Accounting Organizations and Society, Vol. 12, No. 1, pp. 49-61, 1987. 0361-3682/87 53.00+.

00
Printed in Great Britain Pergamon Journals Ltd.

MANAGEMENT CONTROL SYSTEMS AND DEPARTMENTAL

INTERDEPENDENCIES: AN EMPIRICAL STUDY

N. B. MACINTOSH
School of Business, Queen's University, Kingston, Canada

and

R. L. DAFT
Department of Management, Texas A & M University

Abstract

Behavioral accounting research suggests that (1 ) the design and use of a management accounting system
is related to overall characteristics of the organization, and ( 2) a management accounting system is one ele ment
in a package of control systems. The research reported here investigated the relationship between the
organizational characteristic of departmental interdependence and the design and use of three elements in
a package of management controls — the operating budget, periodic statistical reports, and standard
operating policies and procedures. The findings support the hypothesis that departmental interdepen -
dence is related to the emphasis placed on each management control system. Standard operating proce -
dures were an important control device when interdependence was low. The budget and statistical reports
were used more extensively when interdependence was moderate. When interdependence among depart-
ments was high, the role of all three control systems diminished.

The literature on accounting control suggests tems is related to factors such as the external en -
that the design and focus of management ac - vironment (Duncan, 1972; McCann & Selsky,
c o un t in g s ys t e ms m a y b e r e l a t e d t o o ve r a l l 1984), technology (Fry, 1982; Daft & Macin -
characteristics of the organization. From this tosh, 1981) and interdependence among depart -
perspective accounting systems are intertwined ments (Thompson, 1967). Although a growing
with the way organizations function, and they body of research supports the relationship be -
can be studied in conjunction with their organi- tween organization design and environment or
zational setting (Waterhouse & Tiessen, 1979; t e chnolo gy, in te rd ep endenc e a mon g d ep ar t -
Otley, 1980; Ewusi-Mensah, 1981; Birnberg et ments is an area of potential new insight into the
al., 1983; Burchell et al., 1980; Merchant, 1981; design of control systems within organizations.
Hopwood, 1983; Gordon & Narayanan, 1984; Interdependence is the extent to which depart -
Govindarajan & Gupta, 1985). The purpose of ments depend upon each other and exchange in -
the research reported here is to explore the rela- formation and resources to accomplish their re-
tionship of one such chara cteristic — de- spective tasks (Van de Ven et al., 1976; McCann
partmental interdependence — and control sys- & F er r y, 1979) . Th e con ce pt o f int erd ep en -
tem design. dency is proposed as an organizational variable
relevant to control syste ms for t wo rea sons.
First, interdependency reflects workflow and
DEPARTMENTAL INTERDEPENDENCE hence the amount of coordination and feedback
The design of organizations and internal sys - needed among departments. The data available
49
50 N. B. MACINTOSH and R. L. DAFT

for coordination and feedback may be available dinated on the basis of feedback about the pro -
from the management systems used to manage ject and mutual adjustment among the
a n d co n tr o l th o s e d e p a r tme n t s ( Th o mp s o n , specialized departments. In a hospital, for ex -
1967; Van de Yen et al., 1976; Tushman, 1977). ample, a patient may move back and forth among
Second, several accounting researchers have departments (X-ray, surgical, psychiatric) on the
identified interdependency across departments basis of the most current information and feed -
as a potentially important organizational vari - back about the patient's health. Reciprocal inter-
able for future management accounting studies dependence places a heavy demand on manage-
(Watson & Baulmer, 1975; Hayes, 1977; ment for coordination. Standardization and ac -
Ginzberg, 1980; Otley, 1980; Kilmann, 1983; counting information often are not sufficient for
E mma n u el & O tl e y, 1985; Me r ch ant, 1985; coordination, so face-to-face interaction and
Chenhall & Morris, 1986). mutual adjustment may be required.
Thompson (1967) first proposed the role of
interdependence in organizational design, and
defined three types that are widespread in mod - MANAGEMENT CONTROL SYSTEMS
ern organizations — pooled, sequential and re-
ciprocal. Each of these place unique and identifi- The role of management accounting systems
able demands on management systems and pro- in organizational control traditionally has been
cesses. Pooled is the lowest form of interdepen- studied in isolation from characteristics of the
dence. Departments are relatively autonomous total organization and from other non -account-
and little work flows between them. This occurs ing based control systems (Young, 1979;
when departments are self-contained, or when Flamholtz, 1983; Macintosh, 1985). Recently a
they provide services to geographically distri - few researchers have argued that this narro w
buted clients. Branch banks and stores are ex - view is part of the reason that we still do not have
amples of units that operate independently of a good understanding of how management ac -
each other. Operating units share resources (e.g. counting systems function ( Gordon & Miller,
financial, advertising) from a common pool. The 1976; Otley & Berry, 1980; Daft & Macintosh,
low level of interdependence leads to standard - 1984).
ized coordination through rules and procedures. The reason accounting systems alone repre -
Sequential interdependence involves the link - sent a narrow view is that the package of formal
age of organizational departments in serial fash - controls in an organization typically includes:
ion. The output of one department becomes a di- accounting reports, the budget, formal hierar -
rect input to the next department. Each depart - chy and supervision; job descriptions; rules and
ment completes its work by depending on work standard operating procedures; statistics for
from preceding departments. Breakdowns and measuring performance; organization stru cture;
disruptions in this work flow do occur and place employee performance appraisal systems; and
greater coordination and control demands on corporate culture (Lawler, 1976; Flamholtz,
the organization than does pooled interdepen - 1983). These controls may seem an ad hoc col-
-

