There are 42 rms in the sample in which their

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: ndings, we drop 31 firms that are owned by the influential families that own banks from our sample and re-estimate the probit model. Our results remain the same (Specification (4) and (5)). In an unreported regression, we ran a test that controls for the size of the board. The regression was done by using the Tobit model, in which the dependent variable is the ratio of the number of persons from a firm’s board who are serving on the boards of banks divided by the number of persons in the firm’s board. Again, our results are robust in term of signs, statistical significance and magnitudes. To sum up, the presence of the board connections is consistent with the crony view. The probability of having the connections is strongly dependent on the crony relationship. When banks build connections with the crony firms, firm characteristics seem to be much less important factors. For non crony firms, however, banks tend to build a connection with large and less indebted firms in the same manner as in the US. [Insert Table 10 Here] 5.2 Board connections: The case of firms that do not own banks Having shown that the board connection is the most prevalent in the firms in which their controlling shareholders also own banks, we now investigate whether our results on the effects of the crony relationship on obtaining long term lending are robust if we exclude such firms from our sample. By excluding the firms where their controlling shareholders own banks and finance companies, we are able to discern the effects of the crony relationship from the effects of the bank ownership. We repeat the regressions in Section 4.1 and Section 4.2 but exclude 31 firms in which the controlling shareholders own financial institutions. The results are reported in Table 11. Consistent with the previous findings, the results reveal that the board connections both via executive and non executive levels are indeed beneficial to the firms. The connected firms seem to be able to obtain more long term loans. We examine further whether the connections that bring favoritism to the firms are affected by the two characteristics of the connections: When the firms’ controlling shareholders sit on the boards of banks, and when the board connections are formalized through other persons who are not the controlling shareholders. There are 42 firms in the sample in which their controlling shareholders and their families serve in the boards of financial institutions, and 115 firms that 23 have either bankers on the firms’ board, or the firms’ executives and other directors serving on the boards of banks. To understand whether these two characteristics of the connections affect the amount of long term lending, we re-estimate the OLS regressions and include the dummy variables that indicates these two types of board connections. The results are shown in Table 12. Similar to the case of the firms controlled by the influential families with banks, the board connect...
View Full Document

This document was uploaded on 02/28/2014 for the course ECONOMICS fn314 at Harvard.

Ask a homework question - tutors are online