American Academy of Actuaries: Life Underwriting and Risk Classification Work Group's Comments on Accelerated Underwriting

WASHINGTON, Dec. 4 -- The American Academy of Actuaries issued the following letter:

* * *

To: Honorable Mark Afable, Chair, Honorable Grace Arnold, Vice Chair, Accelerated Underwriting (A) Working Group, National Association of Insurance Commissioners (NAIC)

Dear Commissioner Afable and Commissioner Arnold:

One of the goals of the Life Underwriting and Risk Classification Work Group of the American Academy of Actuaries1 is to monitor public policy developments regarding underwriting and risk classification for life insurance and annuities. We have reviewed the Ad Hoc Drafting Subgroup exposure draft of Nov. 8, 2021, and are pleased to have the opportunity to provide the following general comments. Due to the brief comment period, we were unable to provide more detailed comments.

* In general, we believe the recommendations contained in the paper are sound concepts. However, we note that some of the recommendations may be challenging to implement from a practical standpoint and others may require more detail in order to assure that they support the working group's charge related to the use of external data and data analytics in accelerated life underwriting.

* Although many of the comments provided to the working group included consumer benefits of accelerated underwriting, the paper tends to focus on the potential issues. We believe the paper might be more balanced if it included more discussion of the favorable impacts to the consumer and believe the table of contents may be a useful guide in such an effort.

* Throughout the development process of this draft paper, the working group has heard from a variety of stakeholders regarding different practices and with different perspectives. Not all of the information provided should be generalized across the life insurance industry, and we suggest revisions throughout the paper to avoid overgeneralizations. For example:

- The paper generally assumes that accelerated underwriting uses predictive models or machine learning algorithms. However, the use of these tools is not a defining characteristic of accelerated underwriting, because some accelerated underwriting programs use only human underwriters. In addition, the use of these tools is not exclusive to accelerated underwriting, because the tools are used in all types of life insurance underwriting (e.g., simplified issue, full underwriting, etc.).

- The draft paper implies that there is a distinct definition for "traditional" underwriting and that there is a differentiation between traditional underwriting and accelerated underwriting. However, underwriting is an ever-changing continuum, and many of the practices used in accelerated underwriting are also used in full underwriting.

- "FCRA data is already used in property/casualty lines of business" may imply that Fair Credit Reporting Act (FCRA) data has not been used in life insurance, but FCRA data is already used in life insurance.

- Common types of underwriting data are listed with uncommon data types, which may imply that there is a higher prevalence of less common data in the underwriting process.

* The definition of accelerated underwriting used in the draft paper is important, as it potentially paves the way for use of the term in other policy documents. The proposed definition conflates the general concept of accelerated underwriting and the use of data and predictive models in underwriting, but data and predictive models are used in all forms of underwriting. We recommend revisiting this definition and the use of these terms.

* We do not understand why the bullet point "FCRA data may be used to predict mortality, but there may not be a reasonable explanation for that correlation" has a footnoted reference to Actuarial Standard of Practice (ASOP) No. 12. We do not find any portion of the bullet point that is specifically related to ASOP No. 12. The specific section of ASOP No. 12 should either be referenced and quoted if applicable or the reference to the ASOP should be removed.

We are available to support your efforts to refine this paper and would be able to provide a more thorough and nuanced review with an extended comment period.

If you have any questions or would like further dialogue on these topics, please contact Khloe Greenwood, the Academy's life policy analyst, at [email protected].

Sincerely,

Sue Bartholf, MAAA, FSA

Chairperson, Life Underwriting and Risk Classification Work Group

American Academy of Actuaries

CC: Jennifer Cook