Abstract: We present evidence on the duration of firm-bank relationships using a unique panel data set of connections between Oslo Stock Exchange-listed firms and their banks for the period 1979-1994. We focus on the determinants of the duration of a relationship and the causes for ending an existing bank relationship. We find that duration itself does not greatly influence the likelihood of ending a relationship: short-lived relationships are as likely to end as long-lived relationships. We also find firms that maintain simultaneous multiple-bank relationships are more likely to end a bank relationship than a single-bank firm and that small, highly-leveraged "growth" firms are more likely to end a bank relationship than large, low-leveraged "value" firms.
Keywords: banking relationships, hazard models, duration analysis JEL Codes: G21, C41