dence. Here accounting and control syste ms lection of techniques and mechanisms, but in
may be used to facilitate planning and schedul - many cases they are the tangible elements of a
ing, and also encourage feedback to coordinate strategy to create an integrated organizational
workflow between departments. control package (Otley & Berry, 1980; Otley,
Th e th ird and h i ghe st form o f int e rd epen - 1980; Flamholtz, 1983). In this research we
dence is reciprocal. It is characterized by the propo s e to in ves ti gat e the w a y th r ee con trol
movement of work back and forth among de - sub-systems interrelate with one another as well
partments in reciprocal fashion. Organizational as the way the package fits the organizational set-
departments featuring reciprocal interdepen - ting.
dence typically work jointly on the same project
or customer. Services are customized and coor-
 MANAGEMENT CONTROL SYSTEMS 51

Operating budget formance, and feedback for correction. Manage -

The operating budget schedules and records ment control systems should theoretically assist
department revenue and expenditures for managers to perform the control cycle. Control
salaries, noncapital equipment and other operat - systems also help managers evaluate employees
ing expenses. The budget is considered a crucial and coordinate across departments. Depending
component of the control system package upon the emphasis given to a control system
(Horngren, 1982; Kaplan, 1982). Usually a de- within the organization, its scope and frequency
tailed operating budget is drawn up for the forth- may vary, and so may the emphasis given to each
coming year and then periodic budget reports control function, such as target setting, monitor-
are issued (usually monthly) to provide informa- ing, and coordination. Moreover, control sys-
tion to department managers and upper manage- tems have the potential to exert a motivational
ment about progress toward budget targets. force on employees who are affected by the sys -
tem (Hopwood, 1976; Brownell, 1982; Daroca,
Statistical reports 1984; Merchant, 1985). Management control
In addition to the budget, most organizations systems can vary in terms of many characteris -
rely on periodic operational reports that provide tics including: scope; reporting frequency; influ -
u pp e r ma n a ge me n t wi t h i n fo r ma t i o n on d e - ence in target setting; importance for key func-
partmental outputs and performance (Daft & t i on s (p l an n in g, c o or d in a ti o n a nd me a s u r e -
Macintosh, 1984). These reports are composed ment); emphasis, response to variances, and in -
of statistical data such as number of personnel, fluence on day-to-day operations. The three in-
number of new customer contracts, volume of t e r d e p e nd e n c i e s ( poo l e d, s e q u e n ti a l , r e c i p -
orders received and on hand, machinery down rocal) are expected to be related to the design
time and other statistics relevant to the depart - and use of budgets, statistical reports and SOPs
ment. Most of the data in these reports are non fi- along these dimensions. The overall research
nancial and are issued in weekly, monthly or model is illustrated in Table 1. The reasons for
quarterly reports. The specific content of these the hypothesized relationships are as follow s.
reports may differ across departments. These re- Under conditions of pooled interdependency,
ports normally originate in staff units other than organizations rely on standardization for con -
accounting or finance, such as MIS or computing trolling operating departments (Van de Ven et
services. al., 1976; Mintzberg, 1979). Standardization
makes possible the operation of the independent
Standard operating procedures and policies tasks over time by assuring that segments of the
The set of written rules, procedures, policies organization are operating in compatible ways.
and operating manuals (SOPs) are used to guide "It is in such situations that the bureaucratic
managers as they administer their departments. techniques of categorization and impersonal ap -
The SOPs also include general policy guidelines, plication of rules have been most beneficial"
job descriptions and prescriptions for how man - (Thompson, 1967, p. 17). Thus SOP s are ex -
agers should handle operational situations that pected to he a primary control medium when in-
might arise. terdependence is pooled and consequently the
scope, function and motivational impact of SOPs
are expected to be high. Most control and coor-
THE RESEARCH MODEL dination needs can be met through SOPs. Other
forms of control, such as budgets and statistical
The theoretical question to be addressed con - reports, are not expected to be important when
cerns the relationship between the design and i n t e rd e p e nd e n c e i s lo w. H e n c e bu d ge t s a nd
use of these management control systems and statistical reports are expected to receive less
t h e t yp e o f i n t e r d e p e n d e n c e a m o n g d e p a r t - emphasis than SOPs when organizational depart-
ments. Control is typically a three -stage cycle
that includes target setting, monitoring of per -
 52 N. B. MACINTOSH and R. L. DAFT

