We thank the conference participants of the University of Illinois-Chicago Accounting Conference and workshop participants at the University of Notre Dame, Fordham University, Chinese University of Hong Kong for their comments and suggestions. A related version of this paper benefited from the comments of participants at the 2021 HARC Conference and the 2021 FARS Conference, as well as Darren Bernard, Russell Lundholm, Jason Huafeng Chen, Charles Lee, Jesse Chan (discussant), and the workshop participants at the University of Washington, York University, Saint Mary's University, Stanford University, and University of Waterloo. Matsumoto would like to thank the Marion Ingersoll Professorship at the University of Washington for financial support. Zhang acknowledges the financial support from the Social Sciences and Humanities Research Council of Canada (SSHRC) and the Ronald L. Cliff Professorship.
Disclosure rules in many countries require public companies to report fees paid to the principal audit firm and its affiliates, but not to unaffiliated component auditors (UCAs). Reported fees thus depend on component auditor affiliation. We show that UCA involvement is significant in many countries. Using Italian private companies' public filings, we find that UCA involvement is a statistically and economically significant source of bias in audit fee models. This bias is particularly relevant around principal auditor changes, when some subsidiaries temporarily retain UCAs before switching to auditors affiliated with the new principal auditor. The exclusion of fees paid to temporary UCAs explains the initial-year audit fee discount (lowballing) previously documented, which disappears once these omitted fees are accounted for. Our results suggest that lowballing reflects measurement bias rather than systematic strategic pricing. Because UCAs are common internationally and disclosure rules similarly exclude these fees, this bias extends beyond Italy.
