17
PEB
= a dummy variable, 1 if the firm has an AFP representative on the board, or 0
otherwise
Table 6 shows the results of the robustness tests. We find that neither changes in insider
ownership levels nor changes in institutional ownership levels are associated with changes in firm
values (see Table 6, column 1).
We find changes in insider ownership levels are positive and
statistically significantly (and linearly) associated with changes in firm disclosure levels (see Table
6, column 2). Finally, a regression (not reported) of the changes in insider ownership levels on
changes in private pension funds or AFP ownership levels, using the same control variables, finds
that there is a negative and statistically significant relationship between them. This suggests AFPs
may prefer holdings in firms with lower insider ownership levels. It is in these firms that
institutional investor ownership has a positive effect on firm performance.
Therefore, we find
evidence of an endogenous relationship between insider ownership levels and institutional investor
ownership levels based on disclosure levels but not on firm performance.
CONCLUSIONS
Our findings support the corporate governance literature. We find that institutional
ownership levels of publicly traded firms in Chile positively and significantly affect a firm’s
performance, and such effect is stronger when ownership concentration is below 60 percent. The
results also indicate that when insider ownership levels are above 60 percent, they have a positive
effect on firm performance; however, below 60 percent, they have a negative effect on firm
performance. Therefore, we conclude that at high insider ownership levels, there is greater
convergence of interests between inside and outside shareholders. As such, the role of institutional
investors appears to be relatively more important when insider ownership levels are below 60
percent.
Convention