Risk Changes around Calls of Convertible Bonds

Abstract

We examine changes in equity and asset betas around convertible bond calls and report two major findings. First, calling firms exhibit an increase in asset betas following the call. We argue that the finding is consistent with the implications of the sequential financing theory but not of the backdoor equity financing theory. Second, abnormal returns at call announcements are negative only for the subsample of firms that also exhibit an increase in equity beta. We conclude that risk changes help explain the market reaction to convertible bond calls. Copyright (c) 2010, The Eastern Finance Association.

Calls of convertible debt securities: no bad news at all

“…A call event is excluded if (5) the underlying company is in the regulated financial or utility industries. 5 I follow Garcia-Feijóo et al (2010) and King and Mauer (2014) and drop (6) potentially disturbing covenants such as variable coupon bonds and liquid yield option notes, because the latter are putable for cash, which can give rise to different motives for calling the bond. Issuers sometimes clean up their convertibles after most of the holders have already converted so as to avoid small outstanding amounts.…”

“…For example, Brick et al (2007) employ 392 calls made over 19 years (1985)(1986)(1987)(1988)(1989)(1990)(1991)(1992)(1993)(1994)(1995)(1996)(1997)(1998)(1999)(2000)(2001)(2002), an extension of Ederington and Goh (2001) sample. Garcia-Feijóo et al (2010) employ 165 non-financial calls between 1986 and 2001. In contrast to this literature, King and Mauer (2014) use a sample of non-financial bonds issued between 1980 and 2002.…”

“…Certain covenants can generate special incentives to call the bond. I follow Garcia-Feijóo et al (2010) and exclude putable and variable coupon bonds. The opportunity to call a convertible bond generates financial flexibility; for example, it allows responding to cheaper refinancing possibilities due to decreased credit risk.…”

“…The mean abnormal returns of the Dirty sample (Panel A of Table 4) are consistent with the literature. For example, Brick et al (2007) report mean abnormal returns of −0.75 %, Bechmann (2004) of −1.75 %, andGarcia-Feijóo et al (2010) of −1.07 %. The results for the median and the winsorized sample reveal no large outliers as well as a symmetric distribution of the abnormal returns.…”

“…King and Mauer (2014) report that about 31 % of their convertible sample drops due to contemporaneous news. Indeed, several papers (e.g., Garcia-Feijóo et al 2010) find evidence for Mayers (1998) sequential financing hypothesis: companies issue callable bonds to overcome agency conflicts of debt and call their bonds to exercise investment options. The announcement of investment options can be simultaneous with the call event.…”