The Impact of Activity-Based Costing
The Impact of Activity-Based Costing
The Impact of Activity-Based Costing
Volume Thirteen
2001
INTRODUCTION
The availability and relevance of accounting information underlies many business decisions. In recent years, traditional volume-based cost models have been
the subject of much criticism, especially in relation to the accuracy of product
costing. The popularity of activity-based costing (ABC) in the mid-1980s and the
subsequent evolution {Bromwich and Bhimani 1989) vs. revolution (Johnson and
Kaplan 1987) debate has enriched both the management accounting literature
and practice. Research to date, however, has concentrated on assessing the integrity of the ABC process (for example, Foster and Gupta 1990; Noreen 1991; Roth
and Borthick 1991; Banker and Johnston 1993; McGowan 1998; Maher and
Marais 1998); examining its application and implementation in a single case study
situation (for example, Cooper and Kaplan 1999); assessing the degree of interest
and adoption (for example, NichoUs 1992; Armitage and Nicholson 1993; Innes
and Mitchell 1995, 1997; Malmi 1999); and factors impacting the success of implementation (for example, Anderson 1995; Shields 1995; Foster and Swenson 1997;
This paper has benefited from the helpful comments of Roger Mills, Robert Kaplan, Ram Ramanan,
Eamonn Walsh, and seminar participants at Florida State University, the University of Kentucky, and the
University of Notre Dame. We gratefully acknowledge comments from two anonymous remewers and the
editor in revising this paper. Any remaining errors are the sole responsibility of the authors. The financial
support of the University of Limerick is also acknowledged.
20
Anderson and Young 1999). Despite the assertion of Cooper and ICaplan (1992) that
"the goal of ABC is to increase profits, not to obtain more accurate costs," little
attempt has been made to examine empirically the relationship between the adoption of ABC and the creation of shareholder value through increased profits.
Therefore, the prime purpose of this paper is to see whether the many documented
successfiol case study implementations of ABC are, on average, translated into
superior stock perform,ance.
Using mail surveys, we identify a sample of 47 firms listed on the London
Stock Exchange (LSE) that adopted ABC between January 1988 and February
1996. We also identify 183 firms that had not adopted ABC before the end of
February 1996. We match each adopting firm at the beginning of the year in which
they first adopted ABC with a non-adopting firm from the same industry and of
approximately the same market capitalization. Despite the limited number of
confirmed non-adopters in our sample, we were able to match 37 of the 47 ABCadopting firms within a reasonable level of accuracy. Next we computed buy-andhold returns for the ABC-adopting firms and their matched counterparts for the
three-year period beginning in the year of adoption and continuing for the subsequent two years. Our results reveal a three-year return of 61 percent for the ABCadopting firms, compsired to 34 percent for their non-adopting counterparts, the
difference being significant at the 5 percent level.
We perform several robustness checks on our data. First, we use nonparametric methods to confirm that the proportion of ABC adopters that outperform their
matched counterparts is greater than 50 percent. Second, we use two alternative
matching criteria, one based on market-to-book ratio, the other on net total
assets. Third, we use Ritter's (1991) cumulative abnormal return (CAR) method as
an alternate to the buy-and-hold method to measure abnormal returns. Fourth,
we use the FTSE-100 index as a market surrogate to compute market-adjusted
returns rather than matched-firm-adjusted returns. Fifth, we review the two-year
pre-adoption market performance of both the ABC firms and their matched counterparts in order to test for the overreaction/underreaction phenomena. Our
results show that there was no difference in the performance of our ABC firms and
their matched counterparts prior to the adoption of ABC. Finally, we report the
difference between the ABC firms and their matched counterparts under a range
of accounting-based measures of performance. In all cases, we find evidence of
significant superior performance by the ABC-adopting firms.
As with many cross-sectional studies, our results must be interpreted with
caution. Other factors, unknown to us but common to our 47 ABC-adopting firms
may be the underlying cause of the superior performance we document. We
attempt to address these concerns by examining company announcements in the
period surrounding the adoption of ABC. We fmd no difference between our ABCadopting firms and their matched counterparts, either in the year before, or the
three years following the adoption of ABC, in terms of significant corporate events
such as new capital raised, acquisitions, and divestitures. It is possible however,
that the implementation of other strategic initiatives that coincide with the adoption of ABC may be the cause of the abnormal returns we document. Thus, despite
the strong and robust evidence in this paper, it is not possible to prove definitively
that there is a causal link between ABC adoption and subsequent increases in
shareholder value.
The structure of the paper is as follows. Section two presents a brief review of
the ABC literature and a summary of the attributes and characteristics of ABC, as
interpreted in this study. It also sets out the hypothesis and the theoretical model
underpinning this study. Section three documents the research method and data
Produc I/service,
ctulomer and
market profitability
inrormation
business's
acciviiies
Investment
Decisions
ABC
Techniques
Profit and
Cashfkiw
\
Financial and
nonfinancial
peifonnancc
measures
Operating
Dee is ions
Market
Vakie
Objective
o f Finn
22
1991). The activity-based approach has also been used as a base for applications
such as benchmarking, cost reduction, and transfer pricing (Morrow and Ashworth
1994). Further, the evolution of the "balanced scorecard" concept (Kaplan and
Norton 1992), and its subsequent refinement as the organization's central management system (Kaplan and Norton 1993; Hoffecker and Goldenberg 1994),
contextualizes the application of ABC.
