9 / 372
臺大管理論叢
第
26
卷第
2
期
9
3. Research Design
3.1 Measure of Auditor Industry Expertise
Prior research suggests that external audit firms with large market shares are more
likely to be industry experts (De Angelo, 1981) because they are able to increase their
knowledge through significant investments in industry-specific training and from experience
gained through providing services to a large number of clients (Mayhew and Wilkins, 2003).
This study uses the method developed by Krishnan (2003) to measure auditor industry
expertise: one proxy is IMS (Industry Market Shares), and the other one is IPS (Industry
Portfolio Shares). This study sets continuous variables and dummy variables based on the
following calculations:
where
REV
is sales revenue, and the numerator is the sum of sales of all
J
ik
clients of
audit firm
i
in industry
k
. The denominator is the sales of
J
ik
clients in industry
k
summed
over all
I
k
audit firms in the sample with clients (
J
ik
) in industry
k
. To estimate industry
market share for each auditor, this study requires a minimum of five audit firms for each
CSRC
7
(China Securities Regulatory Commission) industry code and calendar year. We also
set a dummy variable
IMS_D
, which equals 1 when
IMS
ik
≥ 10% and 0 if not.
IMS_D
= 1
means audit firm
i
is an expert in industry
k
.
Following Krishnan (2003), we estimate auditor portfolio share as follows:
where
REV
is sales revenue and the numerator is the sum of the sales of all
J
ik
clients of
audit firm
i
in industry
k
. This study uses CSRC codes to identify industry categories. The
denominator is sales of all clients of audit firm
i
summed over all
k
industries. The dummy
variable
IPS_D
equals 1 when
IPS
ik
is the highest in the portfolio of audit firm
i
and equals 0
if not.
7 The industry code is one-digit except manufacturing industry which is two-digit.