This preview shows page 1. Sign up to view the full content.
Unformatted text preview: might
be established unconditionally on ﬁrm characteristic factors. In other words, the crony view
suggests that the crony relationship might overwhelm the ﬁrm characteristic factors. 21 To explore this issue more systematically, we follow the methodology used by Kroszner and
Strahan (2001a and 2001b). We use a probit model in which the dependent variable is one if
the ﬁrm has at least one person on its board connected to those of banks. The probit regression
results on the determinants of the incidence of the board connections are shown in Table 10. In
Speciﬁcation (1), we test how the board connections varies with ﬁrm characteristics. The ﬁrm
characteristics are the same set of variables as described in Section 3.3. The reported results
are the marginal eﬀects of a one unit change from the mean of each independent variable on
the probability of having a board connection with banks.
Similar to the US banks shown in Kroszner and Strahan (2001a and 2001b), Thai banks
also have the connections with larger ﬁrms and less indebted ﬁrms. Asset tangibility, growth
and sale volatility, however, are not related to the probability of having the board connections.
These results somewhat supports the conﬂicts of interest argument.
Next, we investigate whether the crony relationship attributes to the incidence of having
the board connections. In Speciﬁcation (2), we include two dummies representing ﬁrms that
are owned by the two groups of the inﬂuential families: Inﬂuential families with bank and
Inﬂuential families without bank. Interestingly, our results show that the connections are the
most prevalent at the crony ﬁrms aﬃliated to these two groups of inﬂuential families. Further
investigation on the data reveals that ﬁrms owned by the inﬂuential families who own banks
are always connected to the boards of banks. In most of the cases, the persons who serve on
the board of banks and ﬁrms are the ﬁrms’ controlling shareholders and their families.
We investigate further to check whether the eﬀects of ﬁrm characteristics on the likelihood
of having the board connections attenuate or even disappear in the case of crony ﬁrms or
not. To absolutely distinguish the ownership eﬀect from the crony eﬀect, we examine the case
when ﬁrms that are aﬃliated to the inﬂuential families that do not own banks. Similar to the
methodology used in Section 4.2, we interact Inﬂuential families without bank with the ﬁve ﬁrm
characteristics. Consistent with the crony view, the incidence of having the connection with the
inﬂuential families are almost not related to the ﬁrm characteristics (Speciﬁcation (3)). When
banks build a connection with non crony ﬁrms, banks appear to have the board connections
with large and less indebted ﬁrms. In contrast, banks seem to pay less attention to ﬁrm’s size
and the level of indebtedness when they establish a connection with the crony ﬁrms. Banks
appear to have the connection with ﬁrms that owned by the inﬂuential families even if they
are smaller. The eﬀect of leverage on the presence of the connections disappears when the
22 inﬂuential families are concerned.
To test the robustness of this...
View Full Document
- Fall '13