Table 6 examines the consistency of the Big 6 specialization premium across industries by estimating separately the premium for each industry group. Audit fee models will be better specified on an industry-specific basis if there are systematic differences in fees across industry. A potential limitation of such an approach is that the sample size in each industry is relatively small (50 to 100 observations) which lowers the power of the tests. However, the regression estimates reported in Table 6 are all significant and the adjusted R2s in each model exceed 65 percent. Coefficients on the industry specialist variables in table 6 indicate that the specialization premium is significant in both the property and consolidated enterprises industry groups,but is insignificant in the industrial industry group.The coefficient of 0.294 in the property group translates to a 34 percent specialist premium and the coefficient of 0.303 in the consolidated enterprise group translates to a 35 percent specialist premium.Table 6 is important because it demonstrates in a direct industry-by-industry test that an industry -specialization premium exists, at least for two of the three industry group in the study.To the extent the audit fee model is misspecified across industry groups, industry-specific tests will be more robust than a pooled-industry design.