This paper was originally published in Variance: Advancing the Science of Risk, vol. 1 no. 1, a publication of the Casualty Actuarial Society.

While accounting principles and actuarial standards of practice are all well designed, they provide only broad guidance to the actuary on what is "reasonable." This broad guidance is based on the principle that "reasonable" assumptions and methods lead to "reasonable" estimates. Unfortunately, this broad guidance can leave the low end of a range of "reasonable" reserves open to an interpretation that could lead to unintended consequences in practice.

This paper reviews some current actuarial practices and examines how they relate to the question of what is "reasonable" from a statistical perspective. Moreover, it reviews and further develops some statistical concepts and principles that actuaries can add to their repertoires when developing ranges and distributions of liability estimates and then evaluating the "reasonableness" of management’s best estimate within those ranges and distributions.