ments experience pooled interdependence. The greater coordination and control demand
These arguments are summarized in the follow- of sequential interdependence is expected to
ing hypotheses. mean that budgets and statistical reports will re-
ceive greater control emphasis than SOPs as a
Hypothesis 1 a. The use of SOPs for control will be posi-
mechanism of management control. SOPs pro -
tive ly correlated with the extent of pooled interdepen -
dence a mo ng depa rtme nts. vide standardization when interdependence is
low, but they are not expected to be emphasized
Hypothesis lb. The use of operating budget and statisti - or to provide sufficient control when inter -
cal reports for control will he negative ly correlated with
dependence is sequential because each depart -
t he e x te n t o f p o o le d in t e rd e p e n d e nc e a mo n g d e p a rt -
ments.
ment in the chain is performing a different task.
These relationships are reflected in the follow -
For sequential interdependence, performance ing hypotheses.
in a department depends on the work of other
Hypothesis 2a. The use of the Operating budget and statis -
departments in the chain. Planning and schedul - tical reports for control will he positively correlated with
ing are critical to ensure that departments pro - the extent of sequential interdependence among depart -
vide necessary resources for other departments ments.
in the work flow. Measurement of output is im-
Hypothesis 2b. The use of SOPs for control will he nega -
portant so that management can monitor
tive ly corre lated w it h t he e xte nt o f seque nt ia l inter -
whether activities are on schedule and can re - d e p e nd e nc e a mo n g d e p a rt me nt s .
spond to any exceptions or deviations that arise.
Th e p r e s s u r e t o r u n i n t e r d e p e n d e n t d e p a r t - When departments experience reciprocal in -
ments without interruption can lead to a strong terdependence, organizational performance is
control mentality from the top to coordinate dif- based on the ability to fuse diverse departments
feren ces amon g dep artments (Mintzherg, into a joint effort. Coordination and control
1979 ). The primary control systems used by come from rapid mutual adjustment and face-to-
managers under these conditions are expected face communication among departments, and
to be the budget and statistical reports. The feedback from the customer or client
budget provides tor the planning and scheduling (Thompson, 1967). The transformation process
of resources into each department and statistical is difficult to quantify, and performance tends to
reports measure and monitor outputs from each he based on whether the joint activities produce
department. the best outcome rather than on schedules or ef-

TABLE I. Prop osed rela tio ns hip bet ween depart me nt inte rdepe nde nc y a nd use o f ma na ge me nt contro l s yste ms

fIypothesized relationship of
Departmental Preferred means interdepe nde nc y w it h use
inte rdep e nde nc y of control of control syste m

Pooled Standardization ( + ) SOPs

( — ) Operating budget
( — ) Statistical reports
Sequential Pla nning a nd meas ure me nt ( — ) SOPs
( + ) Operating budget
( + ) Statistical reports
Reciprocal Mutual adjustment ( — ) SOPs
( — ) Operating budget
( — ) Statistical reports

( + ) positive relationship.
( — ) negative relationship.
 MANAGEMENT CONTROL SYSTEMS 53

ficiency. Control systems such as standardiz - In one company, for example, plant, marketing,
ation, rules, procedures, planning, budgets and personnel and engineering departments were
formal reports are expected to be less important selected. The final sample included 90 major de-
for control when interdependence is very high. partments from twenty organizations in five sec -
The impersonal control of SOPs and the data in tors of the economy. The organizations were lo -
the budget and statistical reports tend not to cated in the U.S.A. and Canada. Details of the
capture the dynamic nature of mutual adjust - sample are given in Table 2.
ment (Hayes, 1977). Professional norms, super -
vi s i o n a n d o t h e r fo r m s o f p e r s o n a l , d e c e n - Data collection
tralized control become more important under The primary method of data collection con -
conditions of reciprocal interdependence. Thus sisted of a personal interview with each depart-
the scope, function and motivational impact of ment manager. A preliminary interview format
for mal mana ge ment control s ys te ms such as had been pre-tested on managers in several or -
SOPs, budget and statistical reports are expected ganizations, after which the final questionnaire
to be low under conditions of reciprocal inter - was developed. Personal interviews were used
dependence. because they eliminated some of the distortions
ascribed to mail questionnaires when used to
Hypothesis 3. The use of SOPs, budget, and statistical re- measure organizational phenomena.
ports for control will be negatively correlated with the
The interviews were conducted in the offices
extent of reciprocal interdependence among depart -
ments. of the department managers. They were asked to
pull out the actual budget and statistical reports
under investigation, and these documents re -
ma i n e d i n fr o nt o f t h e ma n a ge r s a s th e y r e -
RESEARCH METHOD sponded to the interview questions. The SOPs
were also on hand, usually in the manager's of -
Sample fice or nearby.
The main criterion used to guide sampl e Two types of measurements were obtained.
selection was the need to include a wide range of First, actual physical counts were taken of the
departmental tasks to have variation in inter - number of books, pages and lines of standard
dependencies. To accomplish this, twenty five operating procedures and practices. The second
firms representing each industrial and commer - measurement involved asking managers for their
cial category in a directory of industrial and perception of how the controls were used, such
commer cial enterprise s we re sele cted at ran - as for motivational impact. The managers were
dom and contacted to see if they would partici- asked, for example, to rate the level of difficulty
pate in the study. Organizations from the service of budget targets, using a nine point scale rang -
and public sectors were also included to ensure ing from "very easy to achieve" to "almost im-
that the sample represented a cross section from possible to achieve". For the interdependence
both the private and public sectors. variables the managers were shown diagrams
Eighty percent of the organizations contacted (Appendix A) and asked for their perception of
agreed to participate in the study. During the the work flow betweeen their department and
preliminary visits, the control systems in each o t h e r d ep a r t me n t s . Th i s in s t r u me n t wa s d e -
organization were discussed with the corporate ve lop ed and vali da ted b y V an de V en et a l.
controller or his counterpart. Within each or - (1976).
ganization every attempt was made to obtain a
diverse cross-section of departments. The main
criteria for inclusion in the sample was that the FINDINGS
department managers had a clearly defined re -
sponsibility for meeting their operating budget. The statistical tests of the hypotheses about
 54 N. B. MACINTOSH and R. L. DAFT

