The widespread theft of collection materials, including rare and unique items, continues to be an issue of great concern to libraries of all types. The potential loss of such items threatens not only an institution's operations but, in many cases, global cultural heritage. Despite an increasingly open attitude among institutions regarding sharing information about lost items and suspected perpetrators, little scholarship has examined such thefts quantitatively in an effort to draw conclusions about how such incidents occur and how best to prevent them. This paper describes a project that examines data from over twenty years of reported library theft cases in libraries and special collections to determine how frequently such losses are perpetrated by library insiders.