Description
Previous studies have provided contradictory evidence with respect to the eect of rigid budgetary controls on slack
and other dysfunctional behaviors. One motivation for the current study was to test whether spillover eects exist
between two alleged dysfunctional consequences of a rigid budgetary control style: budget slack creation and man-
agerial short-term orientation. The data support this contention: reducing one form of dysfunctional behavior (slack
creation) through rigid controls seems to spill over into another form (stronger management focus on business matters
that aect short-term results). However, the budgetary control styles that organizations implement, as well as the
behaviors that they encourage, may be aected by two important antecedents: business unit past performance and
competitive strategy.
The relationship between two consequences of budgetary
controls: budgetary slack creation and managerial
short-term orientation
$
Wim A. Van der Stede*
University of Southern California, Leventhal School of Accounting, Los Angeles, CA 90089-0441, USA
Abstract
Previous studies have provided contradictory evidence with respect to the e?ect of rigid budgetary controls on slack
and other dysfunctional behaviors. One motivation for the current study was to test whether spillover e?ects exist
between two alleged dysfunctional consequences of a rigid budgetary control style: budget slack creation and man-
agerial short-term orientation. The data support this contention: reducing one form of dysfunctional behavior (slack
creation) through rigid controls seems to spill over into another form (stronger management focus on business matters
that a?ect short-term results). However, the budgetary control styles that organizations implement, as well as the
behaviors that they encourage, may be a?ected by two important antecedents: business unit past performance and
competitive strategy. The results indicate that business units that either pursue a di?erentiation strategy or have been
more pro®table are subject to less rigid budgetary controls, which augment the propensity to build slack as well as the
tendency for managers to think long-term. These relationships are tested in a structural equation model on survey data
obtained from 153 business unit general managers. # 2000 Elsevier Science Ltd. All rights reserved.
Since Hopwood's (1972) seminal paper, the
budgeting literature has shown great interest in
understanding possible e?ects of budgetary control
styles. It is generally maintained that the incidence
of so-called dysfunctional behavior is a?ected by
the rigidity of budgetary controls. A rigid budget-
ary control style is one in which employees, mostly
at management organization levels, are evaluated
primarily on whether or not they achieved their
budget. When evaluated in this way, managers are
held fully accountable for their performance as
measured by the budget. This implies that salary,
resources, and career prospects become highly, if
not fully, dependent on the managers' ability to
meet the budget. Managers who miss the targets
face the prospect of interventions by upper man-
agement, the loss of organizational resources, the
loss of annual bonuses, and ultimately the loss of
their job (Merchant & Manzoni, 1989). Under
these circumstances, managers may look for ways
to protect themselves from the downside risk of
missing budget targets and the stigma normally
attached to underachievers (Lukka, 1988; Onsi,
1973; Schi? & Lewin, 1970). Possible ways of
protection can be obtained by negotiating for
highly achievable targets (i.e. slack creation) or by
focusing on business matters that improve current
0361-3682/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PI I : S0361- 3682( 99) 00058- 6
Accounting, Organizations and Society 25 (2000) 609±622
www.elsevier.com/locate/aos
$
Data availability: survey data is available from the author
upon request.
* Tel.: +1-213-740-3583; fax: +1-213-747-2815.
E-mail address: [email protected] (W.A. Van der Stede).
period performance while sometimes causing harm
to the long-term e?ectiveness of the ®rm (i.e.
managerial short-term orientation).
1
However, empirical evidence regarding alleged
dysfunctional consequences of a rigid budgetary
control style has been equivocal. In contrast to
Hopwood (1972), Otley (1978) found that rigid
budgetary controls did not lead to increased levels
of budget-related tensions and found only mixed
support for its associated dysfunctional behaviors
(obtaining easy budget targets and having a short-
term view of the job). Instead, Otley (1978) found
that a high emphasis placed on meeting the budget
lead to budgets being more closely met (i.e. higher
budget accuracy). One important variable in this
puzzle was past performance, which seemed to
a?ect both the rigidity of budgetary controls as
well as the incidence of dysfunctional budget-rela-
ted behaviors (budget target manipulation). Otley
(1978) also pointed out that the above relationships
are dependent on the organizational context in
which the budgetary control style is used, such as
an organization's operating environment and size.
Equally to their surprise, Dunk (1993) and
Merchant (1985b) found that budget slack was
low when budget emphasis was high. From their
behavioral priors, they expected budget slack to be
high under a rigid budgetary control style. Orga-
nization economists, however, would argue that
the purpose of rigid budgetary controls is to increase
the likelihood that dysfunctional behaviors get
detected, and therefore, reduced (Merchant, 1985b;
Williamson, 1964).
The current paper maintains that arguments for
a positive and negative relationship between bud-
get emphasis and slack, or other dysfunctional
behaviors, may both contain a grain of truth.
Being able to detect and reduce one form of so-
called dysfunctional behavior might re-emerge
elsewhere in another type, which is not as closely
metered or as easily discernable (Ghoshal &Moran,
1996). Although unexpected, the above-mentioned
budgeting studies did ®nd empirically that rigid
budgetary controls appear to be negatively related
with slack (Dunk, 1993; Merchant, 1985b) and
resulted in apparently improved real budgeting
performance, as opposed to performance due to
manipulated budget standards (Otley, 1978). At
the same time, the managers did not seem to
experience increased stress from the rigid budget-
ary control system (Otley, 1978). However, these
®ndings do not rule out that managers protect
themselves from missing budget targets in other
ways than slack creation, such as by limiting their
exposure to risky, long-term projects. This argu-
ment is consistent with Otley's (1978) ®nding that
managers who are subject to rigid budgetary con-
trols tend to spend a smaller proportion of their
time to long-term planning. But, most studies have
considered only one form of budget-related beha-
vior, which makes an investigation of potential
spillover e?ects impossible. One contribution of
the current study to the management accounting
literature, and the RAPM-literature in particular, is
the inclusion of two types of budget-related
behavior Ðbudget slack and managerial short-term
orientation Ðwhich appear to be negatively related.
However, the budgetary control styles that
organizations implement and the behaviors that
they encourage may be a?ected by contextual
variables, such as business unit past performance
and competitive strategy. First, past performance
is included in the current study as an independent
variable due to its expected e?ect on both the
rigidity of budgetary controls as well as the like-
lihood that managers engage in potentially dys-
®mctional behaviors (Otley, 1978). In line with
these expectations, the results show that business
units that have been more pro®table enjoy more
budget ¯exibility, which provides more leeway for
slack creation which, in turn, tends to reduce the
pressure for short-term results. Second, this study
incorporates competitive strategy as an important
element of an organization's context. The ®ndings
suggest that di?erentiators are subject to less rigid
budgetary controls from upper management allow-
ing them to build slack, which seems in line with the
¯exibility required by their strategy to respond
e?ectively to changes in their environment.
The remainder of the paper is structured as fol-
lows. Section 1 builds on the literature to formulate
1
Research on the organizational and behavioral e?ects of
budgeting, and the reliance on accounting performance measures
(RAPM) for evaluating subordinate managers in particular, is
also known as the RAPM-literature. For a recent review and
critique of this literature, see Hartmann (in press).