TABLE 2. Sample composition

Sector Main activities of each organization

1. Manufacturing a. Machinery (6)

b. Electronics (6)
c. Wood products ( 3)
d . Textiles (4)
e . Oil, gas, and petroc he mica ls ( 3)
f . W i n e (4 )
g. Spirits ( 7)
h . T e l e p h o n e a n d t e l e c o m m u n i c a t i o n s e q u i p me n t ( 5 )
2. Merchandising a. Large, general department store chain ( 7)
b . Sp e c i a l t y d e p a rt me n t s t o r e c ha i n ( 3 )
c . Clothing retail chain (3)
3. Consumer marketing a. Food products (5)
b. Personal, food, and grocery products (4)
4. Service a. Advertising (3 )
b. Te lep ho ne a nd te lec o mmunica t io ns (6 )
c. Bank ( 3)
d. Fina nce (2 )
e . Trust and banking (4)
5. Health care a. Hospital (9 )
b. Faculty of medicine (3)

Figures in parentheses represent the number of departments samp led in each organization.

the design and use of control system and the ex- tial literature indicates that size is associated
tent of interdependence are shown in Tables 3,4 with greater use of bureaucratic rules and other
and 5. The tables report partial correlation coef- fo r m s o f i mp e r s o n a l c o n tr o l ( Ch i ld , 19 72;
ficients that control for department size. The de- Khandwalla, 1974; Bruns & Waterhouse, 1975;
partments varied widely by size, and a substan - Waterhouse & Tiessen, 1979; Merchant, 1981;
TABLE 3. Partial correlations (controlling for size) of standard operating procedures and departmenta l interdependence

Int e rd e p e nd e nc ie s
Pooled Sequential Reciprocal
Characteristics of the standard
operating p roced ures
0.33*
1 . N u mb e r o f b o o k s — 0.22t
2 . N u mb e r o f p a g e s 0.31* — 0.30*
3 . Percentage of departmental work covered 0.29t — 0.241-
4 . Pe rc e nta ge o f ti me ne c e s s a ry to 0.22 — 0.21t
follow SOP's to do work well
5 . Ad he re nc e to SO P's us e d to
eva luate perfor ma nce
6 . In flue nc e o f SO P 's o n d e p a rt me nt 0.27t — 0.32*
activities and operations

* p < 0.01.
< 0.05.
— Not significant.
 MANAGEMENT CONTROL SYSTEMS 55

TABLE 4. Partial correlations (controlling for size) of budget characteristics and departmental interdependence

Interdependencies
Budget characteristics Pooled Sequential Reciprocal

Frequency 0.221 -
Target difficulty - 0 .2 6 f
Influence in target setting
Upper management - 0.23f - 0 .2 3 t
Department manager - - - 0.32*
Department employees - - 0.18*
Importance for
Planning -
Coordination - 0.18* -
Measure and monitor - 0.18$ 0.24f - 0.23t
Emphasis on meeting targets - 0.25t - 0.19*
Response to negative variances - - -
0.
Influence on daily activities - 0 .25 f 0.29t

* p < 0.01. p <

0.05. p < 0.10. -
Not significant.

TABLE 5. Partial correlations (controlling for size) of statistical reports characteristics and departmental interdependence

Interdependencies
Statistical reports characteristics Pooled Sequential Reciprocal

Frequency -0.33* - 0 .22 f

Target difficulty - 0.37* 0.33t -
Influence in target setting
Upper management - 0.19f -
Department manager - 0.33* - 0.21*
Department employees - 0.28* - 0.22*
Importance for
Planning - - 0.36*
Coordination - 0.28* - 0.2I*
Measure and monitor - 0.29* - - 0.29*
Emphasis on meeting targets - 0.25$ -
Response to negative variances - - -
Influence on daily activities - 0 .22 f 0.20$ -

* p < 0.01. t p <

0.05. $ p < 0.10.
- Not significant.

