Document Type

Article

Abstract

Lost in the recent efforts to take political advantage of (or explain away) the rapid rise in liability insurance premiums is any real attempt to understand the underwriting cycle, why it is so severe in medical malpractice insurance, and what it might mean for the ability of malpractice liability to deliver on its risk distribution, loss prevention, and corrective justice objectives. This essay attempts to fill that gap. Part 1 provides a primer on the liability insurance underwriting cycle that draws on the research prompted by the mid-1980s insurance hard market. This part explains that the recent dramatic increases in predicted medical malpractice losses are a result of the insurance cycle, not dramatic changes in medical malpractice claim payments. Part 2 explores why the underwriting cycle is so severe in medical malpractice insurance. This part applies my recent Geneva Lecture analysis of liability risks to the specific problem of medical malpractice, with a fuller consideration of the dynamics of the underwriting cycle. In short, the cycle is so severe because there is a relatively long period between the time that the premiums for a medical malpractice policy are paid and the time that losses under that policy can be known with certainty, and because there is more uncertainty regarding future medical malpractice losses than many other kinds of losses. Part 3 explores whether insurance regulators should consider acting to moderate the underwriting cycle (assuming that they could do so). This part argues that there are good reasons to believe that medical malpractice insurance crises lead medical providers to improve patient safety and, therefore, that efforts to moderate the cycle could have a negative impact on patient safety. Further research is needed before we can draw firm conclusions, but leaving the insurance cycle alone would be the wiser course for now.

Date of Authorship for this Version

May 2005

Keywords

insurance, medical malpractice, torts, underwriting cycle