Examining classical capital structure models of debt utilization decisions in Norwegian SME shipping companies

Abstract

This study aims at identifying variables that have a notable impact on the choice of leverage ratio in 161 Norwegian small and medium-sized shipping companies during the period from 2008 to 2019. We apply linear, logistic and machine-learning (ML) regression models. We interpret our findings in terms of the three major theories of companies’ capital structure choices: The Trade-off Theory, the Pecking Order Theory, and the Market Timing Theory of Capital Structure. Importantly, the ML models support the results of the naive regression models. We find that tangibility has the highest impact on leverage ratio and identify a positive relationship between tangibility and leverage. Also, several models agree on a negative relationship between profitability and leverage. The Logistic Regression (logit) model provided full support to The Pecking Order Theory. Generally, the models lend some support to all three theories of capital structure choice.