Even though the ABC concept was initially developed in a manufacturing
context, it can be applied equally well in the services sector, as activities are
universal to all organizations. Cooper and Kaplan (1992) and King et al. (1994)
document successful applications in the telecommunications, transport, wholesale and distribution, marketing, health, and information services sectors. Innes
and Mitchell (1997) document its successful application in the U.K.'s largest
financial institutions.
Malmi (1999) cautions that there is no common view of what makes an accounting system an ABC system. For the purposes of this study, we therefore
regard ABC as a generic term to describe in altemative paradigm to traditional
volume-based cost models with the following key attributes:
(1) a more equitable allocation of overheads by identifying the underlying
"driver" of activities;
(2) a "new technology^ capable of dealing with an increasingly complex and
opaque cost base;
(3) a process of activity identification with the ability to integrate nonaccoiinting
aspects; and
(4) a control device in the spirit of "panopticon" and built upon the knowledge
created through the management information system.
Within this framework ABC can be viewed as a cost and performance management
model capable of contributing significantly to operational improvements and strategy formulation (Kaplan and Cooper 1998). It is also consistent with Porter's
(1985) value-chain concept and its development by Shank and Govindarajan
(1993) in a strategic management accounting context.
Gosselin (1997) notes that despite the perceived advantages of ABC and the
interest shown by academics and management accountants, adoption has not
been intense (for example. Institute of Management Accountants 1993 [U.S. evidence]; Armitage and Nicholson 1993 [Canadian evidence]; Innes and Mitchell
1995 [U.K. evidence]). Moreover, Homgren (1990) and Nanni et al. (1992) provide
evidence on some firms that have stopped the implementation process. Gosselin
(1997) terms this the ABC paradox, namely, "if ABC has demonstrated benefits,
why are more firms not actually employing it?"
There are several potential answers to this paradox. First, ABC may not be
suitable for every firm. A large strand of recent research shows that successful
implementation depends on many aspects. For example, Anderson (1995) and
Malmi (1997) argue that successful implementation depends on organizational
and technical factors, with Anderson and Young (1999) citing supporting evidence
from studies such as Chenhall and Langfield-Smith (1998), Foster and Swenson
(1997), Knimwiede (1998), and Shields (1995). Similarly, Malmi (1999) argues that
adoption depends on several factors including firm size, production lype, degree of
centralization, product diversity, and the ratio of indirect to total costs. Finally,
Gosselin (1997) suggests that specific characteristics in their business strategy
and organizational structure lead certain firms to adopt and implement ABC and,
by implication, other firms not to implement ABC.
24
examines the performance of ten firms identified as ABC users in a May 30, 1988
article in the magazine, Business Week. While the ABC firms do have positive abnormal returns on the publication date, so do ten size- and industry-matched control
firms. As the difference in the returns to the ABC firms is not significantly different
(at conventional levels) from their matched counterparts, Gordon and Silvester
(2999) conclude that the announcement of ABC use does not affect firm values.
In summary. Cooper and Kaplan (1992) argue that the ultimate aim of ABC is
to increase profits. The associated literature suggests that provided the techniques are implemented effectively, with the support of ail, then increases in sales
and/or reductions in expenses should follow. Given these profit implications,
additional value should accrue to those firms that adopt the techniques. The
general hypothesis examined in this study is, therefore:
Ho: There is no difference in the performance of a portfolio of firms
adopting ABC when compared with a set of matched firms that do
not employ ABC.
26
TABLE 1
Analysis of Survey Results
This Study
Analysis of
Respondents
ABC users
Non-ABC users
Total-usable responses
Too busy
Policy not to respond
Unlikely to respond
Unusable
Other
No response
Total Sample Size
47*
187
234
Innes and
MitcheU (1995)
5.5
49*
202
21.9
27.4
292
203
54
26
NA
44
34.2
23.8
853
251
6.3
3.1
5.2
23
51
NA
31
83
561
100
1,000
Innes et al.
(2000)
4.9
31*
146
20.2
25.1
2.3
177
%
4.0
18.8
22.8
56.1
NA
NA
NA
32
139
427
17.9
55.1
100
775
100
5.1
3.1
8.3
4.1
* 20.1 percent. 19.5 percent, and 17.5 percent of total usable responses, respectively.
FIGURE 2
When was ABC Introduced?
1988
1989
1990
1991
1992 1993
Year of Introduction
1994
1995
1996
implementation rate was to be expected in the early 1990s. The results of our
survey indicate that this was achieved primarily in the years 1992 to 1994. It is
also important to note that none of the respondents to our survey indicated that
they had introduced and subsequently abandoned the ABC approach.
The Matching Process
Our main results are based on a comparison between the ABC-adopting firms
and a matched sample of similar, but non-ABC firms. This is similar to the
approach followed by Gordon and Silvester (1999). From our survey, we identified
47 firms that had adopted ABC, together with the year of adoption for each firm.
We also identified 187 firms that had not adopted ABC. As Easton and Jarrell
27
(1998) note, the use of control firms is critical to control for common factors that
effect similar types of firms in the performance measurement period and are not
related to the adoption of ABC.
