Exploratory analysis of the determinants of audit engagement resource allocations;

              3
An Exploratory Analysis of the Determinants of Audit Engagement Resource Allocations
Timothy B. Bell
KPMG Peat Marwick LLP
W. Robert Knechel
University of Florida
John J. Willingham, Jr.
University of Texas at Austin
Introduction
The purpose of this study was to gain a better understanding of characteristics that impact total audit work performed on domestic financial statement audits. Prior studies have investigated the determinants of audit fees (see, e.g., Elliott and Korpi (1978) and Ashton, Elliott and Willingham (1989)). A more recent study by O'Keefe, Simunic and Stein (1994) (hereafter OSS) examines how client characteristics affect both the amount and mix of labor used on financial statement audits. OSS estimated five regression equations using as the dependent variables each of four types of labor input hours and total audit fees. Our study uses the same data as used by OSS-data from the period 1986 through 1989 for 249 clients of a large international accounting firm with primary operations in the manufacturing, merchandising and high tech­nology
industries.
Our study extends the OSS and other prior studies in the following ways. OSS studied the determinants of total (domestic plus foreign) audit hours for four different personnel levels (partner, manager, senior and staff hours) and total audit fees. We limit our investigation to domestic audit hours for the following reasons: (1) statutory audit requirements may differ across jurisdictions, (2) differences in legal environ­ments
could affect the extent of audit work across jurisdictions, (3) differences in audit market conditions may exist across jurisdictions (e.g., fixed vs. variable fee markets), and (4) technology (e.g., audit processes) may differ across jurisdictions even within the same audit firm.1 We expect the existence of client foreign operations will impact the quantity of domestic audit work performed. Even though domestic auditors might not perform the actual audit work on foreign subsidiaries, consolida­tion
of these subsidiaries into a domestic parent's financial statements could lead to additional domestic audit work, especially administrative work dealing with the coor­dination
of the full audit.
OSS estimated separate models for each of the four labor inputs mentioned above. We primarily focus on total domestic audit hours aggregated across all personnel
1 For example, we know that the calculation of planning materiality for certain foreign jurisdictions differs
from the domestic calculation for the firm whose data are being studied. This could directly affect sample
sizes and the resulting extent of audit work performed.
49