(HORIZON), and earnings volatility (STDROE) are the common control variables across
the two equations. Because auditor choice is an endogenous decision for each client
and the factors that determine the auditor choice can also influence the dependent variables
used in this study, we calculate the inverse Mills ratio (INVMR) and include it in all three
models in order to control for the possible endogeneity in the estimations.10 In addition, to
control for the presence of heteroscedasticity, we apply White’s (1980) heteroscedasticity
consistent standard errors for all regression analyses performed in this study. Furthermore,
we apply the firm-clustering technique for all the analyses because multiple observations
from the same firm (but from different years) are included in our dataset. We use the
following Equation (1) to test H1: