Note(s): The dependent variable for all models is p. Variables are defined in the  Appendix. For multivariate regression results, two-tailed p-values based on robust standard errors clustered by firm and year (Petersen, 2009; Cameron et al., 2011) are reported in brackets, with a correction where the covariance matrix is not positive semi-definite. ***, ** and * denote significance at the 1%, 5% and 10% levels respectively. The Vuong test (Vuong, 1989) is directional, so that a negative test statistic indicates that the second model is superior to the first