This preview shows page 1. Sign up to view the full content.
Unformatted text preview: long term loans might be attenuated or disappear if ﬁrms have strong
connections with banks. In this section, we attempt to shed light on this issue. Speciﬁcally, we
investigate whether the ﬁrm-bank ties have any inﬂuence on the choices of long term loans via
To test this issue, we need to simultaneously incorporate the crony and the ﬁrm characteristic variables. We re-estimate the regressions including the interaction terms between ﬁrm-bank
connection variables and ﬁrm characteristics. The coeﬃcient on a given interaction term measures how the relation between the choice of long term debt and the relevant ﬁrm characteristic
diﬀers for ﬁrms with and without close connections. If the crony relationships overwhelm the
eﬀects of ﬁrm characteristics on the decisions of long term loans, then the estimated coeﬃcient on an interaction variable for a ﬁrm characteristic should be opposite in sign from the
First, we investigate the eﬀects on the ﬁrm-bank connections that is measured by the inﬂuential families. The results of this regression are presented in Table 8. In column (1), (2),
and (3), the ﬁrm characteristics are interacted with Inﬂuential families, Inﬂuential families with
banks, and Inﬂuential families without banks. Overall, the results provide stronger support for 18 the crony lending hypothesis. Interestingly, for non crony ﬁrms that are not aﬃliated to the
inﬂuential families, size appears to be a very important determinant of long term lending. The
estimated coeﬃcients on ﬁrm size are strongly signiﬁcant at the 1 percent level in all models.
However, the coeﬃcients on the interaction term between Inﬂuential families dummies and size
are consistently insigniﬁcant. The results reveal that ﬁrms connected to the inﬂuential families
that both own and do not own banks and ﬁnance companies are able to borrow long term
independently of their size.
Similar to the Mexican lending behavior shown in La Porta et al. (2001), the long term
loans borrowed by ﬁrms connected to the inﬂuential families appear to backed by less collateral.
While the coeﬃcients on the ﬁxed asset ratio are positive and consistently signiﬁcant at the 5
percent level, the estimated coeﬃcients on the interaction terms between the ﬁxed asset ratio
and the dummy variables indicating the eﬀects of the inﬂuential families either turn out to be
signiﬁcantly negative (in Speciﬁcation (1)) or insigniﬁcant (in Speciﬁcation (2) and (3)). This
evidence suggests that long term loans to non crony ﬁrms are always collateralized. In contrast,
the inﬂuential families seem to either post less collateral against their loans or post no collateral
The results in Speciﬁcation (2) showing the eﬀects of the inﬂuential families that control
ﬁnancial institutions are also interesting. Speciﬁcation (2) presents the results on the eﬀects of
the inﬂuential families who own banks. While default proba...
View Full Document
This document was uploaded on 02/28/2014 for the course ECONOMICS fn314 at Harvard.
- Fall '13