Coronavirus Considerations – Underwriting in a Changing Self-Funded Risk Landscape
I have a confession to make. As I watched the Coronavirus pandemic spread throughout the world and into the United States in late January and February, I was openly derisive about the potential impacts. I was convinced this would just be another flu (don’t we have one of those already), and this would not likely be a significant concern, both from a risk/loss perspective and from a societal impact standpoint.
Facing the Unknown
It’s a good thing I learned an important lesson a long time ago: be humble in defeat. As late February rolled into March, there was clearly a shift to this becoming a more serious concern.
I brushed up on my reading and reacquainted myself with terms like R0 (pronounced R naught – which indicates how contagious an infectious disease is) and looked at the numbers related to incidence rate and severity. I also started reviewing numbers generated by various models – both within HM Insurance Group and our parent company Highmark Inc. and with those produced by outside experts. I watched a very good YouTube video with Joe Rogan and Dr. Michael Osterholm from the University of Minnesota. The more I learned, the more uninformed (or, using plain words, wrong) my initial stance was on how the Coronavirus would impact our lives.
One of the most difficult parts of being an underwriting leader is turning volumes of information into the specific tactics used by our underwriters. The deeper I looked, the more difficult it became to develop specific actions. In general, I am distrustful of models – I think they give guidance using a set of assumptions but not “the answer.” There are three reasons I feel this way:
- The inputs may be wrong. For example, it is clear to me that the lack of available testing has meant that many people have had the virus and recovered without ever being tested and having this confirmed and also that many people have sadly passed away due to the Coronavirus without their deaths being attributed to it.
- The assumptions used to generate the model results are not always clear and are subject to change. The initial R0 (contagion level) is influenced by how effective we are at social distancing. If you don’t give a fire fuel, it dies out, and that is what social distancing is attempting to do. So the R0 may be different at the beginning, middle and end of the pandemic, depending on how aggressive we are in our efforts to prevent the spread of the Coronavirus.
- The sensitivity of the model results is not always clear when it comes to changes in inputs or assumptions.
Whatever my thoughts are on models, I’m left with what we should do. Part of what informs my thinking is what we are actually observing. We have received numerous Specific claims notices for insureds with the Coronavirus. Some of these claims have been for amounts larger than we originally anticipated. Those claims have been offset largely by a reduction of medical claims that are not Coronavirus-related.
We have not seen any Aggregate claims related to the Coronavirus yet, but I believe that we will. We have received many requests for expansion of our policy terms, some of which we have allowed and others we have not. Of course, all of this is happening on cases that were underwritten within the last year. Whatever happens with those cases is not risk we can manage today.
I’m really interested in what we should do and how we should manage things going forward. I think the biggest impacts are still to come. There will be ups and downs in the fight against the Coronavirus. Both the numerator and denominator in our gross loss ratio calculation will be affected. The numerator (claims) will change as I described above and will continue to evolve in the coming weeks and months. The denominator (premium) will be reduced, at least temporarily, as many people have been laid off in the last six weeks; how many of them will be rehired remains to be determined. And some of our clients’ businesses may not survive this current economic slowdown.
How this will evolve as we go forward is interesting to consider, but I don’t have any good answers at the moment. There is not enough evidence from what I have seen to act immediately and make significant changes to the way we price and set attachment points. Rather, we will continue as we have and make small changes based on what we observe going forward. On the Aggregate side, there will be months and possibly years of significant enrollment changes and “lumpy claims” by month, which will make setting aggregate attachment points more challenging – though not impossible.
We have given our underwriters a guide to help them navigate this uncertainty, but our primary message is to stay the course, to not make significant changes and to use good underwriting practices (multiple methods, peer review, good communication) to work through the process.
We have not changed any of our pricing guidance. We are not enacting rate increases for the unknown. We will continue to observe the situation, and as more data becomes available, we may need to make adjustments where appropriate. That said, we remain open for business and committed to assuming risk in a responsible and thoughtful way. Our commitment to provide a quality product for our clients remains unchanged.