Lee, Darren David, Robert W. Faff, and Kim Langfield-Smith. "Revisiting the CSP/CFP link: when employing corporate sustainability as a measure of CSP." Working Paper, University of Queensland Business School, 2007.

This complex study tests two hypothesis: First,"the accounting performance of leading CSP [corporate social performance] firms is expected to be no different from those of lagging CSP firms. This is consistent with the view of Ullman (1985), who argues that in light of the many intervening variables that exist between social and financial factors, there is no reason to assume that a firm's level of CSP will noticeably affect its financial performance. However, even if there were a relationship between a firm's CSP and financial performance, it would remain largely undetectable due to the many intervening CSP/CFP variables." Second, market returns of high CSP firms are expected to be lower - "if we recognize that being a leading CSP firm has value, as reflected in a higher share price, which is not captured by standard asset pricing models, then leading CSP firms should have lower ex-post returns."

The authors view as inadequate indexes and social screening techniques that employ "negative exclusionary screens," and therefore use as a CSP proxy membership in the Dow Jones Sustainability index. Equal-weighted portfolios of firms from the Dow Jones Sustainability index and Dow Jones Global index are constructed and analyzed for the 1998-2002 time period, using a variety of regression analyses, including a matched companies analysis. Key findings:

1) "Accounting tests results find no difference with regard to the performance of leading and lagging CSP firms."

2) "Financial markets are able to distinguish between leading and lagging CSP firms, with this being reflected in the lower returns of CSP firms," although this finding was seen only in the matched dataset.

3) "Neither the accounting nor market tests support a systematic difference with regard to leading/lagging industry sorted portfolios."