We thank Brent Ambrose, David P. Brown, Shingo Goto, Brad Jordan, Ken Kavajecz, Ajay Khorana, Steve Mann, Sattar Mansi, Ted Moore, Greg Niehaus, Mark Ready, Ellen Roueche, Martin Ruckes, Stathis Tompaidis, Toni Whited, an anonymous referee, and seminar participants at Georgia Institute of Technology, Notre Dame University, the University of South Carolina, the University of Wisconsin at Madison, the 2004 Financial Management Association annual meeting, and the 2005 Southern Finance Association annual meeting for helpful comments.
What Is the Cost of Financial Flexibility? Theory and Evidence for Make-Whole Call Provisions
Article first published online: 25 AUG 2008
© 2008 Financial Management Association International.
Volume 37, Issue 3, pages 485–512, Autumn 2008
How to Cite
Powers, E. and Tsyplakov, S. (2008), What Is the Cost of Financial Flexibility? Theory and Evidence for Make-Whole Call Provisions. Financial Management, 37: 485–512. doi: 10.1111/j.1755-053X.2008.00022.x
- Issue published online: 25 AUG 2008
- Article first published online: 25 AUG 2008
Firms commonly incorporate make-whole call provisions in their newly issued debt, presumably to improve their ability to retire debt early if circumstances require. In return for increased financial flexibility, firms must compensate bondholders with additional (incremental) yield. To estimate theoretical incremental yields, we use and calibrate a structural model for a large sample of callable and noncallable US corporate bonds issued between 1995 and 2004. In a frictionless model where calls occur only when they are in-the-money, theoretical incremental yields average approximately 2 basis points (bp). In an extended model that incorporates taxes, transactions costs, and randomly occurring exogenous events requiring early bond retirement, incremental yields average approximately 5 bp. Empirical analysis, however, indicates that observed incremental yields are significantly greater than model-generated values, averaging between 13 and 24 bp. In the later years of our sample period, however, observed incremental yields begin to converge to model-generated values.