The precise time of the introduction of ABC is not an easily discernible event,
unlike, for example, an acquisition decision or an earnings announcement. Consequently, there is not a specific, easily identified date that can be associated with
the introduction of an ABC system. Both our survey and the Innes and Mitchell
(1995) survey found that firms consider the adoption of ABC over a period of time,
often with the assistance of external consultants. Even sifter deciding to adopt an
ABC system, it can take many months of planning and development before the
system is finally implemented. Because of the difficulty in establishing an exact
event date, we follow Haka et al. (1985) and choose January of the year in which
the firm indicated they first used ABC as the adoption date. All performance
comparisons are done relative to this January date.
Because firms seldom make a specific public announcement of a change to an
ABC system, an instsmtaneous response to the introduction would not be expected. Hence we do not use a standard event study approach. Rather, we use a
performance assessment window of three years to capture the effect, if any, of the
introduction on firm performance. This recognizes the time it takes to fully implement such a system and the potential "lag" effect discussed in other studies such
as the TQM study of Easton and Jarrell (1998) who examine five-year post-event
performance, and Haka et al. (1985) who use a 48-month period to examine the
impact of a change in capital budgeting methods. It is also consistent with the
growing literature on long-term trends in stock market prices (for example, Ritter
1991; Spiess and Affleck-Graves 1995). While most empirical evidence of this
latter phenomenon is from the U.S., Levis (1993) has shown similar evidence for
the U.K. stock market, the database chosen for this study.
We matched firms on December 31 in the year prior to the adoption of ABC.
Three different matched samples were established, based on market capitalization, market-to-book ratio, and net-total-assets. Market capitalization was used to
control for firm size, a variable that Banz (1981) and others have shown is related
to stock returns. Matches by market-to-book were used in response to the findings
of Fetma and French (1992) and net-total-assets was used as a size measure of
balsince sheet value. Throughout the exercise, no firm was used more than once as
a matched firm within a three-year time frame.
The specific matching procedure was as follows.
(1) ABC firm i was matched with all non-ABC firms in the same level 4
(Datastream) industry classification and within 25 percent of its market
capitalization. If more than one match was found, then the non-ABC firm
with the market capitalization closest to firm i was chosen as the match.
(2) If no match was found in the level 4 industry classification within 25
percent, the process in (1) was repeated using the level 3 industry classification. If no match was again found, then the process was repeated at level 2.
(3) If no match within 25 percent of the market capitalization of firm i could be
found in the 187 non-ABC firms, then the entire process was repeated
using 50 percent of the market capitalization of firm i.
This procedure gave first priority to the market capitalization difference and
secondary status to the industry-level match. Similar procedures were used to
identify the market-to-book and net-total-assets matched samples. The final market capitalization-matched sample contains 37 firms, while the market-to-book
and net-total-asset samples contain 38 and 33 firms, respectively.
28
Table 2 presents an industry profile of the three sets of matched ABC sample
firms, based upon Datastream level-3 industry classification codes. It reveals that
all the major industries in the U.K. economy are represented and confirms that
the process has also been applied outside the manufacturing sector. The results of
the three matching procedures are summarized in Table 3.
It is worth noting from Table 3 that 76% of the firms in the case of market
capitalization, 84% in the case of market-to-book and 82% in the case of net-totalassets were matched within the level 3-industry classification,^
Descriptive statistics indicating the accuracy of our matching procedure are
presented in Table 4. They show that the market capitalization data set achieved
the best match as there were no statistically significant differences (at the .10
level) between the ABC and non-ABC firms in terms of mean market capitalization, market-to-book ratio, or net-total-assets.
We also used the less onerous restriction of matching within 75 percent of market capitalization
and going to industry level 1. This increased the market capitalization-matched sample size by 6
firms (0 and 1 in the case of the market-to-book and net-total-assets samples, respectively). All
results are qualitatively the same between the larger sample and the results we report using the
smaller sample size.
TABLE 2
Profile of ABC Matched Sample Firms
Description
(Level 3)
General Industries
Utilities
Financials
Consumer Goods
Services
Total
Market
Capitalization
No. Firms
14
37.84
18.92
16.22
18.92
8.11
100.00
7
6
7
3
37
Market-to-Book
%
No. Firms
11
9
8
8
2
38
28.95
23.68
21.05
21.05
5.26
100.00
10
7
5
8
3
33
30.30
21.21
15.15
24.24
9.09
100.00
TABLE 3
Summary of Matched Sample Process
Industry
Level
Matched
within %
No. Firms
Matched
% Firms
Matched
Cumulative %
of Sample
27
49
24
27
76
100
100
47
37
13
47
84
97
100
100
33
49
18
100
33
82
100
29
Non-ABC
Firms (Mean)
Match Criteria
(Mean)
Panel A: Market Capitalization Sample (37 firms)
Market Capitalization
1,504.35
1.421.68
Market- to- Book
2.61
3.41
Net-Total-Assets
861,740
955,447
Difference
(t-statistic)
1.1815
-0.4817
0.6642
p-value
.1226
.3167
.2557
1.4401
2.4814
2.5584
.0791*
.0089**
.0074**
1.5757
0.6442
0.2872
.0625*
.2620
.3879
The market-to-book data set is not as well matched with respect to all three
criteria, especially market capitalization and net-total-assets. In this case, the
ABC firms tend to be larger, which may bias the results shown later in Section
four. The net-total-as sets data, set of 33 firms is reasonably matched against the
market capitalization and market-to-book criteria, but the ABC firms have larger
net-total assets on average (at .10 level).
The overall result of the matching process is that high-queility matches were
achieved for both the market capitalization and net-total-assets data sets. Caution
is suggested in regard to the market-to-book data sets due to the less than
satisfactory secondary matching results.
