the same individual is both the CEO and chairperson of the board)
are related to ICD. Subsequently, in the United Kingdom, Li et al.
(2008) confirmed these relationshipswith the exception of CEOduality.
It appears that the study of Hidalgo et al. (2011) is the only empirical
study that explored the relationship between corporate governance
and the extent of ICD in a developing country (i.e., Mexico). They also
introduced a newcorporate governance attribute, that is, family ownership.
Their findings suggested that family ownership does not influence
the extent of ICD in Mexican companies. One possible reason for this
result is that Hidalgo et al. (2011) did not test for a non-linear relationship
between family ownership and the extent of ICD; however, this
relationship can be established via a quadratic specification of family
ownership (which was undertaken in this paper).
This study was conducted in Bangladesh. Scarce empirical evidence
exists in respect of the relationship between corporate governance
attributes and the extent of ICD.4 Further, for the most part, previous
Bangladesh studies have been descriptive in nature and used relatively
small samples; for example, Ali, Khan, and Fatima(2008) found that ICD
is mostly disclosed in a narrative form in annual reports and concluded
that Bangladeshi companies do not have a positive approach to
reporting ICD. Similarly, Khan and Khan (2010), Nurunnabi, Hossain,
and Hossain (2011), and Rashid (2013) found that Bangladeshi companies
provide limited ICD in annual reports, as it is not a mandatory
requirement to report on this in Bangladesh. In a recent study in the
pharmaceutical industry, Abhayawansa and Azim (2014) found that