Disclosure Level and Expected Cost of Equity Capital : An Examination of Analysts' Rankings of Corporate Disclosure and Alternative Methods of Estimating Expected Cost of Equity Capital

This paper examines the association between expected cost of equity capital and three types of disclosure (annual report, quarterly and other published reports, and investor relations). Our sample consists of 3,620 firm/year observations with Value Line data, which are also included in the AIMR's Annual Reviews of Corporate Reporting Practices dated from 1985/86 through 1995/96. The disclosure rankings produced by the AIMR are employed to proxy for disclosure level. In addition, we examine four alternative estimates of expected cost of equity capital: one produced by the classical dividend discount formula (rDIV); one produced by the EBO valuation model and employed in Botosan (1997) (rEBO); one produced by a finite horizon specification of the Gordon growth model and employed in Gordon and Gordon (1997) (rGORDON); and one produced by the approach described in Gebhardt, Lee and Swaminathan (1999) (rGLS). We find that rEBO estimates are highly correlated with and economically similar to the theoretically correct rDIV estimates. rGORDON estimates are also highly correlated with rDIV, however the estimates produced by this approach significantly understate cost of equity capital. Finally, rGLS estimates are not highly correlated with rDIV and significantly understate cost of equity capital.We find that cost of equity capital is decreasing in annual report disclosure level. The magnitude of the difference in cost of equity capital between the most and least forthcoming firms is approximately %BD - 1% points. These results confirm and extend the results of Botosan (1997) to include larger, more heavily followed firms, across a diverse group of industries, over a number of years. Surprisingly, we find a positive association between cost of equity capital and the level of more timely disclosures, such as the quarterly report. The magnitude of the difference in cost of equity capital between the most and least forthcoming firms is approximately 1 - 2% points. This result, while contrary to that predicted by theory, is consistent with managers' claims that greater timely disclosures increase cost of equity capital, possibly through increased stock price volatility. Finally, we find no association between cost of equity capital and the level of investor relations activities. Based on our results we conclude that aggregating across different types of disclosure results in a loss of information and potentially erroneous conclusions