Performance Analysis
As Anderson and Young (1999) note, success in ABC implementation is multidimensional and one needs to be careful in addressing the question of what
constitutes success. In this study, we use a narrow definition of "success" that is
consistent with fmance theory and the goal of shareholder wealth maximization,
namely an increase in shareholders' wealth as measured by stock returns {Haka et
al. 1985). As Easton and Jarrell (1998) note, stock returns provide "a comparatively 'clean' overall performance measure." It is important to stress that other
measures of performance exist and our results only pertain to the narrowly defined stock market return definition of performance.
For both the ABC firms and their matched counterparts, we compute returns
from January 1 in the year of the adoption of ABC. This discrete date was chosen,
in the interests of consistency, as the survey infonnation did not identify the
month of adoption of ABC. We use both three-year average holding period returns
(HPRs) and average cumulative abnormal returns (CARs) to test for difference
between each matched set of firms. The HPR measure is used to overcome the
findings of Conrad and Kaul (1993) and Barber and Lyon (1997), that CAR-based
metrics can lead to biased test statistics in studies of long-term stock performance. The CAR measure (Ritter 1991) is included as a robustness check.
We compute HPRs over the three-year period commencing in the first month of
the year of adoption of ABC as follows:
30
We use the simple t-test to determine whether the monthly AR is different from
0, that is, whether ABC firms on average outperform the non-ABC benchmark
firms in month t:
-r
A
where n^ is the number of observations in month t, and sd^ is the cross-sectional
standard deviation of the abnormal returns for month t.
The cumulative abnormal return (CAR) for firm i is the summation of the
average abnormal return from event month q to event month s:
t=q
The CAR is cross-sectionally averaged and the t-statistic for the cumulative
average abnormal return from month 1 to month t is computed as (Ritter 1991):
1
where: C A R i t = ^
n^ = the number of firms trading in each month;
t (t.var + 2 (t- 1) cov)^/^.
t = the event month;
var = the average (over 36 months) of the monthly cross-sectional
variances; and
cov = the first-order au toco variance of the AR^ series.
31
RESEARCH FINDINGS
Table 5 contrasts the performance of the ABC firms with that of their non-ABC
matched counterparts. Results are presented for the HPR and CAR methods, as
well as a nonparametric test of the proportion of ABC firms that outperformed
their matched counterparts. Using the HPR method, the ABC firms clearly outperform their matched counterparts in the three years commencing with the year of
the introduction of ABC. Over this period, the ABC firms average approximately 61
percent return (58.66 percent to 63.24 percent depending on the sample used) in
contrast to the approximately 34 percent (30.32 percent to 36.02 percent) earned
by their matched counterparts. This 27 percent difference, on average, is statistically significant in all cases (at the 0.05 level).
The CAR method confirms the above results, with the ABC firms outperforming
their non-ABC counterparts, by approximately 25 percent, on average, over the
three-year period. The nonparametric test of proportions confirms that in most
samples, the percentage of ABC firms that outperform their matched counterparts
is above 60 percent and is statistically larger than 50 percent at the 0.10 level in
four of the six cases. Finally, wealth relatives that range from 1.16 to 1.25 indicate
the magnitude of the difference in performance over the three-year period.
Figures 3, 4, and 5 highlight the magnitude of the positive difference in favor
of the ABC firms over the 36 months. The largest positive abnormal returns are
observed in the last 12 months of the study period in each case. A possible reason
for this is the complexity of an ABC system and the lengthy time frame required
for full implementation. This may lead to a delay between implementation and the
realization of the benefits in the stock price.
ROBUSTNESS TESTS
We provide four additional robustness checks to show that the superior returns earned by the ABC firms are not a result of our matched sample method.
First, we compare the performance of the ABC firms with that of the FTSE-100
index, a broad-based market index. Second, we track the performance of the ABC
firms and their matched counterparts in the two-year period prior to the adoption
of ABC. Third, we provide a comparison between the ABC and the non-ABC firms
for a number of accounting-based measures of performance. Finally, we compare
the occurrence of other strategic initiatives such as new capital raised, acquisitions, and divestitures for the ABC and the non-ABC firms both before and after
the adoption of ABC.
Table 6 compares the post-adoption performance of the ABC firms with the
FTSE-100 index. In this case, the HPR for the FTSE-index was computed over the
same 36 months as that used to compute each ABC firm's HPR. The results are
very similar to those found using the matched-firm approach. In summary, the
ABC firms outperform the FTSE index by approximately 20 percent over the three
years beginning on January 1 of the year that ABC is first adopted. This difference
in performance is statistically significant in all cases, at the 0.05 level. In addition,
approximately 60 percent of the ABC firms outperform the FTSE-index over the
three-year period and the wealth relative is in the order of 1.14.
Failure to capture risk differences could explain the superior performeince of our
ABC firms. To address this possibility, we contrast the performance of these firms
over a different time period. The period chosen is the two-year period prior to the
first year that ABC was adopted by the ABC firms. The results in Table 7 show
little difference in performance between the two groups during this two-year period.
The difference in abnormal returns is between -7 percent and +5 percent, depending on the sample used, and is never statistically significant at conventional levels.
32
FIGURE 3
Monthly HPRs of ABC Firms vs. Non-ABC Firms
Market Capitalization Sample
0.7(X)0 .
