analysis. In addition, characteristics of the board of directors, such as board size, board composition and interlocking
directorates were collected to measure the degree of independence of the board of each company.
Dependent Variable - Expropriation of Minority Shareholders’ Rights
Measuring expropriation of minority shareholders’ rights can be difficult given its numerous definitions and
manifestations. However, emerging markets research suggests that concentrated ownership is correlated with a lack of
investor protection (Claessens, Djankov, & Lang, 2000; Denis & McConnell, 2003; La Porta, Lopez-De-Silanes, & Shleifer,
1999; La Porta et al., 1998; Shleifer &Vishny, 1997). Thus, a measure of ownership concentration of the top five
blockholders may be used to assess the degree of expropriation of minority shareholders’ rights.
Following prior research on ownership concentration (Claessens et al., 2000; Demsetz & Lehn, 1985; Goergen &
Renneboog, 2001; Nenova, 2003), this study utilizes the Herfindahl index (HI) to measure ownership concentration and to
proxy for the expropriation of minority shareholders’ rights. This index captures both the inequality of shares among
stockholders and the number of shareholders as well as better reflecting the true levels of ownership concentration in any
company (Barabanov & McNamara, 2002). The HI is usually calculated as the sum of squares of the shareholdings of a
certain number of stockholders. For this study, the top ultimate share blockholders holding at or above the five percent level
is used.
To calculate the HI, the ultimate ownership of both direct and indirect control and cash flows rights of each firm in
the sample was traced for each of the three years (2000 – 2002) under consideration. To do so, the direct ownership of
control rights for all owners with stakes at or above a five percent threshold was determined. Then the ultimate control of
these direct owners was traced, using the same threshold of five percent ownership. Form 20-F usually traces the identity of
ultimate control owners with at least five percent ownership. When the Form 20-F did not provide the necessary information
for determining the ultimate owners, other sources of information were utilized, such as the web pages of the companies.
Once these ultimate owners are identified, control rights are determined for the sample company and categorized
into one of the following groups: family-management ownership group, non-affiliated company ownership group,
government ownership group, institutional ownership group, individual ownership group, and miscellaneous ownership
group, following Lins (2003). Included in the first group are family members (based on overlapping last names or known
familial relations) and any management official or director. The non-affiliated company ownership group consists of
ownership by other companies not affiliated with family-management. Direct and indirect ownership by all the agencies and
companies identified as state-owned comprise the government ownership group. The institutional ownership group includes
ownership held by pension funds, insurance companies, and/or banks. Individuals who are not family members or mangers
Convention