This table reports estimation results from probit regression models. The dependent variable ΔPD equals 1 if the estimated probability of default increases for firms that default or decreases for firms that do not default, and 0 otherwise, when we add four factors on business credit information to the baseline default prediction model. Explanatory variables are split into tercile dummies (T2, T3) with tercile T1 as reference category to allow for non-monotonic effects (except LIMITED_LIABILTY which is a dummy variable). We include dummy variables for years, the first two digits of the industry classification code and the first two digits of the ZIP Code in all models to control for year, industry and regional fixed effects. Marginal effects (mfx) are reported as average marginal effects. We employ robust standard errors clustered within firms.
*Significance at the 10%-level.
⁎⁎
Significance at the 5%-level.
⁎⁎⁎
Significance at the 1%-level.
Table options