610 W.A. Van der Stede / Accounting, Organizations and Society 25 (2000) 609±622
the hypotheses. Section 2 presents the survey data
collected from 153 business units and develops the
measures. The results obtained by means of a
structural equation model are discussed in Section
3. The ®nal sections conclude, discuss the limita-
tions of the present study, and provide directions
for future research.
1. Hypotheses
1.1. Budgetary control style and its consequences
This paper considers two types of potentially
dysfunctional consequences of a rigid budgetary
control style: budget slack creation and managerial
short-termorientation. Slack is generally de®ned as
resources and e?ort toward activities that cannot
be justi®ed easily in terms of their immediate con-
tribution to organizational objectives (March,
1988, p. 4). If the slack is created during the bud-
geting process, then it is commonly called budget
slack. Managerial short-term orientation is
de®ned as the extent to which managers focus on
business matters that will a?ect their performance
within the current budgeting period (1 year).
It is important to emphasize at this point that
there is much disagreement in the literature about
whether slack creation and managerial short-term
orientation are always dysfunctional. As discussed
below, slack sometimes has bene®cial e?ects (e.g. in
strategies that require innovation and experimenta-
tion) and some seemingly short-term management
actions can help both short-term and long-term
results simultaneously (e.g. scrapping develop-
mental projects with little promise).
When business unit managers create slack, they
exploit their position of superior knowledge about
business possibilities vis-a -vis corporate manage-
ment to get performance targets that are deliber-
ately lower than their best guess forecast about the
future (Lukka, 1988). Business unit managers may
bene®t from creating budget slack in one or more
ways. Slack protects them against unforeseen
contingencies and improves the probability that
the budget target will be met, thus increasing the
likelihood of receiving a favorable evaluation (and
associated performance-dependent rewards).
Slack creation, as a means of protection from
the downside potential of an uncertain future, is
particularly valuable in ®rms that treat the budget
as a strong commitment from the manager to the
corporation and use the budget as a primary, if not
exclusive, tool to evaluate management perfor-
mance. Indeed, rigid budgetary controls imply that
salary, resources, and career prospects become
highly dependent on the ability to meet the bud-
get. Therefore, a positive relationship between
emphasis on meeting the budget (a rigid budgetary
control style) and the propensity of managers to
build slack is expected. Slack increases the chances
of making the budget, and thus avoids interven-
tions by upper management, reduces the risk of
being ®red, etc. (Merchant & Manzoni, 1989).
However, the literature has not produced con-
clusive evidence with respect to the e?ect of budget-
ary control style on slack. As aforementioned, Dunk
(1993)
2
and Merchant (1985b)
3
generally found that
slack was lowwhen budget emphasis was high. This
®nding is in line with economic, as opposed to
behavioral, theory which maintains that rigid
budgetary controls should increase the likelihood
that slack gets detected and, therefore, curtailed
(Williamson, 1964). A reduction in managerial
opportunism should, in turn, be associated with
superior performance (Williamson, 1975) since
economists generally view slack as an ineciency
that detracts from the value of the ®rm (Leiben-
stein, 1966). But, organizational economists have
not reached agreement on this, as some studies
found that pro®tability actually deteriorates as
corporate management puts more emphasis on
2
Dunk's (1993) ®ndings involved a three-way interaction of
budget emphasis, budget participation, and information asym-
metry on slack. His results showed that slack is low when par-
ticipation, information asymmetry and budget emphasis are all
high, contrary to the expectation from the literature. From a
three-way interaction it is dicult to infer anything about the
main e?ect of budget emphasis on slack in isolation. However,
across all levels of information asymmetry (from low to high)
and across all levels of budget participation (from low to high),
slack was lower when budget emphasis was high rather than low.
3
Speci®cally, Merchant (1985b) found that two out of three
components that measure the importance placed on meeting
the budget were signi®cantly and negatively related to the pro-
pensity to create budgetary slack. Only one component of
budget emphasis, ``reactions to budget overruns,'' was posi-
tively related to budgetary slack.
W.A. Van der Stede / Accounting, Organizations and Society 25 (2000) 609±622 611
bottom-line ®nancial performance (Hill, 1988). To
explain this result, Hill speculated that rigid
accounting performance-based controls might
encourage short-run pro®t maximization at the
expense of long-run pro®tability, which is the issue
of managerial short-term orientation.
Several studies in the management literature
have indeed documented how an exclusive focus
on accounting-based controls may encourage the
worst practices of management aiming for short-
run pro®t maximization at the expense of long-
term e?ectiveness and competitive strength ( Hayes
& Abernathy, 1980; Laverty, 1996; Merchant,
1990; Merchant & Bruns, 1986). Pressures on
management to perform in the present reduce the
probability of initiating and funding new projects
(Jaeger & Baliga, 1985), sti¯e creativity (Miller,
1986), and undermine commitment to innovation
(Hitt, Hoskisson, Johnson & Moesel, 1996; Hos-
kisson & Hitt, 1988; Hoskisson, Hitt & Hill, 1991).
In short, if budgetary controls are too rigid and
business unit managers have too little discretion,
corporate management may choke all initiatives
that promise long-term or less certain payo?s, as
reported by Merchant (1990).
This discussion indicates that while rigid bud-
getary controls may reduce slack, as empirically
supported by Dunk (1993) and Merchant (1985b),
it also may drive managers to become concerned
primarily about short-term results. This leads to
the expectation that:
H1: Rigid budgetary controls are:
a. Negatively related to budgetary slack;
b. Positively related to managerial short-term
orientation.
One would expect from H1 that budgetary slack
and managerial short-term orientation are nega-
tively related. This is in line with Merchant (1985b)
who suggested that slack provides a cushion in
response to uncertainty through the provision of
resources free from short-term commitment. As
such, slack resources permit managers to more
safely experiment with, for example, new product
introductions and other uncertain innovative pro-
jects (Hambrick & Snow, 1977; Nohria & Gulati,
1996). More direct evidence for a negative rela-
tionship between slack and managerial short-term
orientation was provided by Merchant and Man-
zoni (1989) who found that corporate managers
sometimes allow slack in the budget to reduce the
incentives for business managers to engage in
other dysfunctional behaviors, such as short-term
earnings management. Hence, it is expected that:
H2: Budgetary slack and managerial short-term
orientation are negatively related.
1.2. Impact of business unit competitive strategy
It is generally maintained in the management
accounting literature that the way in which a
business unit competes in its market Ð that is, its
competitive strategy Ð in¯uences the design of the
management control system (Lang®eld-Smith,
1997). Moreover, organization theory has sug-
gested that (some) slack may be needed to success-
fully pursue strategies that require a high degree of
¯exibility to respond e?ectively to changes in the
environment (Bourgeois, 1981). Hence, competitive
strategy is expected to be an important antecedent
of both the budgetary control style implemented
and the occurrence of, or need for, slack.
Porter's (1980) low cost vs. di?erentiation and
Miles and Snow's (1978) defender vs. prospector
strategy-typologies have been used most often in
management control research. Although both
typologies have their own nuances, I assume a
certain degree of isomorphism and comparability
between both typologies Ð as documented in the
management and strategy literatures (Segev,
1989) Ð in the discussion of the literature below.
Generally speaking, cost leaders/defenders have
a narrow product range and undertake little pro-
duct or market development (Miles & Snow,
1978). They focus primarily on achieving a low
cost position relative to competitors and therefore
pursue cost reduction, exploit economies of scale,
standardize the task environment, and produce
standard, undi?erentiated products (Porter, 1980).
Di?erentiators/prospectors, in contrast, actively
engage in market and product development (Miles
& Snow). They strive to create something that is
perceived by the customer as unique by pursuing
superior product features, product innovation, cus-
tomer service, brand image, etc. (Porter).
612 W.A. Van der Stede / Accounting, Organizations and Society 25 (2000) 609±622
Porter (1980) argues that the key to e?ectively
implementing strategy is to explicitly recognize
that di?erent competitive strategies require di?er-
ent organizational arrangements. Management
control is one important organizational arrange-
ment. The focus of the current paper is on
accounting-based budgetary controls, which are
an integral part of the management control system
in most for-pro®t ®rms (Merchant, 1998).
Prior research has not produced conclusive evi-
dence on the relationship between management con-
trol systems and competitive strategy, however
(Lang®eld-Smith, 1997). Simons (1987, 1988), for
instance, found that prospectors emphasize rigid
budgetary controls to a greater extent than defen-
ders. This ®nding con¯icts with the widely held view
that innovation and di?erentiation is best achieved in
organizations that minimize formal controls. Other
studies usually support this view, however, and the
balance of evidence for superior performance is gen-
erally against the combination of di?erentiation/
prospector strategies with rigid formal (budgetary)
controls (Govindarajan, 1988; Govindarajan &
Fisher, 1990). Hence, evidence seems to suggest that
di¯erentiators/prospectors put less emphasis on
budgetary controls or apply themless rigidly.
In the above studies, uncertainty is the main
mechanism through which management control
systems are hypothesized to vary across competi-
tive strategies. Di?erentiators/prospectors face
much uncertainty because they have broad pro-
duct lines, engage in product innovation, deal with
products that have not yet crystallized, etc. Cost
leaders/defenders, in contrast, keep their essentially
undi?erentiated product o?erings relatively stable
over time (Fisher & Govindarajan, 1993; Govin-
darajan, 1986, 1988). In sum, the critical success
factors associated with di?erentiation/prospector
strategies, such as new product development and
innovation, are of a long-termnature and dicult to
quantify (Lang®eld-Smith, 1997), which makes
reliance on formal accounting-based budgetary con-
trols less suitable (Merchant, 1985b; Simons, 1988).
Besides the bounded ability for di?erentiators/
prospectors to rely heavily on accounting-based
budgetary controls, the uncertainty surrounding
their strategy also requires a higher degree of
¯exibility to respond e?ectively to changes in the
environment. One important way to hedge against
uncertainty is the creation of slack resources,
which provide a cushion to support the exploita-
tion of market opportunities and a source of funds
to experiment with product innovations (Bour-
geois, 1981; Cyert & March, 1963). Thus, di?er-
entiators/prospectors may need some slack to
pursue the critical success factors on which their
strategies are build.
4
In a budgeting context, slack is built by setting
budget targets so that they become easier to achieve
(Lukka, 1988). For most divisionalized organiza-
tions, slack is embodied in the budget (Schi? &
Lewin, 1970). Merchant (1985b) argued that the
ability to set accurate budget targets and to mea-
sure performance precisely, which is likely to be the
case for low cost/defender business units, provides
the opportunity to prevent the introduction of slack.
The environments in which di?erentiators/pro-
spectors operate, on the other hand, make it more
dicult for corporate management to detect slack.
Furthermore, Williamson (1964) maintained that
slack creation is potentially restricted as cost cut-
ting, standardization, economies of scale, etc., are
emphasized, which is again the case for low cost/
defender business units. For di?erentiation/pro-
spector business units, on the other hand, corpo-
rate management simply may not wish to reduce
slack to the point where it chokes innovation or
prevents managers from exploring new market
opportunities. Thus, for low cost/defender busi-
ness units there is both a lesser need for slack and
a higher chance of detecting it. The opposite holds
for di?erentiation/prospector business units. This
discussion leads to the following hypotheses:
H3: Relative to cost leadership strategies,
di?erentiation strategies are:
a. Negatively associated with rigid
budgetary controls;
b. Positively associated with budgetary slack.
There is no theory that relates competitive strat-
egy to managerial time-orientation (short-term vs.
4
It has been suggested that there may be an optimal level of
slack and that either too little or too much slack is dysfunctional
only. The literature remains troubled about this issue (Nohria
& Gulati, 1996).
W.A. Van der Stede / Accounting, Organizations and Society 25 (2000) 609±622 613
long-term). A crucial argument in Porter's theory
of competitive strategy is that both cost leadership
and di?erentiation should result in a ``sustainable''
competitive advantage (Porter, 1980). Due to lack
of theory, no formal hypothesis is formulated.
1.3. Relationship with performance
The control±performance link has been notor-
iously problematic in the management control lit-
erature (Merchant & Simons, 1986; Otley, 1980).
Moreover, it is not always clear to what extent the
so-called control dysfunctions are truly undesir-
able or harmful for performance (Jaworski &
Young, 1992; Nohria & Gulati, 1996).
Otley (1978) and Merchant (1985a) have called
for using performance as an independent variable
because the adoption of certain controls is likely
to be in response to low or high past performance.
As such, it is expected that business units that
have been more pro®table may be less a?ected by
rigid budgetary controls and enjoy more ¯exibility
in the expenditures they make (Merchant, 1985a).
Similarly, organization theorists have called for
using slack as a dependent variable and discover-
ing its antecedents rather than being preoccupied
with eliminating it based on theories of organiza-
tional eciency (Bourgeois, 1981). Past perfor-
mance is one such antecedent of slack. Empirical
studies have shown that good performance
increases slack and bad performance decreases it
(Onsi, 1973; Schi? & Lewin, 1970).
Moreover, Merchant and Manzoni (1989) found
that managers who have demonstrated good per-
formance over an extended period enjoy the lar-
gest amounts of budgetary slack. Alternatively,
they found that an urgent need for immediate
pro®ts is an important reason for corporate man-
agement to reduce slack, even if it comes with
short-term management actions that may be
harmful in the long-run (e.g. slashing develop-
mental expenditures). Overall, in their ®eld study,
the inverse relationship between slack and pres-
sure for short-run pro®t is manifest, as well as the
tolerance for slack in good times and the attenua-
tion of slack during bad times, even when it
encourages potentially harmful short-term man-
agement actions. This trade-o? is also mentioned
by Simons (1988, p. 278) who argues that success-
fully eliminating slack may improve performance,
at least ``on a short-term basis.'' This discussion
translates into the following hypotheses:
H4: Past business unit performance is
a. Negatively related to rigid budgetary
controls;
b. Positively related to budgetary slack; and,
c. Negatively related to managerial short-term
orientation.
2. Data, questionaire, and measures
2.1. Sample and data
Data were collected as part of a larger study
that investigates management control issues in
large, diversi®ed ®rms headquartered in Belgium.
5
A questionnaire was developed following the
Total Design Method, the details of which are
given in Dillman (1978). Respondents are business
unit general managers with a direct reporting line
to corporate (i.e. the level immediate below cor-
porate with strategic business unit responsibility).
Prior to mailing it, the survey was submitted to the
scrutiny of three faculty colleagues, eight business
unit managers, and three controllers for pre-test-
ing. Two follow-ups were administered to non-
respondents: 2 weeks (reminder only) and 4 weeks
(replacement questionnaire) after the original
mail-out. In total, 341 surveys were mailed to
business unit managers in May 1996.
6
Response to
5
The corporate sample consists of 37 independent compa-
nies headquartered in Belgium with average corporate con-
solidated sales of 65.5 billion Belgian francs (2.2 billion US
dollars). ``Independent'' means that no other company holds a
stake in the ®rm equal to or greater than 50%.
6
Geographical coverage of the business units was as fol-
lows: 238 (70%) in Belgium; 95 (28%) in other European
countries; and 8 (2%) in the rest of the world (mainly USA and
Canada). The high English literacy rate among the target
managers enabled the survey to be conducted in English.
Indeed, 134 respondents (88%) indicated that they use English
in business-related communication at least occasionally (91
managers or 60% use English at least frequently). Moreover,
112 respondents (74%) have a college degree or higher. The use
of English as the common language of the research overcomes
interpretation problems associated with the translation of
questionnaires into other languages.
614 W.A. Van der Stede / Accounting, Organizations and Society 25 (2000) 609±622
the questionnaire was acceptable: 190 managers
replied (56%). Of these replies, 37 were not valid
(usable response is 153 or 45%).
On average, respondents have been employed by
their current corporation for about 13.7 years and
have been head of their business unit for 5.7 years.
The business units represent a variety of indus-
tries,
7
have average sales of 4.8 billion Belgian
francs (160 million US dollars); and employ an
average of 427 employees.
2.2. Development of measures
2.2.1. Budgetary control style
A rigid budgetary control style is one in which
managers are evaluated primarily on whether or not
they achieved their budget. The rigidity of budget-
ary control is measured by the emphasis placed on
meeting the budget. The scale has seven items:
. I am constantly reminded by the corporate
parent of the need to meet budget targets;
. Corporate superiors judge my performance
predominantly on the basis of attaining
budget goals;
. Control over my business is achieved by the
corporate parent principally by monitoring
how well my budget is on target;
. In the eyes of my corporate superiors,
achieving the budget is an accurate re¯ection
of whether I am succeeding in my business;
. Not achieving my budget has a strong
impact on how my performance is rated by
my corporate superiors;
. My promotion prospects depend heavily on
my ability to meet the budget;
. In the eyes of my corporate superiors, not
achieving the budget re¯ects poor performance.
Scores range from 1 to 7 (de®nitely false to
true). The higher the score, the more achieving the
budget is emphasized. Cronbach / for the 7-item
scale is 0.83. The 7-item scale was computed by the
equally weighted average of the standardized item
scores (zero mean and unit variance) associated
with each of the seven items. This standardized
composite variable ranges from À2.10 to +1.29
with m=0.00 and s=0.71. A positive score indi-
cates a budgetary control style that is ``rigid,'' i.e.
one in which managerial performance is evaluated
predominantly based on meeting the budget. A
negative score indicates a budgetary control style
that is more ``¯exible,'' i.e. one in which budgeting
information is less relied upon and used only in
conjunction with other sources of information to
evaluate managerial performance.
2.2.2. Budgetary slack
Budget slack has been referred to in the litera-
ture under a variety of labels (Merchant, 1985b).
A budget contains slack if the business unit man-
agers have intentionally set their budget targets
lower than their best guess forecast about the future
so that the budget becomes easier to achieve (Lukka,
1988). Stated di?erently, there is slack in the bud-
get if the business unit managers have been able to
negotiate easy budget targets. Conversely, a budget
has little slack if the probability that it will be met
is low (Merchant & Manzoni, 1989) or if it requires
serious e?ort and a high degree of eciency in
accomplishment (Simons, 1988). Based on these de®-
nitions, budgetary slack was measured by ®ve items:
. I succeed to submit budgets that are easily
attainable;
. Budget targets induce high productivity in
my business unit (reverse coded);
. Budget targets require costs to be managed
carefully in my business unit (reverse coded);
. Budget targets have not caused me to be
particularly concerned with improving e-
ciency in my business unit. [Scores range
from 1 (de®nitely false) to 7 (de®nitely true).]
. The ®fth item is a fully-anchored question
asking whether the budget is (1) very easy to
attain; (2) attainable with reasonable e?ort; (3)
attainable with considerable e?ort; (4) practi-
cally unattainable; or (5) impossible to attain.
Cronbach / for the 5-item scale is 0.68, which is
acceptable for scales with relatively few items
(Hinkin, 1995; Nunnally, 1978). The scale for
7
The business units represent a variety of industries that
include both manufacturing (steel, lumber, paper, plastics,
chemicals, machinery, electronics, textiles, cement, construc-
tion, and food) and service [transportation, (non)durable goods
wholesale, merchandise and food stores, auto dealers and gas
stations, and hotels and lodging].
W.A. Van der Stede / Accounting, Organizations and Society 25 (2000) 609±622 615
budget slack was computed by the equally weigh-
ted average of the standardized item scores (zero
mean and unit variance) associated with each of
the ®ve items. This standardized composite vari-
able ranges from À1.32 to +2.12 with m=0.00
and s=0.69.
As a matter of construct validation, a factor
analysis was performed on all items representing
emphasis on meeting the budget and budget slack
to ensure that both variables cover separate con-
structs. This analysis resulted in two factors. All
items representing emphasis on meeting the budget
loaded on factor 1 and all items representing bud-
get slack loaded on factor 2 (all above 0.50). This
con®rms that the two variables measured by these
items clearly represent two di?erent constructs.
2.2.3. Managerial time-orientation
This scale asks managers to indicate the time
spent on matters that will appear in the pro®t and
loss statement within 1 month or less, 1 month to 1
quarter, 1 quarter to 1 year, and 1 year to 5 years.
Merchant (1990) and Otley (1978) previously used
this instrument also. As in Merchant (1990), the
sum of the percentages of the ®rst three categories
(e?ect within one year) was used as an indicator of
the managers' short-term orientation (m=78.31,
s=16.65).
2.2.4. Competitive strategy
Business unit competitive strategy was oper-
ationalized as in Govindarajan and Fisher (1990),
using Porter's (1980) low cost vs. di?erentiation
typology. A cost leader aims to achieve low cost
relative to competitors and vigorously pursues cost
reduction, exploits economies of scale, standardizes
the task environment, and produces standard
undi?erentiated products. A di?erentiator, on the
other hand, creates something that is perceived by
customers as unique and pays more attention to
superior product features, customer service, brand
image, etc. (Porter). Respondents were asked to
indicate the percentage of their business unit's
current sales accounted for by either of these stra-
tegies. In addition, the business unit managers
were asked to position their business relative to
competitors in terms of (i) product selling price;
(ii) R&D expenditures; (iii) product quality; (iv)
brand image; and (v) product features. The scale
ranged from 1 (my business unit is positioned
signi®cantly lower) to 7 (signi®cantly higher). Each
of these ®ve items is individually correlated with the
percentage di?erentiation (two-tail P
Previous studies have provided contradictory evidence with respect to the eect of rigid budgetary controls on slack
and other dysfunctional behaviors. One motivation for the current study was to test whether spillover eects exist
between two alleged dysfunctional consequences of a rigid budgetary control style: budget slack creation and man-
agerial short-term orientation. The data support this contention: reducing one form of dysfunctional behavior (slack
creation) through rigid controls seems to spill over into another form (stronger management focus on business matters
that aect short-term results). However, the budgetary control styles that organizations implement, as well as the
behaviors that they encourage, may be aected by two important antecedents: business unit past performance and
competitive strategy.
