The (Changing) Role and Function of Dutch Controllers: a Survey Study

(1)

The (Changing) Role and Function of Dutch

Controllers: a Survey Study

University of Groningen

Faculty of Economics and Business

Accountancy & Controlling

Specialization: Controlling

June 2012

Wouter Moossdorff

1934465

Supervisor: Prof. dr. D.M. Swagerman

(2)

ii

Preface

The finalization of my research on the role and function of Dutch controllers also means completing my study on Controlling. The purpose of my research was to contribute to the current knowledge about this role and function by giving new insights in the organization and interpretation of the controller's job by suggesting a distinction between this role and function. Previous research studied the role and function as one common part of the job. Based on indications in literature, I chose to study them separately. This concept resulted in unique results that might tell us that a distinction between the role and function can and should be made.

My interest in conducting research increased during the writing of this thesis. In particular, there was one person that really opened my eyes and made me interested in research; my supervisor Dirk Swagerman. He inspired me every time we met and my head was a pleasant mess once I left his office. His open-minded approach has led to many new directions within the research, which were useful to increase the quality of the writings. I would like to thank you for that, Dirk, and I would be honored to work with you again in the future.

Another person I would like to thank is Paula van Veen-Dirks. She was, in first instance, voluntarily involved in my research project. She also wrote several articles on the role and function of Dutch controllers and this made her interested in my research. Our conversations were mainly focused on the general design of the research, but she gave me valuable feedback on the statistical side of the research. I would like to thank you for this.

My research has a survey-based character and therefore I would like to thank all the participants that reserved some time to fill in my questionnaire. Without your cooperation, I would never have been able to complete my study on the role and function of Dutch controllers. The participants that were interested have been thanked by a copy of either the thesis, a summary of the results or both. Next to this, I would like to thank the VRC, and in particular Mr. Smit and Mrs. Herklots, for their collaboration in contacting their members.

Now that I have completed my Controlling study, I want to explore my interest in research by applying for the Research Master at the University of Groningen. This master was recommended by both Dirk Swagerman and Paula van Veen-Dirks and I would like to thank them for that and for their efforts in promoting me within this master by, for example, writing a recommendation letter.

Enjoy reading my thesis!

Yours sincerely,

(3)

iii

Abstract

This research studied the role and function of the Dutch controller. The goal is to examine whether or not a distinction between those two parts of the controller's position should be made. Previous studies aimed mainly at the function side of the controller's job (i.e. the activities the controller performs in). In this research both sides, as well as the function side as the role side (i.e. how controllers relate to other officers in the organization), were initially studied separately. Later on, both studies were combined to examine the extent in which the role and function correspond. In order to do so, data was gathered using a survey. Controllers were grouped for their role part, a role classification, and for their function part, a function classification. In both classifications controllers were described as bean-counters or business-partners. The final step was to compare both classifications. The results on these analyses showed that most controllers were categorized as business-partners in the role classification, while business-partner oriented and bean-counter oriented activities were evenly balanced in the function classification. Subsequently, by performing a comparison analysis between the role and the function classification, evidence was found that a distinction between the role and function should be made. This analysis showed a lot of 'mismatches' in which a bean-counter role classification corresponds with a business-partner function classification and the other way around. Most mismatches were found in the category bean-counter function / business-partner role. These findings differ from earlier research to this subject in such a way that previous research did not suggest such a distinction. Future research should help to find more evidence on the presence of this distinction. For practitioners it is very important to pay attention to the organization and interpretation of the controller's job, because they should meet the organizations' needs.

Keywords

(4)

iv

1. Introduction

1.1 Background

More than half a century ago, Van der Schroeff (1953) introduced an officer named 'controller' in the Netherlands. In his advocacy, Van der Schroeff (1953) pleaded for an important organizational role for the controller. It was supposed to be an officer who is financially high educated and someone who could handle organizational problems. Besides that, verbal skills were also important, because controllers were considered to be team players. Van der Schroeff (1953) ascribed four key tasks of the controller, based on the Controller Institute of America: the construction and organization of the administration (1), to manage the administration and to produce figures, to interpret and analyze data (3) and to provide data and to advise (4). The first three tasks were typical tasks for a head administrator or chief accountant. The fourth task, advising, however, is clearly a different task and was seen as a task that distinguished the controller from a head of administration (Kuiper, 1969).

Exactly 25 years earlier, Kester (1928) highlighted the importance of the controller. In his paper, Kester (1928) distinguishes two different kinds of controllers, that of a 'glorified bookkeeper' and that of a 'real accountant'. The bookkeeping role consists of the recording and reporting of the accomplishment within each of the operating departments of the business and judging the efficiency of those accomplishments (i.e. the monitoring role of the controller). Besides these working activities, the controller has

to report to operating executives by using (non-)personalized management reports. Kester (1928) allocates an important role for the controller in the interpretation process of these reports, because he/she is in a better position to do so. The controller should discuss the significance of the reports with the operating executives. With his/her knowledge, the controller can often express an opinion about the best way to use the information. This makes the controller a real counselor to all operating executives.