Kimberly, 1983; Daft, 1986). Otley (1978) effect of size should provide a better test of
found that operating unit size had a signi ficant whethe r s yste matic di ffer ences in interdepen -
impa ct on budget behavior. Controlling for the dence are relat ed to co ntrol syst e m scope and
56 N. B. MACINTOSH and R. L. DAFT

use. The overall tendency of controlling for size Hypothesis 3 stated that managers in reciproc-
is to weaken the correlations slightly, although a ally interdependent departments will rely less
few small correlations for the budget and statisti- on the three formal control systems than mana -
cal reports disappeared. gers in departments with low interdependence.
The findings in Table 3 support Hypothesis 1 a The data generally support this hypothesis for
that the use of SOPs will have a positive correla- the SOPs and the budget. Five of the six SOP
tion with pooled interdependence. Five of the characteristics (Table 3, column 3) and four
six SOP characteristics are significantly as - budget characteristics (Table 4, column 3) are
sociated with pooled interdependence, includ - negativel y associat ed with reciprocal inter -
ing the quantity and coverage of SOPs, the neces- d ep enden c e. S OP s and bud get s thu s ar e r e -
sity to follow them to do the job well, and SOP ported to be used less as the level of interdepen -
influence on daily operations. dence increases.
Table 3 also indicates that the extent of se- Contrary to Hypothesis 3 some characteristics
quential interdependence has no relationship of the statistical reports are positively associated
with standard operating procedures, but that the with reciprocal interdependence (Table 5, col -
e xt e n t o f r e c ip r o c a l in t e rd e p e nd e n c e h a s a umn 3). Statistical reports appear to be used
negative relationship. These findings are consis - more for planning and coordination when inter -
tent with the theoretical argument that as the
dependence is high. Department managers and
level of interdependence increases, the size and
lower level employees also have more influence
use of SOPs as a primary control device will de-
in target setting. Thus SOPs and budgets are used
cline.
less under conditions of reciprocal interdepen -
The results in the first column of Tables 4 and dence while the statistical reports seem to play
5 provide modest support for Hypothesis 1 b that an expanded role in planning, target setting and
the operating budget and the statistical reports coordination. Planning and target setting also
will be negatively correlated with pooled inter - seem to entail more participation from lower
dependencies. Four budget and seven statistical level employees.
r e p o r t s ch a r a c t e r i s t i c s we r e n e ga t i ve l y a s -
s o c i a t e d wi t h poo l e d in t er d e p e nd e n c e . Th e
overall finding is that the use of SOPs appears to DISCUSSION AND IMPLICATIONS
he an important control for departments having
pooled interdependencies while the budget and Th e r e s e a r c h r e p o r t e d h e r e ga t h e r e d d a t a
the statistical reports are less so. fro m 90 departments in 20 organizations to
Hypothesis 2a proposed that under condi - examine the relationship between departmental
tions of sequential interdependence organiza - i n t e r d e p e n d e n c e a n d c o n t r o l s ys t e m s . T h e
tions will emphasize the operating budget and hypothesized pattern was that increasing levels
statistical reports for control. The results in the of interdependence would be associated with
second column of Tables 4 and 5 provide some differential use of SOPs, the budget and statisti -
support for this relationship. Five budget charac- cal reports. All three control systems were used
teristics and four statistical report characteris - in the organizations, but the scope and extent of
tics were positively associated with sequential use did differ by extent of departmental inter -
interdependence. In general, for sequential in - dependence.
terdependency it appears that top management Under conditions of pooled interdependence,
have a lot of in fluence in setting targets, and the organizations in the sample seemed to rely
these controls have motivational force on more on SOPs and less on budget and statistical
employees. This is consistent with Mintzberg's reports. A requirement of pooled interdepen -
( 1979 ) idea that a strong top-down control men- dence is that departments operate in compatible
tality prevails in sequentially interdependent de- ways, but frequent adjustments are not needed,
partments. For example, Daft (1986 ) cites the McDonald's
 MANAGEMENT CONTROL SYSTEMS 57