The relationship between two consequences of budgetary
controls: budgetary slack creation and managerial
short-term orientation
$
Wim A. Van der Stede*
University of Southern California, Leventhal School of Accounting, Los Angeles, CA 90089-0441, USA
Abstract
Previous studies have provided contradictory evidence with respect to the e?ect of rigid budgetary controls on slack
and other dysfunctional behaviors. One motivation for the current study was to test whether spillover e?ects exist
between two alleged dysfunctional consequences of a rigid budgetary control style: budget slack creation and man-
agerial short-term orientation. The data support this contention: reducing one form of dysfunctional behavior (slack
creation) through rigid controls seems to spill over into another form (stronger management focus on business matters
that a?ect short-term results). However, the budgetary control styles that organizations implement, as well as the
behaviors that they encourage, may be a?ected by two important antecedents: business unit past performance and
competitive strategy. The results indicate that business units that either pursue a di?erentiation strategy or have been
more pro®table are subject to less rigid budgetary controls, which augment the propensity to build slack as well as the
tendency for managers to think long-term. These relationships are tested in a structural equation model on survey data
obtained from 153 business unit general managers. # 2000 Elsevier Science Ltd. All rights reserved.
Since Hopwood's (1972) seminal paper, the
budgeting literature has shown great interest in
understanding possible e?ects of budgetary control
styles. It is generally maintained that the incidence
of so-called dysfunctional behavior is a?ected by
the rigidity of budgetary controls. A rigid budget-
ary control style is one in which employees, mostly
at management organization levels, are evaluated
primarily on whether or not they achieved their
budget. When evaluated in this way, managers are
held fully accountable for their performance as
measured by the budget. This implies that salary,
resources, and career prospects become highly, if
not fully, dependent on the managers' ability to
meet the budget. Managers who miss the targets
face the prospect of interventions by upper man-
agement, the loss of organizational resources, the
loss of annual bonuses, and ultimately the loss of
their job (Merchant & Manzoni, 1989). Under
these circumstances, managers may look for ways
to protect themselves from the downside risk of
missing budget targets and the stigma normally
attached to underachievers (Lukka, 1988; Onsi,
1973; Schi? & Lewin, 1970). Possible ways of
protection can be obtained by negotiating for
highly achievable targets (i.e. slack creation) or by
focusing on business matters that improve current
0361-3682/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PI I : S0361- 3682( 99) 00058- 6
Accounting, Organizations and Society 25 (2000) 609±622
www.elsevier.com/locate/aos
$
Data availability: survey data is available from the author
upon request.
* Tel.: +1-213-740-3583; fax: +1-213-747-2815.
E-mail address: [email protected] (W.A. Van der Stede).
period performance while sometimes causing harm
to the long-term e?ectiveness of the ®rm (i.e.
managerial short-term orientation).
1
However, empirical evidence regarding alleged
dysfunctional consequences of a rigid budgetary
control style has been equivocal. In contrast to
Hopwood (1972), Otley (1978) found that rigid
budgetary controls did not lead to increased levels
of budget-related tensions and found only mixed
support for its associated dysfunctional behaviors
(obtaining easy budget targets and having a short-
term view of the job). Instead, Otley (1978) found
that a high emphasis placed on meeting the budget
lead to budgets being more closely met (i.e. higher
budget accuracy). One important variable in this
puzzle was past performance, which seemed to
a?ect both the rigidity of budgetary controls as
well as the incidence of dysfunctional budget-rela-
ted behaviors (budget target manipulation). Otley
(1978) also pointed out that the above relationships
are dependent on the organizational context in
which the budgetary control style is used, such as
an organization's operating environment and size.
Equally to their surprise, Dunk (1993) and
Merchant (1985b) found that budget slack was
low when budget emphasis was high. From their
behavioral priors, they expected budget slack to be
high under a rigid budgetary control style. Orga-
nization economists, however, would argue that
the purpose of rigid budgetary controls is to increase
the likelihood that dysfunctional behaviors get
detected, and therefore, reduced (Merchant, 1985b;
Williamson, 1964).
The current paper maintains that arguments for
a positive and negative relationship between bud-
get emphasis and slack, or other dysfunctional
behaviors, may both contain a grain of truth.
Being able to detect and reduce one form of so-
called dysfunctional behavior might re-emerge
elsewhere in another type, which is not as closely
metered or as easily discernable (Ghoshal &Moran,
1996). Although unexpected, the above-mentioned
budgeting studies did ®nd empirically that rigid
budgetary controls appear to be negatively related
with slack (Dunk, 1993; Merchant, 1985b) and
resulted in apparently improved real budgeting
performance, as opposed to performance due to
manipulated budget standards (Otley, 1978). At
the same time, the managers did not seem to
experience increased stress from the rigid budget-
ary control system (Otley, 1978). However, these
®ndings do not rule out that managers protect
themselves from missing budget targets in other
ways than slack creation, such as by limiting their
exposure to risky, long-term projects. This argu-
ment is consistent with Otley's (1978) ®nding that
managers who are subject to rigid budgetary con-
trols tend to spend a smaller proportion of their
time to long-term planning. But, most studies have
considered only one form of budget-related beha-
vior, which makes an investigation of potential
spillover e?ects impossible. One contribution of
the current study to the management accounting
literature, and the RAPM-literature in particular, is
the inclusion of two types of budget-related
behavior Ðbudget slack and managerial short-term
orientation Ðwhich appear to be negatively related.
However, the budgetary control styles that
organizations implement and the behaviors that
they encourage may be a?ected by contextual
variables, such as business unit past performance
and competitive strategy. First, past performance
is included in the current study as an independent
variable due to its expected e?ect on both the
rigidity of budgetary controls as well as the like-
lihood that managers engage in potentially dys-
®mctional behaviors (Otley, 1978). In line with
these expectations, the results show that business
units that have been more pro®table enjoy more
budget ¯exibility, which provides more leeway for
slack creation which, in turn, tends to reduce the
pressure for short-term results. Second, this study
incorporates competitive strategy as an important
element of an organization's context. The ®ndings
suggest that di?erentiators are subject to less rigid
budgetary controls from upper management allow-
ing them to build slack, which seems in line with the
¯exibility required by their strategy to respond
e?ectively to changes in their environment.
The remainder of the paper is structured as fol-
lows. Section 1 builds on the literature to formulate
1
Research on the organizational and behavioral e?ects of
budgeting, and the reliance on accounting performance measures
(RAPM) for evaluating subordinate managers in particular, is
also known as the RAPM-literature. For a recent review and
critique of this literature, see Hartmann (in press).
