The financial health of Norwegian crowdlending borrowers : comparing crowdlending borrowers and borrowers in traditional banks by applying three bankruptcy prediction models

The purpose of this thesis is to examine whether companies obtaining loans at Norwegian crowdlending platforms are more likely to default than companies obtaining loans in traditional Norwegian banks. Crowdlending, the concept of lending directly from large groups of investors through a digital platform, has experienced an explosive growth in Norway and worldwide over the last years. Despite this, research regarding this topic is limited. To the best of our knowledge, this is the first study to examine the financial health of Norwegian crowd borrowers. In our empirical study, we use three proven models for prediction of bankruptcy and analyze whether Norwegian crowd borrowers’ credit quality is different than that of regular Norwegian borrowers. The analysis is conducted on a dataset constructed from manual collection of Norwegian crowd borrower information from the three Norwegian crowdlending platforms Monner, FundingPartner and Kameo. To create an approximately randomized experiment with suitable comparable traditional borrowers we applied the matching approach of Coarsened Exact Matching. Based on our analysis, there is not enough evidence to suggest that Norwegian crowdlending borrowers are riskier than borrowers in traditional banks. Although the mean and median credit scores of crowd borrowers are generally worse than those of the regular borrowers, thus suggesting a difference in credit quality between the groups, these differences prove to be statistically insignificant.