Sharfman and Fernando (2007)

Sharfman, Mark P. and Chitru S. Fernando. 2008. "Environmental Risk Management and the Cost of Capital." Strategic Management Journal, 29: 569–592,
DOI: 10.1002/smj.678.

This study estimates the impact of environmental risk management activities on firm cost of capital, leverage, and equity ownership for the 1999-2002 time period. The companies analyzed were publicly-traded U.S. firms (subset of the S&P 500).

The independent environmental variable is derived from a varimax rotation of a subset of the EPA Toxic Release Inventory (TRI) data compiled by IRRC, and KLD environmental ratings.

Eight distinct hypotheses and sub-hypotheses are tested:

- Perhaps the most important hypothesis is that companies with strong environmental risk management will have a lower weighted average cost of capital. Four weighted cost of capital measures are developed to address this question: WACC-1 uses firm betas calculated (annual data) by Research Insight, from the Compustat database. WACC-2 uses Bloomberg weekly betas. WACC-3 recalculated WACC-1 using Bloomberg's firm-specific risk premium. These are then aggregated using a varimax rotation to create a summary measure, WACC-4. WACC-1 and WACC-4 are the measured used in the regressions, and the authors note that "this hypothesis was supported regardless of the WACC measure."

- That companies with strong environmental risk management will have higher financial leverage. This is confirmed in the analysis: "firms with better environmental risk management benefit by being able to carry higher levels of debt."

- That companies with strong environmental risk management will have a lower cost of debt. This was not confirmed --> "Our results suggest a significant positive relationship between the cost of debt and environmental risk management, even after controlling for size, leverage, and industry effects... Because leverage also increases with environmental risk management, it is possible we are unable to control properly for the effect of leverage on the cost of debt."

- That companies with strong environmental risk management will have a lower cost of equity capital. "...[I]n both equations the coefficients for the indicator are in the correct direction and significant."

- That companies with strong environmental risk management will have a lower non-levered beta. This is supported by a separate regression, but also by definition, given the equity cost of capital result.

- That companies with strong environmental risk management will have more dispersed share ownership. The coefficient was in the correct direction, but only significant at the 10% level.

- That companies with strong environmental risk management will have more institutional shareholders. "Our test of [this hypothesis] did not yield significant results."

The study also includes a useful conceptual map of the hypothesized linkages.