Three essays in empirical finance.

Summary This thesis consists of three separate chapters. In the first chapter, we examine whether the explanatory factors for debt maturity previously identified in the literature have impacts that vary with the level of debt maturity, focusing on the extreme cases. We find that the effects of the classical determinants vary significantly with the distribution of debt maturity. These effects are much weaker for the lowest and highest percentiles. This indicates that the refinancing risk is much more binding in the very short term and much less so in the very long term. On the other hand, the fact of having access to public financing or not accentuates this phenomenon of heterogeneity in the impact of the determinants according to the level of maturity of the debt. This last point can be explained by the fact that the refinancing risk is much greater for firms without access to public financing. In summary, our results confirm our intuition regarding the heterogeneous impacts of the determinants of debt maturity as a function of the level of debt maturity and in particular in extreme cases. In the second chapter, we examine firms' debt maturity choices from a dynamic perspective. First, our results highlight sheep-like effects. Both in terms of debt maturity levels and in terms of changes in debt maturity, firms replicate the behavior of firms in the same sector. This herding behavior explains much more of the variation in debt maturity than the firms' own characteristics. After removing the impact of changes in the term structure of interest rates, this herding behavior in response to changes in the debt maturity of firms in the same sector is even more consistent. Second, we find persistent levels of debt maturity over time, especially for firms with very low debt maturities. The third section analyzes the impact of market timing on debt maturity. We argue that large firms with strong fundamentals tend to issue long-term debt rather than short-term debt in the event of a temporary overvaluation of these firms' securities. In particular, for such companies, the effect of timing dominates that of herding behavior during periods of heavy refinancing. For small companies with weak fundamentals, the effect of market timing is small, while the effect of herding is large.