0.6000 0.5000 0.4000
0.3000 0.2000 0.1000
0.0000
ABC Fitms
FIGURE 4
Monthly HPRs of ABC Firms vs. Non-ABC Firms
Market-to-Book Sample
0.7000
0.6000
0.5000
0.4000
0.3000
0.2000
0.1000
0.0000
-1
ABC Firms
FIGURE 5
Monthly HPRs of ABC Firms vs. Non-ABC Firms
Net-Total-Assets Sample
0.7000 n
ABC Firms
0.6000
0.5000
0.4000
v'
0.3000
0.2000
Non-ABC Firms
0.1000
0.0000 .
h-
<n
<-
fo
f^
^ ^ i i i
C M C s l f s i f s J C J C O M C O
33
CN
O O
in
(N
CN OJ
CO
.3
1a3
*
00 CN
Ol I1
in in
Non ABC
i
1
Finn
b.
b)
S!
CN
CN
r^
O
VO
o
t^
10
vO
in
00
CJ*
00
00
N
V
1
I ABC
CO in
00 ^^
CO in
in CN
in
vO in vO
in
00
\O o
CN
a
o
o
00
CN
in
in
B
0
CQ
as
CO
in
oq
oj
* *
00
o
00
in
O) CN
o
oi O)
V
9
-a
to
J CJ
en o
vo oq
\D
\D
CO
6 tf
O ^
(S O
00 oi
Ol
O 00
CO 6 -I
^ vD ^
CN vp Ol
oi ocj CO
\Q if^ \^
a
o
Xu
34
35
-^ a
in in in
in
o CN
o o o
o o o
o o
O
o
O
in
00
CO
q
d 6
^5
yD
m
in
in
>1
-'
f^
o
in CO
o
\
1
vO
vo
t i-H
11
CO
(y\
\D
1;
CO
^^
If
Tj-
o o o
I-H
I-H
1 1
I
11
t1
CN
fO in
O tn
in in o
00 CN in
CN CO
O o Ol
in in in
o CN
o o o
CO 00
t ^
oJi '
C
CO c
in
o o o
I-H
.1
o o o
s
v
oo c
in
cs
in
c^
1 1 1
in
1 1
o
o
o o o
CO t1
CO CN 1
CO
vo
fO r-t
1 1
in
o N
vo CO
-H
CO
^^ o o
O
o o
CN o
CO O 00
o O cs
o o
in
vo m
o o o
o o o
in
cs
c
s o CO
00
c^ in
o o o
ON
I1
1-H
cs <1
-H
1 1 I
in
in
CO
CN
1 1 1
-H
CN 1
-H
o
o
o o o
11
(O CO
CO
o . in
CO CN CN
1 1
\O
vD sP -^
-H ^ <N
Ol in CO
CN CO ^
CO CO CN
CO
o
00 in
O
0^ CO
CO
vo
00
vO t^
O Q\ ON
CO
'-I
^
in
o
o
vO
o
I-H
o o o
-3
CO o
CO in
CN
o in tn
N
o
vo cs
o o o
o
o
CO o
vo CO
O
o
o
in
in
11 '
1 1
I1
71
<1
CO 0 CN
00 o
I1
CN
vo
CO
oj
^
00
cji t^ q
-J CO ^'
5
CN 00
oi CN CN
<.^
sp s ^
O ^ vO
CN 't
CO CO ^
O vo CN
in CO
E
a
V
B
GQ
vo CO O
-H p CO
CN ^ in
O
M
o o g
q cs \D
od CO t^
ON in 00
vo o vo
CO q a;
-^ iri 'J-'
M cs ^
00 h- CO
CO CO ^
in N N
l!
CO
in
in
CO
Ol
0C3
o^
CN
CO
lO
lo
O)
vO
b cs ^
o
OJ
O
00
ro
0.
v P
O\ O
o 00
o t-
CN 00
in fON ON ON
CN O) CN
o ^ o cV
CN
f- in CN
-K -H ^
in CO ^
CO CO CO
Q.
q
d
o
t
1
u
o c
o
>N a
c
o
0 0
M
V
o
o
33 m
"d
a
ou
rt
*->
-
S
c
CO
tn
3
ti
0 OJ
Oi
m V
o (/}
UcQ
<
*.!3 oV
<
a
cuu U
*-l
.l-t
0) V
2
O
tal
o
CO
CO
Tf
CO
tal
o
O
O;
d
CO
< ^
CO CO 00
o
.<..>
I*
2
36
the Global Access online database provided by Primark. Table 9 tabulates our
results for these three important strategic initiatives. It documents both the number of firms engaged in each activity and the dollar value of the activity expressed
as a percentage of the firms' market capitalization. There is no evidence of significant differences between the number of our ABC firms and the number of nonABC firms that are engaged in these three activities, either before or after the
adoption of ABC. {Note that the higher number of firms in the post-ABC period is
due to the three-year period as compared to the one-year pre-ABC period that we
examine.) Similarly when we examine activity as a percentage of each firm's
market capitalization, we fmd no differences between the ABC firms and their
matched counterparts, either before or after the introduction of ABC.
We also read the Chairman eind Chief Executive's annual report for the majority of our ABC and non-ABC firms. There are several consistent themes in these
reports, including regional development, some element of re-organization and staif
changes, changes in directors, encouraging trade performance, increased emphasis on marketing and new product development, emphasis on maximizing shareholder value through long-sighted programs of organic growth coupled with carefully chosen acquisitions, etc. We were, however, unable to discern any marked or
obvious differences between the ABC and non-ABC firms.