610 W.A. Van der Stede / Accounting, Organizations and Society 25 (2000) 609±622
the hypotheses. Section 2 presents the survey data
collected from 153 business units and develops the
measures. The results obtained by means of a
structural equation model are discussed in Section
3. The ®nal sections conclude, discuss the limita-
tions of the present study, and provide directions
for future research.
1. Hypotheses
1.1. Budgetary control style and its consequences
This paper considers two types of potentially
dysfunctional consequences of a rigid budgetary
control style: budget slack creation and managerial
short-termorientation. Slack is generally de®ned as
resources and e?ort toward activities that cannot
be justi®ed easily in terms of their immediate con-
tribution to organizational objectives (March,
1988, p. 4). If the slack is created during the bud-
geting process, then it is commonly called budget
slack. Managerial short-term orientation is
de®ned as the extent to which managers focus on
business matters that will a?ect their performance
within the current budgeting period (1 year).
It is important to emphasize at this point that
there is much disagreement in the literature about
whether slack creation and managerial short-term
orientation are always dysfunctional. As discussed
below, slack sometimes has bene®cial e?ects (e.g. in
strategies that require innovation and experimenta-
tion) and some seemingly short-term management
actions can help both short-term and long-term
results simultaneously (e.g. scrapping develop-
mental projects with little promise).
When business unit managers create slack, they
exploit their position of superior knowledge about
business possibilities vis-a -vis corporate manage-
ment to get performance targets that are deliber-
ately lower than their best guess forecast about the
future (Lukka, 1988). Business unit managers may
bene®t from creating budget slack in one or more
ways. Slack protects them against unforeseen
contingencies and improves the probability that
the budget target will be met, thus increasing the
likelihood of receiving a favorable evaluation (and
associated performance-dependent rewards).
Slack creation, as a means of protection from
the downside potential of an uncertain future, is
particularly valuable in ®rms that treat the budget
as a strong commitment from the manager to the
corporation and use the budget as a primary, if not
exclusive, tool to evaluate management perfor-
mance. Indeed, rigid budgetary controls imply that
salary, resources, and career prospects become
highly dependent on the ability to meet the bud-
get. Therefore, a positive relationship between
emphasis on meeting the budget (a rigid budgetary
control style) and the propensity of managers to
build slack is expected. Slack increases the chances
of making the budget, and thus avoids interven-
tions by upper management, reduces the risk of
being ®red, etc. (Merchant & Manzoni, 1989).
However, the literature has not produced con-
clusive evidence with respect to the e?ect of budget-
ary control style on slack. As aforementioned, Dunk
(1993)
2
and Merchant (1985b)
3
generally found that
slack was lowwhen budget emphasis was high. This
®nding is in line with economic, as opposed to
behavioral, theory which maintains that rigid
budgetary controls should increase the likelihood
that slack gets detected and, therefore, curtailed
(Williamson, 1964). A reduction in managerial
opportunism should, in turn, be associated with
superior performance (Williamson, 1975) since
economists generally view slack as an ineciency
that detracts from the value of the ®rm (Leiben-
stein, 1966). But, organizational economists have
not reached agreement on this, as some studies
found that pro®tability actually deteriorates as
corporate management puts more emphasis on
2
Dunk's (1993) ®ndings involved a three-way interaction of
budget emphasis, budget participation, and information asym-
metry on slack. His results showed that slack is low when par-
ticipation, information asymmetry and budget emphasis are all
high, contrary to the expectation from the literature. From a
three-way interaction it is dicult to infer anything about the
main e?ect of budget emphasis on slack in isolation. However,
across all levels of information asymmetry (from low to high)
and across all levels of budget participation (from low to high),
slack was lower when budget emphasis was high rather than low.
3
Speci®cally, Merchant (1985b) found that two out of three
components that measure the importance placed on meeting
the budget were signi®cantly and negatively related to the pro-
pensity to create budgetary slack. Only one component of
budget emphasis, ``reactions to budget overruns,'' was posi-
tively related to budgetary slack.
W.A. Van der Stede / Accounting, Organizations and Society 25 (2000) 609±622 611
bottom-line ®nancial performance (Hill, 1988). To
explain this result, Hill speculated that rigid
accounting performance-based controls might
encourage short-run pro®t maximization at the
expense of long-run pro®tability, which is the issue
of managerial short-term orientation.
Several studies in the management literature
have indeed documented how an exclusive focus
on accounting-based controls may encourage the
worst practices of management aiming for short-
run pro®t maximization at the expense of long-
term e?ectiveness and competitive strength ( Hayes
& Abernathy, 1980; Laverty, 1996; Merchant,
1990; Merchant & Bruns, 1986). Pressures on
management to perform in the present reduce the
probability of initiating and funding new projects
(Jaeger & Baliga, 1985), sti¯e creativity (Miller,
1986), and undermine commitment to innovation
(Hitt, Hoskisson, Johnson & Moesel, 1996; Hos-
kisson & Hitt, 1988; Hoskisson, Hitt & Hill, 1991).
In short, if budgetary controls are too rigid and
business unit managers have too little discretion,
corporate management may choke all initiatives
that promise long-term or less certain payo?s, as
reported by Merchant (1990).
This discussion indicates that while rigid bud-
getary controls may reduce slack, as empirically
supported by Dunk (1993) and Merchant (1985b),
it also may drive managers to become concerned
primarily about short-term results. This leads to
the expectation that:
H1: Rigid budgetary controls are:
a. Negatively related to budgetary slack;
b. Positively related to managerial short-term
orientation.
One would expect from H1 that budgetary slack
and managerial short-term orientation are nega-
tively related. This is in line with Merchant (1985b)
who suggested that slack provides a cushion in
response to uncertainty through the provision of
resources free from short-term commitment. As
such, slack resources permit managers to more
safely experiment with, for example, new product
introductions and other uncertain innovative pro-
jects (Hambrick & Snow, 1977; Nohria & Gulati,
1996). More direct evidence for a negative rela-
tionship between slack and managerial short-term
orientation was provided by Merchant and Man-
zoni (1989) who found that corporate managers
sometimes allow slack in the budget to reduce the
incentives for business managers to engage in
other dysfunctional behaviors, such as short-term
earnings management. Hence, it is expected that:
H2: Budgetary slack and managerial short-term
orientation are negatively related.
1.2. Impact of business unit competitive strategy
It is generally maintained in the management
accounting literature that the way in which a
business unit competes in its market Ð that is, its
competitive strategy Ð in¯uences the design of the
management control system (Lang®eld-Smith,
1997). Moreover, organization theory has sug-
gested that (some) slack may be needed to success-
fully pursue strategies that require a high degree of
¯exibility to respond e?ectively to changes in the
environment (Bourgeois, 1981). Hence, competitive
strategy is expected to be an important antecedent
of both the budgetary control style implemented
and the occurrence of, or need for, slack.
Porter's (1980) low cost vs. di?erentiation and
Miles and Snow's (1978) defender vs. prospector
strategy-typologies have been used most often in
management control research. Although both
typologies have their own nuances, I assume a
certain degree of isomorphism and comparability
between both typologies Ð as documented in the
management and strategy literatures (Segev,
1989) Ð in the discussion of the literature below.