fast-food chain as an example of pooled inter - cipro cal interdependence dominates ( Ha yes,
dependence where each component used pro - 1977). Our findings indirectly support Govin -
cedures and reports as a way of standardizing darajan's (1984) finding that business units fac -
outcomes. ing higher environmental uncertainty rely more
For sequentially interdependent departments on subjective performance evaluations than on
the findings indicated that managers tended to formula-based ones. A basic ten et of manage-
use budgets and statistical reports more than ment accounting is that managers should be held
SOPs. The sequential linkage among depart - accountable for aspects of performance, such as
ments appears to require more emphasis on re - costs, over which they have control. Yet this ten-
sources, targets, scheduling, monitoring and dency may not fit reciprocal interdependence.
feedback, which are accomplished with budgets The task of each department is highly dependent
on the on-going work of several other depart -
and periodic operational reports. Management
ments (Hayes, 1977). Formal reporting systems
accounting systems and statistical reports are
may be used for planning, but they are not em -
well-suited to measuring the performance of de-
phasized for current detailed data for coordinat -
partments when internal variables are the major
ing or measuring the effectiveness of highly in -
explanators of effectiveness (Hayes, 1977), and
terdependent groups (H ayes, 1977). Although
in large, process do minated organizations
we did not test mutual adjustment directly, the
(Bruns & Waterhouse, 1975).
requisite coordination and control seems to be
The findings also indicated that under condi - achieved through personal interaction, frequent
tions of reciprocal interdependence the budgets communication, and mutual adjustment by the
and SOPs were used less than when interdepen- various managers and employees involved (Van
dence was low. Reciprocal interdependence in - de Ven et al., 1976; Daft & Macintosh, 1978;
creases uncertainty, and managers used statisti - Cheng, 1983).
cal reports for planning and coordination. Statis -
In conclusion, we want to address the ques -
tical reports probably were used less for measur -
tions raised at the beginning of this paper — do
ing and monitoring because work flow is hard to
m a n a ge m e n t c o n t r o l s ys t e m s v a r y w i t h d e -
quantify and measure under reciprocal inter -
partmental interdependence, and do accounting
dependence. Management accounting tools and
systems fit within a larger organizational control
other traditional controls seem to lose their pri -
package? Watson & Baumler ( 1975), Bruns &
macy under the uncertainty and rapid adaption
W a t e r h o u s e ( 1 9 7 5 ) , H a ye s ( 1 9 7 7 ) , G i n z b e r g
needed for reciprocal coordination. The find -
( 1980), and Otley (1980) proposed that de -
in gs ar e con si st ent with M int zbe r g's (1979)
partmental interdependence may be a design
suggestion that the need for quick response and
parameter that influences the use of accounting
sophisticated innovation, with different depart -
and information systems as integrating and con -
ments joining forces around specific projects,
trol devices. The findings reported in this paper
means the organization cannot rely on rules,
support this line of thinking and indicate how
standardization and all the regular bureaucratic
each of three control systems are used depend -
trappings. Expert knowledge, mutual adjust -
ing upon the level of departmental interdepen -
ment, and full -time project managers replace dence.
formal reporting systems, SOPs, and emphasis on
One interpretation of the findings is that the
hierarchical arrangements (Thompson, 1967).
role of each control system reflects a fit between
Formal controls are still used, of course, but
the need for information created by interdepen -
more o ften for help with planning and coor -
dence and the supply of information provided by
dinating by lower managers than for top level
the control system. SOPs are a standing body of
performance monitoring.
kno wl ed ge th at i s approp ria t e for sp e ci fyi n g
Reciprocal interdependence appears to pose
the greatest challenge to formal management
control systems. Financial data and budgets ap
parently do not play as large a role for depart -
ments such as R&D and Marketing in which re-
58 N. B. MACINTOSH and R. L. DAFT

standard behaviours across relatively stable and motivational force. Moreover, department man -
ind ep end ent d ep ar t men ts . S equ ent i al in t er - agers reported a high degree of satisfaction with
d ep enden c e c r e at es a n e ed fo r mo re d at a to all three control systems.
schedule, plan, and monitor the flow of material One realistic implication of the control pack -
and activities between departments. Budgeting age concept is that accounting systems design -
and statistical reports, which can provide data ers may want to consider the management ac -
on a monthly, weekly, or even a daily cycle, pro- counting system as part of a larger organizational
vide data that are more current and more relev - c o n t r o l s ys t e m . Th e p r e s e n c e o f S O P s a n d
ant than SOPs for the short time horizon needed periodic statistical reports may influence how
for coordination. When interdependence the accounting system is used. Accountants tend
among departments is reciprocal, the informa - to focus on the financial aspects of the organiza-
tion requirements begin to outstrip the data tion, and pay less attention to other control in -
supplied by formal control systems. These sys - formation needed by managers. If management
tems may be used for planning and scheduling, accountants are aware of the unique control
but special emphasis is given to face -to-face problems posed by organization characteristics
coordination and mutual adjustment. Since man- such as pooled, sequential, and especially recip-
agers cannot predict in advance the problems rocal interdependence, they will be in a position
that may arise and the information required, the to provide leadership in the selection of the re -
data contained in formal reports will not cover quisite control framework.
all problems, and hence will receive less em- In conclusion, this research attempted to in -
phasis than in the case of pooled and sequential crease understanding of control with in the or-
interdependence. Thus as interdependence in - ganizations by examining a package of control
creases, data are needed that are current, timely, systems and comparing their use to departmen -
and pertain to unpredictable events. These data tal interdependency (Otley, 1980). Previous
are supplied in turn by SOPs, budgets, statistical studies have found that forces such as environ -
reports and by direct managerial involvement in mental uncertainty, technology, business
face-to-face coordination. strategy and decentralization are related to over-
Th e qu e stion o f an o r gani z at iona l con tro l all structure and control processes. Interdepen -
package also seems to have received a partial dence among departments seems related to the
answer from our analysis of the data. Although use of control systems at the department level.
we can't say whether managers consciously de- Taken together the accumulated findings
signed each control system to play a different suggest an agenda for future research: begin to
role, we did find that control systems were used combine separate models into a larger theory,
in different yet complementary ways. SOPs were and undertake new empirical studies that integ-
used to direct behaviour for stable, independent rate several organizational variables and man -
departments, while budgets and statistical re - agement control systems into a unified frame -
ports were used to plan, monitor, and correct ac- work. Although a single study cannot assess all
tivities associated with resource inputs into de - variables, efforts to integrate past empirical find-
partments and outputs from those departments. ings may provide both new theoretical knowl -
Although we focused on only three control sys - edge and normative applications for control sys -
tems, we found that all three were rela ted to tem design.
management functions, and had a good deal of

