concern.

Several survey takers commented that their views could change as more DTC tests (especially of a genetic persuasion) enter the marketplace.

Just as with the potential for tobacco nondisclosure hijinks, one wonders if the dominant viewpoint here is at least a bit naïve.

Marijuana Use

In the last few years there has been a gradual shift in how American insurers regard occasional recreational cannabis use by adults.

Instead of charging them the same base premium designated for cigarette smokers (indeed, most if not all tobacco devotees), a courageous few were countenancing non-tobacco rates for selected pot smokers, largely on a case-by-case basis.

In the new survey, 59% said “yes” when asked if they “permit adults who occasional smoke marijuana…to be eligible for ‘non-tobacco user’ rates on some basis.”

Only 12% had considered and then rejected this practice whereas 9% were weighing the merits of joining the majority. The rest had not as yet given thought to the question.

Our July CE course on Marijuana and Underwriting is based on 243 studies and review papers addressing every aspect of cannabis use and what is known of its links to various ailments, adverse motor vehicle consequences and, most importantly, mortality.

Our research shows that the weight of evidence strongly favors liberalization of underwriting of recreational marijuana at least to the extent that the majority of carriers are presently accommodating.

The matter of medicinal pot use is more problematic if only because many of the ostensible indications for cannabis therapy substantially impact mortality risk. In addition, there is markedly greater exposure when marijuana is smoked in a therapeutic context as compared to the doses absorbed by most recreational users.

Opioid Use

Roughly one-third (1/3^rd) of US companies now screen for opioid use on an age/amount and/or elective (underwriter discretion) basis. An equal share of carriers are considering this practice and no doubt most will chose to proceed.

Needless to say, this is being propelled by concern for the so-called “opioid epidemic”. Over half of respondents who felt they could speak to their life claims experience reported at least one recent opioid-related death claim.

Heroin-related deaths were rarely an issue for insurers until the last few years when they became as prevalent in the suburbs as inter city urban areas. Most of the larger policies are sold to persons living in the ‘burbs, as they say.

Insurance opioid testing protocols cover all major forms except tramadol and tapentadol. These two are distinct from other opioids in their mode of action but both can be abused. If the highly sensitive screening tests are positive, confirmatory tests are done before the insurer is advised of a positive result.

Underwriting Calculators

When I made what was likely my last-ever Association of Home Office Underwriters (AHOU) presentation at our 2018 conference, I went off on a wee tirade denouncing the use of so-called calculators to “aid underwriters” in assessing insurability of cancer risks.

My views are less than warmly received by some reinsurers given how widely calculators have metastasized within underwriting guidelines.

In the survey, we asked several questions about respondents’ perceptions regarding these confounded devices.

The substantial majority concurred that calculators increase the likelihood of avoidable errors by underwriters and also promote the deeply flawed perception that cancer cases can be routinely engine/machine underwritten sans human involvement.

Touché!

“Selfies”

An underwriting services-wannabe came upon the scene recently, pitching the proposition that one could distinguish biological from chronological age based on facial features on a “selfie”.

At this writing, we are aware of just one insurer using this resource.

When we asked if survey completers felt that selfies were “a viable source of accurate information for underwriting purposes,” two of 93 respondents replied in the affirmative.

Based on our research (which we wager is the most extensive yet undertaken), we congratulate the 91 chief underwriters who look askance of this proposition.

Meanwhile, there is at least one new service provider on the cusp of introducing epigenetic testing. This is a far more intriguing proposition than conventional genomic testing because it does not involve germline (inherited) genetic changes. Epigenetic changes arise over the course of one’s lifetime.

Our research to date suggests that epigenetic testing may indeed confer substantial value as a mortality marker. Certainly biological vs. chronological age distinctions would be far more credible on this basis as compared to “selfies”.

This said, there are many issues to be resolved (several of which could rule out epigenetic testing based on cost and other considerations). The jury, if you will, is still out.

Accelerometers

While just two companies currently offer lower premium rates and/or other perks to applicants willing to wear accelerometers (FitBit, et al), 18% are studying the proposition and just 4% have nixed the notion.

We appreciate the initial appeal of this concept.

What its promoters may not appreciate is the weight of evidence in published studies showing substantial attrition in device deployment that evolves over some interval of time…and regardless of incentives to continue wearing such devices.

As a sales gimmick, this gambit may make sense.

As a legitimate approach to improving mortality outcomes, a healthy skepticism must be maintained until such time as hard evidence – not merely wishful projections – confirms a sufficient and sustained payoff.

Fair vs. Unfair Discrimination

Respondents were shown seven potential risk assessment criteria and asked if they considered them to constitute fair vs. unfair discrimination as potential arbiters of insurability.

The strongest consensus was with religious affiliation, where 94% considered it to be an unfair form of discrimination. This was followed by race (90%), ethnicity (89%) and sexual orientation (86%).

Only 74% held that marital status was flawed risk assessment criterion and just 66% took this position regarding the use of personal purchase records in underwriting.

The last criterion scrutinized was zip code (our term for postal zone).

It is disconcerting that 1 in 4 respondents did not regard this as a grossly inappropriate underwriting consideration.

In most large America cities there are one or more postal zones wherein a high proportion of residents come from a single racial or ethnic group, making this a virtual poster child for unfair discrimination.

Underwriting Engines

Over half (55%) of 93 US life insurers currently have engines that facilitate “straight-through processing” of new applications, making at least some less-complex underwriting decisions without human intervention.

Half of these carriers created their engine internally, 30% acquired it from a reinsurer and the remainder from an independent software firm.

Only 5 insurers either stopped using or decided against deploying an engine. Most of the others were actively exploring this proposition.

Working Remotely

Approximately two-thirds (2/3^rd) of US life insurers allow at least some of their underwriters to work remotely (that is, from home) on some basis.

In our currently prevailing “sellers’ market” for underwriting talent, those who do not accommodate working remotely are at a formidable disadvantage when seeking to hire veteran underwriters.

A slight majority (55%) of chief underwriters tell us they do not get significantly greater productivity from remote (vs. in-office). If you had asked this question 10 years ago, 80% would have said that underwriters working remotely were anywhere from 10% to 25% more productive.

Why did this changed?

In the early years when carriers were still experimenting with this practice, those who worked from home feared that they might be forced to forfeit this option. Therefore they worked harder and were thus conspicuously more productive than their peers laboring in the confines at head office.

Now that roughly half of all U.S. underwriters work from home there is no longer a perceived risk of losing this beloved option and thus little incentive to sustain higher levels of productivity. Indeed, there are substantial advantages for insurers even in the absence of heightened remote underwriter case output.

Some 34% of survey takers that embrace working remotely are presently amenable to underwriters with less than three years’ experience working remotely. A decade ago most required at least five years’ seasoning and above average performance ratings as eligibility criteria for working from home.

The success of work-from-home underwriting is underscored by the fact that 99% acknowledged being “pleased” with the impact of this practice.

Artificial Intelligence (AI)

Asked to opine on the future impact of AI, the majority concurs that human underwriters will continue to have the dominant role, with AI assisting rather than replacing most of their staff.

They also largely agree that underwriters will need to become more fluent with predictive analytics and gear up for greater contact with both advisors and their clients.