Abstract Previous research on corporate average effective tax rates (ETRs) has consistently shown that ETRs vary across firms and over lime and has explained this finding solely in terms of the cross-sectional and intertemporal differences in firms' "tax preferences." This paper challenges the completeness and reliability of this explanation by developing and empirically testing a model of the ETR which shows that variations in ETRs are caused by differences in both tax preferences and income whenever tax preferences and Income are not perfectly correlated. Further, it thaws that cross-sectional differences in intra-industry ETR variances are caused; not by differences in the variation of tax preferences alone, but by differences in the variation In income and in the level of tax preferences and Income as well.
Patrick J. Wilkie (1988) studied this question.