The zero-leverage phenomenon: A bivariate probit with partial observability approach

The empirical literature on zero leverage investigates why some firms are debt-free using standard logit and probit specifications. However, such models are not suitable to provide a direct answer to the main research question that arises in this context: is zero leverage a financial decision of the firm or an imposition raised by creditors? This paper examines the factors that affect the demand for debt and the supply of debt using bivariate probit models with partial observability in the sense of Poirier (1980), providing empirical evidence on the zero-leverage phenomenon for European listed firms during the period 2001-2016. We find that some variables influence in opposite directions debt demand and supply, or affect significantly only of them. In particular, firms' profitability affects negatively debt demand but positively debt supply; asset tangibility increases the willingness of creditors to grant debt but does not influence debt demand; and the recent European crises reduced the propensity of firms to resort to debt but did not affect debt supply. We also find that firms in countries with common law systems, market-based financial systems and stronger protection to investors' and creditors' rights are more prone to have zero leverage due to both demand and supply effects.