Generally speaking, cost leaders/defenders have
a narrow product range and undertake little pro-
duct or market development (Miles & Snow,
1978). They focus primarily on achieving a low
cost position relative to competitors and therefore
pursue cost reduction, exploit economies of scale,
standardize the task environment, and produce
standard, undi?erentiated products (Porter, 1980).
Di?erentiators/prospectors, in contrast, actively
engage in market and product development (Miles
& Snow). They strive to create something that is
perceived by the customer as unique by pursuing
superior product features, product innovation, cus-
tomer service, brand image, etc. (Porter).
612 W.A. Van der Stede / Accounting, Organizations and Society 25 (2000) 609±622
Porter (1980) argues that the key to e?ectively
implementing strategy is to explicitly recognize
that di?erent competitive strategies require di?er-
ent organizational arrangements. Management
control is one important organizational arrange-
ment. The focus of the current paper is on
accounting-based budgetary controls, which are
an integral part of the management control system
in most for-pro®t ®rms (Merchant, 1998).
Prior research has not produced conclusive evi-
dence on the relationship between management con-
trol systems and competitive strategy, however
(Lang®eld-Smith, 1997). Simons (1987, 1988), for
instance, found that prospectors emphasize rigid
budgetary controls to a greater extent than defen-
ders. This ®nding con¯icts with the widely held view
that innovation and di?erentiation is best achieved in
organizations that minimize formal controls. Other
studies usually support this view, however, and the
balance of evidence for superior performance is gen-
erally against the combination of di?erentiation/
prospector strategies with rigid formal (budgetary)
controls (Govindarajan, 1988; Govindarajan &
Fisher, 1990). Hence, evidence seems to suggest that
di¯erentiators/prospectors put less emphasis on
budgetary controls or apply themless rigidly.
In the above studies, uncertainty is the main
mechanism through which management control
systems are hypothesized to vary across competi-
tive strategies. Di?erentiators/prospectors face
much uncertainty because they have broad pro-
duct lines, engage in product innovation, deal with
products that have not yet crystallized, etc. Cost
leaders/defenders, in contrast, keep their essentially
undi?erentiated product o?erings relatively stable
over time (Fisher & Govindarajan, 1993; Govin-
darajan, 1986, 1988). In sum, the critical success
factors associated with di?erentiation/prospector
strategies, such as new product development and
innovation, are of a long-termnature and dicult to
quantify (Lang®eld-Smith, 1997), which makes
reliance on formal accounting-based budgetary con-
trols less suitable (Merchant, 1985b; Simons, 1988).
Besides the bounded ability for di?erentiators/
prospectors to rely heavily on accounting-based
budgetary controls, the uncertainty surrounding
their strategy also requires a higher degree of
¯exibility to respond e?ectively to changes in the
environment. One important way to hedge against
uncertainty is the creation of slack resources,
which provide a cushion to support the exploita-
tion of market opportunities and a source of funds
to experiment with product innovations (Bour-
geois, 1981; Cyert & March, 1963). Thus, di?er-
entiators/prospectors may need some slack to
pursue the critical success factors on which their
strategies are build.
4
In a budgeting context, slack is built by setting
budget targets so that they become easier to achieve
(Lukka, 1988). For most divisionalized organiza-
tions, slack is embodied in the budget (Schi? &
Lewin, 1970). Merchant (1985b) argued that the
ability to set accurate budget targets and to mea-
sure performance precisely, which is likely to be the
case for low cost/defender business units, provides
the opportunity to prevent the introduction of slack.
The environments in which di?erentiators/pro-
spectors operate, on the other hand, make it more
dicult for corporate management to detect slack.
Furthermore, Williamson (1964) maintained that
slack creation is potentially restricted as cost cut-
ting, standardization, economies of scale, etc., are
emphasized, which is again the case for low cost/
defender business units. For di?erentiation/pro-
spector business units, on the other hand, corpo-
rate management simply may not wish to reduce
slack to the point where it chokes innovation or
prevents managers from exploring new market
opportunities. Thus, for low cost/defender busi-
ness units there is both a lesser need for slack and
a higher chance of detecting it. The opposite holds
for di?erentiation/prospector business units. This
discussion leads to the following hypotheses:
H3: Relative to cost leadership strategies,
di?erentiation strategies are:
a. Negatively associated with rigid
budgetary controls;
b. Positively associated with budgetary slack.
There is no theory that relates competitive strat-
egy to managerial time-orientation (short-term vs.
4
It has been suggested that there may be an optimal level of
slack and that either too little or too much slack is dysfunctional
only. The literature remains troubled about this issue (Nohria
& Gulati, 1996).
W.A. Van der Stede / Accounting, Organizations and Society 25 (2000) 609±622 613
long-term). A crucial argument in Porter's theory
of competitive strategy is that both cost leadership
and di?erentiation should result in a ``sustainable''
competitive advantage (Porter, 1980). Due to lack
of theory, no formal hypothesis is formulated.
1.3. Relationship with performance
The control±performance link has been notor-
iously problematic in the management control lit-
erature (Merchant & Simons, 1986; Otley, 1980).
Moreover, it is not always clear to what extent the
so-called control dysfunctions are truly undesir-
able or harmful for performance (Jaworski &
Young, 1992; Nohria & Gulati, 1996).
Otley (1978) and Merchant (1985a) have called
for using performance as an independent variable
because the adoption of certain controls is likely
to be in response to low or high past performance.
As such, it is expected that business units that
have been more pro®table may be less a?ected by
rigid budgetary controls and enjoy more ¯exibility
in the expenditures they make (Merchant, 1985a).
Similarly, organization theorists have called for
using slack as a dependent variable and discover-
ing its antecedents rather than being preoccupied
with eliminating it based on theories of organiza-
tional eciency (Bourgeois, 1981). Past perfor-
mance is one such antecedent of slack. Empirical
studies have shown that good performance
increases slack and bad performance decreases it
(Onsi, 1973; Schi? & Lewin, 1970).
Moreover, Merchant and Manzoni (1989) found
that managers who have demonstrated good per-
formance over an extended period enjoy the lar-
gest amounts of budgetary slack. Alternatively,
they found that an urgent need for immediate
pro®ts is an important reason for corporate man-
agement to reduce slack, even if it comes with
short-term management actions that may be
harmful in the long-run (e.g. slashing develop-
mental expenditures). Overall, in their ®eld study,
the inverse relationship between slack and pres-
sure for short-run pro®t is manifest, as well as the
tolerance for slack in good times and the attenua-
tion of slack during bad times, even when it
encourages potentially harmful short-term man-
agement actions. This trade-o? is also mentioned
by Simons (1988, p. 278) who argues that success-
fully eliminating slack may improve performance,
at least ``on a short-term basis.'' This discussion
translates into the following hypotheses:
H4: Past business unit performance is
a. Negatively related to rigid budgetary
controls;
b. Positively related to budgetary slack; and,
c. Negatively related to managerial short-term
orientation.
2. Data, questionaire, and measures
2.1. Sample and data
Data were collected as part of a larger study
that investigates management control issues in
large, diversi®ed ®rms headquartered in Belgium.
5
A questionnaire was developed following the
Total Design Method, the details of which are
given in Dillman (1978). Respondents are business
unit general managers with a direct reporting line
to corporate (i.e. the level immediate below cor-
porate with strategic business unit responsibility).
