10
We start by introducing the dependent variables to be used in our empirical analysis; they
are firm performance and disclosure levels. Firm performance is used to test the first two
hypotheses, and disclosure levels are used to test the third hypothesis. The definitions of
dependent variables are provided next.
Dependent Variables
Firm Performance is measured by Tobin’s Q, the ratio of the market value of a firm’s
debt and equity to the current replacement cost of its assets. This measure is a proxy for firm
growth and has been widely used in previous corporate governance studies (e.g., Lindberg and
Ross, 1981). Woidtke (2002) argues that Tobin’s Q would reflect market valuation adjustments
associated with the effects of institutional investor holdings. We argue institutional investors’
holdings are adjusted depending upon insider ownership levels. To handle the presence of extreme
values, firm measures of Tobin’s Q are winzorized. The values in the bottom and top one percent
are set to the mean values in the adjacent percentiles.
Disclosure levels are measured by an index. We construct a disclosure index that captures
the behavior of a firm’s board of directors regarding transparency and the disclosure of pertinent
information in the following areas 1) external auditor’s recommendations; 2) calendar of activities;
3) accounting policies; 4) interim reviews; 5) financial forecasts; 6) regulatory requirements; and
7) the appointment of advisors. We review each firm’s annual financial reports from 1996 to 2005
and measure disclosure levels using the following score system: 1 indicates the item is not
disclosed, 2 indicates the item is slightly disclosed, 3 indicates the item is fairly disclosed, 4
indicates the item has been more than fairly disclosed but there is still certain information missing,
and 5 indicates the item has been fully disclosed. Haat et al. (2006) use a similar disclosure level
index in Malaysia to study corporate governance. In Chile, we use a slightly modified version.
Convention