While we have not examined all possible extraneous events that may be the
underl5dng cause of the superior performance we document for our ABC sample,
the above analysis indicates very little difference between our ABC firms and their
matched counterparts in terms of several important strategic actions. Nevertheless, we cannot rule out the possibility that some unidentified correlated variable
is the cause of our results.
Finally, this study asked firms to rate their experience of introducing the ABC
techniques. We found that a substantial majority at 77 percent (n = 36) considered
the experience to have been successful. This is consistent with Innes and MitcheU
(1995) at 69 percent (n = 34) and the 3.9 Likert scale rating (with 5 being very
successful) reported by Innes et al. (2000). It is also consistent with a number of
studies (for example. Shields 1995; Swenson 1995; Shields and McEwen 1996;
Krumwiede 1998; Anderson and Young 1999) that have used key organizational
and social variables to measure the success of ABC implementation. Despite the
considerable variation in the degree of success reported by these studies, on average, firms experienced moderate levels of success emd perceived a financial benefit
fi-om the adoption of ABC. Worth noting is the possibility that respondents fi-om
ABC-adopting firms may be exhibiting biased discrimination in their judgment.
37
a^
o
o
U\ 00
o
o
in CS
in o
CS 00
o
o
in <n
in in
CS 00
O 00
in rj;
a
(N \D
CO vb
dd
00
ci
O CS
CN CN
Tf in
in CO
d^
*'
o
o
in in
O O
oo
>> 00
t
fO t-
00 t^
'- CO
OO VO
tn
\6
oi
o
00
CN
CS
00 O\
V V
a9
O O
O O
in
aid
00 00
CS
CS
J^
CS CS
I1
4)
ta
o> oa
u
>
V ^ CO
Adoption
2 Adoption
99
<,
ior /
^4
*3
ars >ost-,
\-/
o\
t
>'
\^
9m
in
CO
Adoption
- Adoption
Adoption
:; Adoption
:ars
Ml
ars
qui
oo
in 00
CO \D
al
ior /
0
M
d
Flrir
VD CO
CO
o
\D
C7;
ro
Iptl
q vD
01 in
t"
CO
j; ^
O
(O
38
Our evidence of superior performance for ABC-adopting firms provides tangential support for two other studies related to the impact and benefits of ABC.
First, Malmi (1999) explores four possible reasons for adopting ABC: efficient
choice, forced selection, fads, and fashion. Superior subsequent performance
suggests that the adopting firms made a rational value-enhancing choice when
adopting ABC. Consequently, our results provide support for Malmi's (1999) evidence in favor of the efficient choice hypothesis, although we are unable to determine if this is limited to the early adopters ("take-off stage) as he suggests.
Second, Shields and McEwen (1996) report that 75 percent of responding companies report a financial benefit from implementing ABC. Our evidence of superior
stock returns following adoption suggests that the companies' perception of financial benefit is justified.
In contrast to the support for the above studies, our results do not clarify
Gosselin's (1997) ABC paradox, but rather accentuate it. If ABC-adopting firms
have better stock performance in addition to the other benefits cited in the literature, then why have more firms not implemented the approach? Moreover, why do
many firms that adopt ABC stop using it? While none of our survey firms indicated
that they had introduced and subsequently abandoned ABC, Ness and Cucuzza
(1995) suggest that only 10 percent of firms that adopt ABC continue to use it.
This is clearly inconsistent with our value-enhancing evidence and we are unable
to shed further light on Gosselin's (1997) paradox.
Our results are also somewhat inconsistent with those in Gordon and Silvester
(1999). Using a sample often U.S. firms, they find no evidence of an increase in
firm value associated with the announcement that the firms are ABC users. Our
tests, however, have several advantages over the Gordon and Silvester (1999)
approach. First, we have a much larger sample. Second, our sample does not have
the extreme event-clustering present in the Gordon and Silvester (1999) study.
Both of these should lead to more powerful test statistics. Third, we examine longterm performance and, hence, we do not require an exact announcement date or
that the market immediately and fully understands the implications of the ABC
adoption (Gordon and Silvester 1999, 236). Combining these two market-based
studies, we can conclude that there is no evidence that the adoption or use of ABC
leads to a decrease in firm value as suggested in a worst-case scenario by Bromwich
and Bhimani (1989) and Dopuch (1993).
There are many important limitations and caveats to our study. First, there
are several issues related to the sample we use. Our sample is Eimsdl (33 to 38
firms), which limits the power of the statistical tests. It is, however comparable to
simileir studies (for example, Haka et al. [1985] use 30 firms, while Easton and
Jarrell [1998] have a sample of 108 TQM firms). The small sample size also makes
our results sensitive to the selection of our sample firms and their matched
counterparts. In addition, we use a survey method that allows firms to self-report
the adoption of ABC and this may bias our sample to certain types of firms. We
attempted to limit the impact of these problems through the use of a controlled
matched-pairs design and several robustness checks. However, sample limitations and bias remain a valid concern.
Second, the sample firms have adopted ABC in a wide variety of ways (Friedman and Lyne 1995) and we are not able to classify the degree of implementation
by our ABC firms. In particular, we have no indication of each firm's commitment
to ABC, or the extent to which it is implemented. Shields and McEwen (1996) and
McGowan and Klammer (1997) both indicate that factors such as top management support, compensation, and tredning are important in determining the success of ABC implementation.