Prior to mailing it, the survey was submitted to the
scrutiny of three faculty colleagues, eight business
unit managers, and three controllers for pre-test-
ing. Two follow-ups were administered to non-
respondents: 2 weeks (reminder only) and 4 weeks
(replacement questionnaire) after the original
mail-out. In total, 341 surveys were mailed to
business unit managers in May 1996.
6
Response to
5
The corporate sample consists of 37 independent compa-
nies headquartered in Belgium with average corporate con-
solidated sales of 65.5 billion Belgian francs (2.2 billion US
dollars). ``Independent'' means that no other company holds a
stake in the ®rm equal to or greater than 50%.
6
Geographical coverage of the business units was as fol-
lows: 238 (70%) in Belgium; 95 (28%) in other European
countries; and 8 (2%) in the rest of the world (mainly USA and
Canada). The high English literacy rate among the target
managers enabled the survey to be conducted in English.
Indeed, 134 respondents (88%) indicated that they use English
in business-related communication at least occasionally (91
managers or 60% use English at least frequently). Moreover,
112 respondents (74%) have a college degree or higher. The use
of English as the common language of the research overcomes
interpretation problems associated with the translation of
questionnaires into other languages.
614 W.A. Van der Stede / Accounting, Organizations and Society 25 (2000) 609±622
the questionnaire was acceptable: 190 managers
replied (56%). Of these replies, 37 were not valid
(usable response is 153 or 45%).
On average, respondents have been employed by
their current corporation for about 13.7 years and
have been head of their business unit for 5.7 years.
The business units represent a variety of indus-
tries,
7
have average sales of 4.8 billion Belgian
francs (160 million US dollars); and employ an
average of 427 employees.
2.2. Development of measures
2.2.1. Budgetary control style
A rigid budgetary control style is one in which
managers are evaluated primarily on whether or not
they achieved their budget. The rigidity of budget-
ary control is measured by the emphasis placed on
meeting the budget. The scale has seven items:
. I am constantly reminded by the corporate
parent of the need to meet budget targets;
. Corporate superiors judge my performance
predominantly on the basis of attaining
budget goals;
. Control over my business is achieved by the
corporate parent principally by monitoring
how well my budget is on target;
. In the eyes of my corporate superiors,
achieving the budget is an accurate re¯ection
of whether I am succeeding in my business;
. Not achieving my budget has a strong
impact on how my performance is rated by
my corporate superiors;
. My promotion prospects depend heavily on
my ability to meet the budget;
. In the eyes of my corporate superiors, not
achieving the budget re¯ects poor performance.
Scores range from 1 to 7 (de®nitely false to
true). The higher the score, the more achieving the
budget is emphasized. Cronbach / for the 7-item
scale is 0.83. The 7-item scale was computed by the
equally weighted average of the standardized item
scores (zero mean and unit variance) associated
with each of the seven items. This standardized
composite variable ranges from À2.10 to +1.29
with m=0.00 and s=0.71. A positive score indi-
cates a budgetary control style that is ``rigid,'' i.e.
one in which managerial performance is evaluated
predominantly based on meeting the budget. A
negative score indicates a budgetary control style
that is more ``¯exible,'' i.e. one in which budgeting
information is less relied upon and used only in
conjunction with other sources of information to
evaluate managerial performance.
2.2.2. Budgetary slack
Budget slack has been referred to in the litera-
ture under a variety of labels (Merchant, 1985b).
A budget contains slack if the business unit man-
agers have intentionally set their budget targets
lower than their best guess forecast about the future
so that the budget becomes easier to achieve (Lukka,
1988). Stated di?erently, there is slack in the bud-
get if the business unit managers have been able to
negotiate easy budget targets. Conversely, a budget
has little slack if the probability that it will be met
is low (Merchant & Manzoni, 1989) or if it requires
serious e?ort and a high degree of eciency in
accomplishment (Simons, 1988). Based on these de®-
nitions, budgetary slack was measured by ®ve items:
. I succeed to submit budgets that are easily
attainable;
. Budget targets induce high productivity in
my business unit (reverse coded);
. Budget targets require costs to be managed
carefully in my business unit (reverse coded);
. Budget targets have not caused me to be
particularly concerned with improving e-
ciency in my business unit. [Scores range
from 1 (de®nitely false) to 7 (de®nitely true).]
. The ®fth item is a fully-anchored question
asking whether the budget is (1) very easy to
attain; (2) attainable with reasonable e?ort; (3)
attainable with considerable e?ort; (4) practi-
cally unattainable; or (5) impossible to attain.
Cronbach / for the 5-item scale is 0.68, which is
acceptable for scales with relatively few items
(Hinkin, 1995; Nunnally, 1978). The scale for
7
The business units represent a variety of industries that
include both manufacturing (steel, lumber, paper, plastics,
chemicals, machinery, electronics, textiles, cement, construc-
tion, and food) and service [transportation, (non)durable goods
wholesale, merchandise and food stores, auto dealers and gas
stations, and hotels and lodging].
W.A. Van der Stede / Accounting, Organizations and Society 25 (2000) 609±622 615
budget slack was computed by the equally weigh-
ted average of the standardized item scores (zero
mean and unit variance) associated with each of
the ®ve items. This standardized composite vari-
able ranges from À1.32 to +2.12 with m=0.00
and s=0.69.
As a matter of construct validation, a factor
analysis was performed on all items representing
emphasis on meeting the budget and budget slack
to ensure that both variables cover separate con-
structs. This analysis resulted in two factors. All
items representing emphasis on meeting the budget
loaded on factor 1 and all items representing bud-
get slack loaded on factor 2 (all above 0.50). This
con®rms that the two variables measured by these
items clearly represent two di?erent constructs.
2.2.3. Managerial time-orientation
This scale asks managers to indicate the time
spent on matters that will appear in the pro®t and
loss statement within 1 month or less, 1 month to 1
quarter, 1 quarter to 1 year, and 1 year to 5 years.
Merchant (1990) and Otley (1978) previously used
this instrument also. As in Merchant (1990), the
sum of the percentages of the ®rst three categories
(e?ect within one year) was used as an indicator of
the managers' short-term orientation (m=78.31,
s=16.65).
2.2.4. Competitive strategy
Business unit competitive strategy was oper-
ationalized as in Govindarajan and Fisher (1990),
using Porter's (1980) low cost vs. di?erentiation
typology. A cost leader aims to achieve low cost
relative to competitors and vigorously pursues cost
reduction, exploits economies of scale, standardizes
the task environment, and produces standard
undi?erentiated products. A di?erentiator, on the
other hand, creates something that is perceived by
customers as unique and pays more attention to
superior product features, customer service, brand
image, etc. (Porter). Respondents were asked to
indicate the percentage of their business unit's
current sales accounted for by either of these stra-
tegies. In addition, the business unit managers
were asked to position their business relative to
competitors in terms of (i) product selling price;
(ii) R&D expenditures; (iii) product quality; (iv)
brand image; and (v) product features. The scale
ranged from 1 (my business unit is positioned
signi®cantly lower) to 7 (signi®cantly higher). Each
of these ®ve items is individually correlated with the
percentage di?erentiation (two-tail P