Frontiers of Entrepreneurship Research
1996 Edition

Back to Index96
Order hard copy editions of Frontiers of Entrepreneurship Research by mail


(1) Jan P. Warhuus
(2) Jonathan Levie

(1) Southern Denmark Business School
Engstein 1
6000 Kolding, Denmark

(2) INSEAD, Dept. of Entrepreneurship and Leadership
Boulevard de Constance
77305 Fountainebleau, France

(2)+33 1 60724133

(2)+33 1 60724242

Principal Topics

Though banks are the most common source of external finance for young ventures, relatively little research has been conducted on the entrepreneur/banker relationship. This paper reports on an exploration of this relationship in Denmark, which has a relatively fragmented banking industry. The research question was: does bank size predict bank credit agreements for young growing firms? Both sides of the entrepreneur/banker relationship in a sample of 11 young growing ventures stratified by bank size were explored in a multi-method study through interviews with their entrepreneurs and their bankers. The interviews were guided by hypotheses drawn from economics and organisation behavior literature, but the inductive study was also undertaken.


The 11 firms include 8 profitable and 3 unprofitable young firms from different manufacturing industries which had grown to employ at least 50 people by 1995. The ventures were stratified by size/age of firm and size/catchment area of bank. In-depth, semi-structured interviews were conducted with both the entrepreneurs and their bankers. For the deductive study, four hypothesis were tested on the small sample case data using the Mann-Whitney U test and Fisher’s exact test. For the inductive study, emergent patterns from the case data were analyzed visually, and Fisher’s exact test was used as an aid to pattern recognition.

Major Findings

We found that for this small sample, bank size, or more precisely, bank catchment area, negatively predicts credit agreements for start-up. In other words, the entrepreneurs in our sample had a significantly higher chance of a credit agreement at start-up, but not during periods of growth or distress, if they went to a small local bank than any other bank. The inductive study provided a plausible explanation: local bank managers take the job-creation and other potential spin-off benefits to the local company into account when assessing credit requests. This assessment criterion is to the advantage of new high potential ventures rather than "the eight pizza bar in town or the third hairdresser in the street" type of new venture(quote from large bank branch banker). Larger branch bank managers do not take this factor into account. We expected, but did not find, statistical evidence of an association between credit agreement and by level of information that the banker had on the entrepreneurs’s business, or between credit agreement and social familiarity between the banker and the entrepreneur. Bank size did not associate strongly with either information level or social familiarity as measured. All bankers reported a high degree of decision-making authority. This is also contrary to our expectations.


The results suggest that entrepreneurs of high potential ventures should exploit their job-creating potential when they approach a local banker for start-up finance. Large branch bank managers claim to have much higher decision-making authority than in the past, so it now seems possible for entrepreneurs to directly influence decision-makers in large Danish banks. The banker’s first few meetings with the entrepreneur appear to be crucial to his or her perception of the entrepreneur as someone who is competent and trustworthy. Careful preparation for these meetings should help the entrepreneur’s chances further.

The decline in the population of local banks in Denmark, given their apparently significant role in backing new high potential ventures, should be a source of concern.

The theory we used is either inadequate to explain the results of our deductive study, or our sample is biased, or our construct measures are poor. We argue that our sample, though it is small and would benefit from enlargement, replicates variability in the population. We have tested other construct measures for the constructs and obtained similar results. Possibly, in a large sample study, information and familiarity will be shown to have significant, if weak, effects. Our inductive study suggests that the central finding, that credit agreements do vary significantly by bank size, is valid. The theoretical explanation for this association is however not supported by the inductive study. A refined model is proposed for large scale testing. We also propose to replicate the study in other nations with different financing systems and where centralization of decision-making varies between banks.

Back to Index96