For some time, researchers have been investigating the relationship between a firm's corporate financial performance (CFP) and its corporate social performance (CSP). Although most studies indicate that CSP is a determinant of CFP, other aspects of this research have been inconsistent. Some studies are criticized for using unreliable CSP measures; others for missing control variables; and still others for assuming linearity without valid testing. This paper responds to these issues with an updated study of the CSP–CFP relationship, testing two approaches to measuring CSP, controlling for key variables identified in the literature, and testing for nonlinearity of certain independent variables. Chief among our findings is a positive CSP–CFP relationship, which supports proponents of stakeholder theory. We also determine that empirical models specifying two CSP component measures are stronger than those using a fully aggregated measure. Lastly, we find that control variables must be properly specified to avoid bias and that some of these measures are quadratically related to CFP. Copyright © 2009 John Wiley & Sons, Ltd and ERP Environment.