What I would say about that paper is that the methodology was incorrect and the data used was incorrect. They used numbers that included the disability business and a bunch of other things that are not related to supplemental health. We've communicated that to the authors, and they're well aware of that.
I would say that the difference shown there was predominantly due to falling interest rates and their implications for the disability side of the business. It really had nothing to do with the supplemental benefits.