This paper quantifies the causal effect of borrowing cost on firms’ investment decisions. To overcome the empirical challenge due to a possible reverse causality where firms’ investment prospects affect their borrowing costs, I apply an instrumental variable methodology where the identification comes from insurance companies’ regulatory constraints regarding the credit rating of their bond holdings. Rating- based regulatory constraints are more binding for insurers with a weaker capital position. For this reason, bonds upon downgrades face different degrees of selling pressure depending on the different capital positions of their holders. Such differences are presumably not correlated with issuers’ investment prospects. Using data from 2004-2010, I estimate that a one percentage-point increase in bond spread reduces investment during the same year by 12 percent. Moreover, a five percentage-point increase in bond spread halves the probability of new debt issuance.
This paper quantifies the causal effect of borrowing cost on firms’ investment decisions. To overcome the empirical challenge due to a possible reverse causality where firms’ investment prospects affect their borrowing costs, I apply an instrumental variable methodology where the identification comes from insurance companies’ regulatory constraints regarding the credit rating of their bond holdings. Rating- based regulatory constraints are more binding for insurers with a weaker capital position. For this reason, bonds upon downgrades face different degrees of selling pressure depending on the different capital positions of their holders. Such differences are presumably not correlated with issuers’ investment prospects. Using data from 2004-2010, I estimate that a one percentage-point increase in bond spread reduces investment during the same year by 12 percent. Moreover, a five percentage-point increase in bond spread halves the probability of new debt issuance.
(0)
(0)