BIBLIOGRAPHY

Birnberg, J. G., Turopolec, L. & Young, S. M., The Organizat ional Context of Accounting, Accounting,
O r g a n iz a tio n s a n d S oc ie ty (1 9 8 3 ) pp. 1 1 1 -1 2 9 .
 MANAGEMENT CONTROL SYSTEMS 59

Brownell, P., Participation in the Budgeting Process: When it Works and When it Doesn't, Journal of
Accounting Literature (1982) pp. 124 - 153.
Bruns, J. C. & Waterhouse, J. I - I., Budgeting, Control and Organizational Structure,Journal of Accounting
Research (Autu mn 1 97 5 ) p p. 1 7 7 -2 03.
Burchell, S., Clubb, C., Hopwood, A. G., Hughes, J. & Nahapiet, J., The Role of Accounting in Organizations
and Soc ie ty, Accounting Organizations and Society (1980) pp. 5 - 27.
Chenhall, R. H. & Morris, D., The Impact of Structure, Environment, and Interdependence on the Perceived
Usefulness of Ma na ge me nt Acco unting Syste ms, The Accounting Review (Janua ry 1986 ) pp. 58 -75.
Cheng, J. L., Interdependence and Coordination in Organiza tions: A Role-System Analysis, Academy of
Manag emen tJou ma l ( 1 9 8 3 ) p p . 1 5 6 - 1 6 2 .
Child, J., Organization Structure and Strategies of Control: A Replication of the Aston Study, Administrative
Science Quarterly ( 1972 ) pp. 163 - 177.
Daft, R. L., Organizational Theory and Design (Saint Paul, MN: West Publishing, 1986).
Daft, R. L. & Macintosh, N. B., A New Approach to Design and Use of Management Information, California
Management Review (1978) pp. 8 2 - 92.
Daft, R. L. & Macintosh, N. B., A Tentative Exploration into the Amount and Equivocality of Informat ion
Processing in Organizational Work Units, Administrative Science Quarterly (1981) pp. 207-224.
Daft, R. L. & Macintosh, N. B., The Nature and Use of Formal Control Systems for Management Control and
Strategy imp le mentatio n, Jounia/ of Management (1984) pp. 43-66.
Daroca, F. P., Informational Influences on Group Decision Making in a Participative Bud geting Context,
Accounting Organizations and Society (1984) pp. 13 - 32.
Duncan, R. B., Characteristics of Organizational Environment and Preceived Environmental Uncertainty,
Administrative Science Quarterly (1972) pp. 313 - 327.
Emma nue l, C. & Otle y, D., Accounting for Management Control (New York : Va n Nostra nd Re inho ld,
1985).
Ewusi-Mensah, K., The External Organizational Environment and Its Impact on Management Information
Syste ms, Accounting Organizations and Society (1981) pp. 301 - 316.
Flamholtz, E. G., Accounting, Budgeting and Control Systems in Their Organizational Context, Accounting
Organizations and Society (1983) pp. 153 - 169.
Fry, L. W., Tec h no lo g y -Structure R esea rc h: Thre e C rit ica l Iss ues, Academy of Management Journal
(1 9 8 2 ), p p . 5 3 5 -5 2 2 .
Ginzberg, M. J., An organizational Contingencies View of Accounting and Information Systems Implement -
atio n, Accounting Organizations and Society (1980) pp. 369 - 382.
Gordon, L. A. & Miller, D., A Contingency Framework for the Design of Accounting Informati on Systems,
Accounting Organizations and Society (1976) pp. 59 - 69.
Gordon, L. A. & Narayanan, V. K., Management Accounting Systems, Perceived Environmental Uncertainty
a nd O r ga n iz a t io na l S tr u c t u re : A n E mp i r i c a l I n v e s t i g a t i o n , Acco u n tin g Or ga ni z at io ns a nd S o ci et y
( 1984) pp. 33-47.
Govindarajan, V., Appropriateness of Accounting Data in Performance Evaluation: An Empirical Examin ation
of Environmental Uncertainty as an Intervening Variable, Accounting Organizations and Society ( 1 9 8 4 ) p p .
125-135.
Govindarajan, V. & Gupta, A. K., Linking Control Systems to Business Unit Strategy: Impact on Performance,
Accounting Organizations and Society (1985) pp. 5 1 - 5 6.
Hayes, D. C., The Contingency Theory of Management Accounting, The Accounting Review (January 1977 )
pp. 22-39.
Hopwood, A. G., On Trying to Study Accounting in the Contexts in which it Operates, Accounting Organ-
izations and Society ( 1983) pp. 287 - 305.
Ho pw oo d, A. G., Accounting and Human Behavior (Engle w oo d C l iffs, NJ: Pre nt ic e-Ha l l 1 97 6 ).
Horngren, C. T., Cost Accounting: A Managerial Emphasis 5th edn. (Englewood Cliff's', NJ: Prentice -Hall
1982).
Kaplan, R. S., The Evolution of Management Accounting, The Accounting Review (July 1984 ) pp. 390-418.
Kap la n, R. S., Advanced Management Accounting (En gle w oo d C li ffs, NJ : Pre nt ic e -Ha l l, 1 98 2 ).
K ha nd w a lla , P. N ., M as s O utp ut O rie nta t io n o f O p e ra t io ns Te c h no lo g y a nd O r ga n iz a t io na l St r uc t ure ,
Administrative Science Quarterly (March 1974) pp. 74 - 97.
Kilmann, R. H., The Costs of Organization Structure: Dispelling the Myths of Independent Divisions and
Orga nizatio n-W ide Dec is io n Mak ing, Accounting Organizations and Society (1983 ) pp. 341-357.
Kimberly, J. R., Orga nizatio na l Siz e and t he Struc t ura list P e rspect ive : A Re view, Crit iq ue a nd Proposa l,
60 N. B. MACINTOSH and R. L. DAFT

