Can the trade-off theory better match empirical evidence by implementing mean reverting earnings? A theoretical study on the trade-off theory

Abstract

The traditional trade-off theory predicts a positive relation between earnings and leverage ratio, contradictory to empirical evidence which suggests a negative relationship between the two. The objective of this dissertation is to analyse whether introducing mean reverting earnings to the traditional trade-off theory turns the model results more in line with empirical evidence.In the dissertation, I implement the Sarkar and Zapatero (2003) model, which considers mean reverting earnings before interest, taxes and depreciations in a setup similar to the proposed by Leland (1994). I use this model to calculate the optimal leverage ratio. The original input parameters presented by Sarkar and Zapatero (2003) were considered and equal results were obtained. The model was solved numerically for the optimal default barrier, optimal coupon level, and finally for the optimal leverage ratio. Furthermore, different initial earnings levels, different speeds of mean-reversion and different mean earnings-levels were considered to observe how the model handles different input values. As argued by Sarkar and Zapatero (2003), the consideration of mean-reversion leads to a negative relationship between the optimal leverage ratio and earnings, which is in line with empirical evidence. This result should change the generalized idea that the trade-off theory is not consistent with data regarding the earnings-leverage relationship.

Date of Award	6 Jun 2022
Original language	English
Awarding Institution	Universidade Católica Portuguesa
Supervisor	Nuno Silva (Supervisor)