40
APPENDIX
Activity-Based Costing (ABC) Techniques Questionnaire
(Confidential
The introduction of Activity-Based Costing (ABC) techniques has met with
mixed success. Some firms have had very successful implementations whilst
others have adopted and later abandoned the approach. This questionnaire is the
basis of a research study aimed at establishing if the introduction of ABC techniques makes a significant difference to firm performance.
ABC is defined by Raffish and Tumey (1991) as "a methodology that measures
the cost and performance of activities, resources and cost objects." The cumulative cost of each activity is then traced to products or services that make that
activity necessary. By so doing ABC recognises the causal relationship of cost
drivers to activities.
Please return the completed questionnaire by using the attached prepaid envelope or fax before
Friday, March 22 to: RESEARCH CENTRE, HENLEY MANAGEMENT COLLEGE. Fax: 01491-571454
.Respondent:
Firm Name:
(optional)
Position in Organisation:
(Please print the above)
1. Has your firm ever used ABC techniques?
Yes
No
2. When were the ABC techniques introduced and what proportion of your firm's
Sales Value, Total Assets and Total Costs were affected by their introduction?
Show multiple introduction dates and data separately, if appropriate.
WHEN
Month
Year
PROPORTION AFFECTED
% of Sales
% of Total
% Total
Value (approx.)
Assets (approx.)
Costs (approx.)
3. Please complete if your firm has used ABC techniques and abandoned them.
WHEN
Month
Year
PROPORTION AFFECTED
% of Sales
Value (approx.)
% of Total
Assets (approx.)
% Total
Costs (approx.)
Implementation
Design
Consultants
In-house accountants
Production personnel
Systems personnel
Marketing personnel
Any other (please specify)
5. For which of the following purposes were the ABC techniques introduced and
are they still used for that purpose? (Tick as appropriate)
Basis of Decision
to Introduce
Purpose
Stock Valuation
Product Pricing
Product Output Decisions
Cost Management and Reduction
Cost Budgeting
New Product Design
Customer Profitability Analysis
Performance Measurement
Cost Modelling
Any other purpose (please specify)
Yes
No
Current Use
Yes
No
No
42
7. How would you rate the overall success of the introduction of the ABC techniques in your firm? (Tick as appropriate)
Degree of Success
Very successful
Fairly successful
Neutral
Fairly unsuccessful
Very unsuccessful
8. Did you experience any unexpected outcomes from the inti^oduction of the
ABC techniques?
Yes
No
10. Into which of the following categories does your firm fall and why? (Ignore if
answered Yes to Question 1)
Category
We have considered using ABC
techniques and rejected them.
Yes
No
Why (optional)
43
REFERENCES
Anderson, S. W. 1995. A framework for assessing cost management system changes: The
case of activity based costing implementation at General Motors, 1986-1993. Joumal
of Management Accounting Research 7: 1-51.
, and S. M. Young. 1999. The impact of contextual and process factors on the evaluation of activity based costing systems. Accounting, Organizations and Society 24; 525559.
Armitage, H. M., and R. Nicholson. 1993. Activity based costing: A survey of Canadian
practice. Issue Paper No 3. Mississauga, Ontario, Canada: Society of Management Accountants of Canada.
Asquith, P., and D. W. Mullins. 1986. Equity issues and offering dilution. Joumal of Financial Economics 15: 61-89.
Ball, R., and P. Brown. 1968. An empirical evaluation of accounting income numbers.
Joumal of Accounting Research 6: 159-178.
Banker, R. D., and H. H. Johnston. 1993. An empirical study of cost drivers in the U.S.
airline industry. The Accounting Review 68: 576-601.
Banz, R. 1981. The relationship between return and market values of common stock. Journal of Financial Economics 9: 3-18.
Barber, B. M., and J. D. Lyon. 1996. Detecting abnormal operating performance: The empirical power and specification of test-statistics. Joumal of Financial Economics 41:
359-399.
, and
. 1997. Detecting long-run abnormal stock returns: The empirical power
and specification of test-statistics. Joumal of Financial Economics 43: 341-372.
Bentley, T. 1993. Activity-based computing. Management Accounting (UK) (December): 32.
Brim.son, J., and R. Fraser. 1991. The key features of ABB. Management Accounting (UK)
(January): 42-43.
Bromwich, M., and A. Bhimani. 1989. Management Accounting: Evolution not Revolution.
London, U.K.: Chartered Institute of Management Accountants.
, and
. 1994. Management Accounting: Pathways to Progress. London, U.K.:
Chartered Institute of Management Accountants.
Brown, S. J., and J. B. Warner. 1980. Measuring security price performance. Joumal of
Financial Economics 8: 205-258.
, and
. 1985. Using daily stock returns: The case of event studies. Joumal of
Financial Economics 14: 3-31.
Campi, J. P. 1992. ABM: It's not as easy as ABC. Joumal of Cost Management 6: 5-12.
Chenhall, R, H., and K. Langfield-Smith. 1998. Adoption and benefits of management accounting practices: An Australian study. Management Accounting Research 9: 1-19.
Clarke, P. 1994. Activity based cost management. Accountancy Ireland (October): 16-17.
Cobb, 1., J. Innes, and F. Mitchell. 1992. Activity Based Costing: Problems in Practice. London, U.K.: Chartered Institute of Management Accountants.
