Throwing Good Money After Bad: Board Connections and Conflicts in Bank Lending

Working Paper: NBER ID: w8694

Authors: Randall S. Kroszner; Philip E. Strahan

Abstract: This paper investigates the frequency of connections between banks and non-financial firms through board linkages and whether those connections affect lending and borrowing behavior. Although a board linkages may reduce the costs of information flows between the lender and borrower, a board linkage may generate pressure for special treatment of a borrower not normally justifiable on economic grounds. To address this issue, we first document that banks are heavily involved in the corporate governance network through frequent board linkages. Banks tend to have larger boards with a higher proportion of outside directors than non- financial firms, and bank officer-directors tend to have more external board directorships than executives of non- financial firms. We then show that low- information cost firms -- large firms with a high proportion of tangible assets and relatively stable stock returns -- are most likely to have board connections to banks. These same low- information cost firms are also more likely to borrow from their connected bank, and when they do so the terms of the loan appear similar to loans to unconnected firms. In contrast to studies of Mexico, Russia and Asia where connections have been misused, our results suggest that avoidance of potential conflicts of interest explains both the allocation and behavior of bankers in the U.S. corporate governance system.

Keywords: No keywords provided

JEL Codes: G3


Causal Claims Network Graph

Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.


Causal Claims

CauseEffect
board linkages (Y80)information flow (D83)
board connections (Y80)increased borrowing from connected banks (F65)
connected banks (G21)loan terms similar to unconnected firms (G32)
conflicts of interest (G34)cautious lending behavior (G21)
regulatory environment (G38)lending behavior (G21)

Back to index