The Corporate Governance Structures of GLCs and NGLCs and Firm Performance in Malaysia

This study has examined the relationship between corporate governance structures and the performance of a sample of companies listed on the Malaysian Stock Exchange in the post-Asian financial crisis period beginning 2001 and ending in 2003. The sample was selected to provide matched-pairs of government linked companies (GLCs) and non-government linked companies (NGLCs), as it was anticipated that these groups would have different governance structures and that these differences might impinge on the association being studied. The empirical results indicate that there were eight statistically significant differences between the corporate governance structures of GLCs and NGLCs, thus supporting the contention that it would be useful to examine the association between corporate governance structures and performance for these two distinct groups separately. Accordingly, univariate and multivariate tests were performed on three samples: (i) GLCs; (ii) NGLCs and (iii) a combination of these referred to as All Companies. The multivariate tests were performed both on the dataset collected from published sources and this data transformed using the Van Der Waerden approach. Although tests for conformity suggested no significant statistical problems with the original data-set for multivariate analysis, it was hoped that the transformation would provide better models for explaining the relationship between corporate governance variables and performance. The multivariate analyses on both the standard and transformed data-set found no empirical evidence of a consistent relationship between corporate governance structures and performance measured by return on assets and return on equity in GLCs, NGLCs or the combined sample over the period. Statistically significant relationships were found across the groupings and for different performance measures but were not sustained across all years. The implications for corporate governance research of these inconsistent results across different measures of performance and different samples across the three year period is that empirical research may reach conclusions based on statistically significant results at a point in time that are only relevant for this historic context and may not persist. The results also indicate that, despite the eight differences in governance structures of GLCs and NGLCs, the observed differences in the performance could not be explained by their governance structures. This finding supports the view that these structures provide appropriate monitoring on company’s management rather than improving performance. These findings are consistent with the ambivalent position on the relationship between performance and corporate governance variables observed by the literature which exhibits conflicting arguments about the direction of relationships and empirical results that are extremely variable. This study also found that the relatively poor performance of GLCs in Malaysia which has been associated with government influence on the board structures such as the appointment of a bumiputra as director (BUM), a senior government officer as director (SGO) and a politician as director (POL) were basically unfounded because these variables have no statistically significant adverse impact on performance.