Conrad, J., and G. Kaul. 1993. Long-term market overreaction or biases in computed returns? Joumal of Finance 48: 39-63.
Cooper, R., and L. A. Weiss. 1985. Schrader Bellows. Harvard Business School Case Series,
186-050. Ceimbridge, MA: Harvard Business School Press.
, and P. B. Tumey. 1988. Tektronix: Portable Instruments Division. Harvard Business School Case Series, 188-142. Cambridge, MA: Harvard Business School Press.
, and R. Kaplan. 1992. From ABC to ABM. Management Accounting (US) 74: 54-57.
-, and
1999. The Design of Cost Management Systems: Text and Cases. Second
edition. Englewood Cliffs, NJ: Prentice Hall.
Easton, G. S., and S. L. Jarrell. 1998. The effects of total quality management on corporate
performance: An empirical investigation. Joumal of Business 71: 253-307.
Dopuch, N. 1993. A perspective on cost drivers. The Accounting Review 68 (3): 615-620.
Fama, E. F., L. Fisher, M. Jensen, and R. Roll. 1969. The adjustment of stock prices to new
information. International Economic Review 10: 1-21.
, and K. R. French. 1992. The cross-section of expected stock returns. Joumal of
Finance 47: 427-465.
44
Joumal of Management Accounting Research, 2001
Foster, G., and M. Gupta. 1990. Manufacturing overhead cost driver analysis. Joumal of
Accx)unting and Economics 12: 309-337.
, and D. W. Swenson. 1997. Measuring the success of activity-based cost management and its determinants. Joumal of Management Accounting Research9: 109-142.
Friedman, A., and S. Lyne. 1995. The organisational consequences of implementing activity-based techniques. The Newsletter of the CIMA Research Foundation 4 (January).
Gordon, L. A., and K. J. Silvester. 1999. Stock market reactions to activity-based costing
adoptions. Joumal of Accounting and Public Policy 18: 229-251.
Gosselin, M. 1997. The effect of strategy and organizational structure on the adoption and
implementation of activity-based costing. Accounting, Organizations and Society 22: 105122.
Haka, S. F., L. A. Gordon, and G. E. Pinches. 1985. Sophisticated capital budgeting selection techniques and firm performance. The Accounting Reidew 60: 651-669.
Hoffecker, J., and C. Goldenberg. 1994. Using the balanced scorecard to develop company
wide performance measures. Joumal of Cost Management 8; 5-17.
Homgren, C. 1990. Contribution margin analysis: No longer relevant. Strategic cost management: The new paradigm. Joumal of Management Accounting Research 2: 1-32.
Ikenberry, D., J. Lakonishok, and T. Vermaelen. 1995. Market underreaction to open market share repurchases. Joumal of Financial Economics 39: 181-208.
Innes, J., and F. Mitchell. 1990a. ABC: Activity-based costing research. Managem.ent Accounting (UK) May: 28-29.
, and
. 1990b. ABOAReiAew with Case Studies. London, U.K.: CIMA Publication.
, and
. 1995. A survey of activity based costing in the U.K.'s largest companies.
Management Accounting Research 6: 137-153.
, and
. 1997. The application of activity-based costing in the United Kingdom's
largest financial institutions. The Service Industries Joumal 17: 190-203.
-, and D. Sinclair. 2000. Activity-based costing in the U. K.'s largest companies: A comparison of 1994 and 1999 survey results. Management Accounting Research
11: 349-362.
Institute of Management Accountants (IMA). 1993. Cost Management Update. Montvale,
NJ: IMA.
Johnson, H. T., and R. S. Kaplan. 1987. Relevance Lost: The Rise and Fall of Management
Accounting. Boston, MA: Harvard Business School Press.
Kaplan, R. S. 1987. John Deere components works. Harvard Business School Case Series,
187-107/108. Cambridge, MA: Harvard Business School Press.
, and D. P. Norton. 1992. The balanced scorecard: Measures that drive performance.
Harvard Business Review 70: 71-79.
, and '
. 1993. Putting the balanced scorecard to work. Harvard Business Review 71: 134-147.
, and R. Cooper. 1998. Cost & Effect: Using Integrated Cost Systems to Drive Profitability and Performance. Boston, MA: Harvard Business School Press.
King, M., I. Lapsley, F. Mitchell, and J. Moyes. 1994. Activity-Based Costing in Hospitals.
London, U.K.: Chartered Institute of Management Accountants.
Krumwiede, K. R. 1998. The implementation stages of activity based costing and the impact
of contextual and organizational factors. Joumal of Management Accounting Research
10: 239-277.
Levis, M. 1993. The long-run performance of initial public offerings: The UK experience
1980-1988. Financial Management 22: 28-41.
Maher, M., and M. L. Meirais. 1998. Afieldstudy on the limitations of activity-based costing
when resources are provided on a joint and indivisible basis. Joumal of Accounting
Research 36: 129-141.
Malmi, T. 1997. Towards explaining activity-based costing failure: Accounting and control
in a decentralized organization. Management Accounting Research 8: 459-480.
. 1999. Activity-based costing diffusion across organizations: An exploratory empirical analysis of Finnish firms. Accounting, Organizations and Society 24: 649-672.
McGowan, A. S., and T. P. Klammer. 1997. Satisfaction with activity-based cost management implementation. Joumal of Management Accounting Research 9: 217-237.