Over the past twenty years many actuaries have claimed and argued that the chain-ladder method of loss reserving is biased; nonetheless, the chain-ladder method remains the favorite tool of reserving actuaries. Nearly everyone who acknowledges this bias believes it to be upward. Although supporting these claims and beliefs, the author proposes herein to deal with two deeper issues. First, does something inherent in the chain-ladder method dispose it to bias? Is there a diagnostic whereby one can predict how the chain-ladder method will fare with a particular loss triangle? To resolve this issue basic regression theory will suffice, specifically, the much misunderstood concept of regression toward the mean. And second, what lessons can we learn from the phenomenon of bias; in particular, is there a difference between actuarial methods and statistical models? These two issues constitute the reason and meaning of chain-ladder bias.
NOTE: Article updated April 1, 2008.
