Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
This comment highlights different ways of coding crisis episodes in Cerra and Saxena (2008) (CS). The comment shows that the coding used for civil war implies a misrepresentation of its impact. A correct coding of civil war reveals that the average civil war leads to a loss in output of 18 percent. This makes civil wars more devastating than all other crisis studied by CS.