Administrative Science Quarterly (1983) pp. 571 - 597.

Lawler, E. E, III, Control Systems in Organizations, in Dunnette, M. D. (ed.), Handbook of Industrial and
Organizational Psychology, (Skokie, IL: Rand McNally, 1976).
McCann, J. E. & Ferry, D. L., An Approach For Assessing and Managing Inter -unit Interdependencies,
Academy of Management Review (1979) pp. 113 - 119.
McCann, J. E. & Selsky, J., Hyperturbulence and the Emergence of Type 5 Environments, Academy of
Management Review (1984), pp. 4 60 - 4 7 0.
Macintosh, N. B., The Social Software of Accounting and Information Systems (Chichester: John Wiley,
1985).
Merchant, K. A., The Design of the Corporate Budgeting System: Influences on Managerial Behavior and
Perfo rma nce, The Accounting Review (October 1981) pp. 813 -82 9.
Merchant, K. A., Budgeting and the Propensity to Create Budgetary Slack, Accounting Organizations and
Society (1985 ) pp. 201 - 210.
Mintzherg, H., The Structuring of Organizations (Englewood Cliffs, NJ: Prentice -Hall 1979).
O tle y, D . T. , Bud ge t U s e a nd M a na ge r ia l Pe r fo r ma nc e j o u r na / of Account ing Res earch (Sp ri n g 1 9 7 8 )
p p . 1 2 2 -1 4 9 .
Otley, D. T., The Contingency Theory of Management Accounting, Accounting Organizations and Society
(1980) pp. 4 13-428.
Otley, D. T. & Berry, A. J., Control, Organisation and Accounting, Accounting Organizations and Society
(1980 ) p p . 2 3 1 -2 4 4 .
Tho mpson," D., Organizations in Action (New York: McGraw -Hill 1967).
Tushman, M. L., Special Boundary Roles in the Innovation Process, Administrative Science Quarterly
D e c e mb e r 1 9 7 7 ) p p . 5 8 7 -6 0 5 .
Van dc Ven, A. H., Delbecq, A. L. & Koenig, R. Jr., Determinants of Coordination Modes Within Organiz -
ations, American Sociological Review (1976) pp. 322 - 338.
Waterhouse, J. H. & Tiessen, P., A Contingency Framework for Management Accounting Systems Resear ch,
Accounting Organizations and Society (1979 ) pp. 65 - 76.
Watson, D. J. H. & Baumler, J. V., Transfer Pricing: A Behavioral Context, The Accounting Review (July
1 9 7 5 ) p p . 4 6 ( -4 7 4.
Young, 1). W., Administrative Theory and Administrative Systems: A Synthesis Among Diverging Fields of
Inq uiry, Accounting Organizations and Society (1979) pp. 235 - 244.
 MANAGEMENT CONTROL SYSTEMS 6i

APPENDIX A. INTERDEPENDENCIES QUESTIONNAIRE FORM

How much work normally flows between your

I. Independent work flow unit and other units in this manner ?
Almost About Almost
Work enters organization
t
Some A lo
none half all
I 2 3 4 5 6 7 8 9

Your unit Work leaves organization

••
2. Sequential work flow
Work enters
I 2 3 4 5 6 7 8 9

Work leaves
3. Reciprocal work flow
Work enters
I 2 3 4 5 6 7 8 9

Work leaves