Van der Schroeffs (1953) and Kesters (1928) description of the controller have a lot in common. Both mentioned the importance of the advising role and being a partner to executives. However, Van der Schroeff (1953) only noticed a couple of controller tasks, where Kester (1928) also described two different 'roles' of the controller, which led to different (types of) controllers. Both 'roles' are associated with

different aspects of the overall

controllerstask. A glorified bookkeeper is more focused on recording and reporting data, this can be compared with the head administrator role mentioned by Kuiper (1969), and the real accountant was seen as an officer who really worked with, and not only for, the executives. An advising and interpreting role arose.

At this point a distinction can be made between the role and the function of controllers. Kester (1928) already mentioned two roles and the different tasks that are performed within those roles, form the function of controllers. Riedijk et al. (2002) define the role of controllers as the way in which the controller operates towards other members of the organization

(6)

2

(i.e. how do controllers relate to other officers). The function of controllers is defined as the scope of the controller's tasks (i.e. in what kind of activities does the controller perform). Besides a distinction in the role and function, literature often uses different definitions for an officer that is responsible for (managerial) accounting and the wide communication of management accounting information around the organization (Granlund and Lukka, 1997). Management accountant is often used as a synonym of controller for similar officers in different parts of the world (read; the United States of America). Therefore, both terms are used as synonyms in my research. The first section of chapter two describes the

positioning of the controller in

management accounting.

After the foundations of the controller's position were laid by Kester (1928) and Van der Schroeff (1953), the role and function of controllers received a great deal of attention in research, as well as in the Netherlands as in other parts of the world. For example Hopper (1980) and Sathe (1983). Hopper (1980) studied the conflicts between two contrasting roles (i.e. the book-keeper and the service-aid role) of management accountants. His research proves that the book-keeping role was seen as less important than the service role and accountants would prefer greater involvement in management decision making teams. Sathe (1983) also makes this distinction in his four ideal types of controllers. He also noted the increasing importance for controllers to be more involved in the business decision making process. Both roles (i.e. the book-keeper and the service-aid role) are comparable

with Kesters (1928) description of 'the glorified bookkeeper' and 'the real accountant'.

Research in these changing role of controllers have intensively increased since the publication of 'Relevance Lost' by Johnson and Kaplan (1987). They stated that management accountants were failing to provide business managers with the information they demanded - management accounting was in 'crisis' (Burns and Vaivio, 2001).

This publication made academics,

accountants in business and consultants to develop new and so-called 'advanced' management accounting techniques and new management accounting systems - a management accounting that meets the information requirements of business managers in today’s global, technology driven world (Burns and Vaivio, 2001). As a consequence, the (changing) role and function of controllers and/or management accountants has received a lot of attention in research (Burns, Scapens and Turley, 1996; Cooper, 1996a; Granlund and Lukka, 1997, 1998a, 1998b; Burns and Vaivio, 2001; Malmi et al., 2001; Riedijk et al., 2002; Kocks, 2003; Burns and Baldvinsdottir, 2005; Järvenpää, 2007; Maas and Matějka, 2009). These studies all notice or focus on the increasing importance of the controller's participation in the management decision making processes. The aforementioned two roles (i.e. the book-keeping role and the service-aid role) are most frequently studied in research, also under different descriptions like 'bean-counter' to describe the book-keeping role and 'business-partner' to describe the service-aid role. Both roles

(7)

3

have been studied independently and in a process in which the book-keeping role develops or transforms into a service-aid role. Besides attention for the role of management accountants, the function was also studied broadly. For example Malmi et al. (2001) who studied the most common work activities of management accountants in Finland. They hoped to find a shift from a focus on 'book-keeping activities' to a focus on 'business-partner activities', a shift that eventually was not found.

Research on the role and function of Dutch controllers consists of studies conducted by

Leeuwen and Wemmenhove (2001),

Riedijk et al. (2002), Verstegen et al. (2007) and De Loo et al. (2011). Riedijk et al. (2002) mentioned the lack of practical foundation of theoretical developments of the role and function of controllers. Verstegen et al. (2007) developed a new

concept in management accounting

research to the role and function of controllers. They stated that previous studies used very different definitions of controllers, controller roles and triggers. As a consequence, as they say, it is difficult to get a coherent, complete picture of the controller's profession. Their

research aimed to produce a

comprehensive conceptual framework that is empirically informed. This framework ultimately leads to a classification of controllers. De Loo et al. (2011) used the

same framework to check if this

classification might have changed during 2004-2007 and what triggers may have caused this change.

1.2 Problem and purpose

As Burns and Vaivio (2001) describe, new management accounting techniques and

systems are needed to meet the information requirements of business managers in today's global, technology driven world. Besides this, Granlund and Lukka (1997) and Järvenpää (2001) give several key drivers of the increased

business-orientation amongst management

accountants. Among them are changing business market conditions, organizational re-design, new managerial philosophies, increased business complexity, systems

development, management technique

innovations and human resource

developments.

Moreover, during the past few years, a lot has changed. The world was in a financial crisis which affected most organizations.

Through global and/or European

harmonization the world is getting smaller and therefore more complex. The annual conference of Dutch Controllers (see also www.controlling.nl), held early 2011, focused on this increasing complexity. The increasing uncertainty in the world was the central theme during this day. This uncertainty asks for less predicting and more anticipating in order to react more responsively and to face the challenges and threats that affect a company. As an organization, and controller, these developments and/or uncertainties have to be faced and an appropriate response is required.

These changing conditions might have changed the role and function of controllers. As the latest study on this field dates from 2007 (De Loo et al, 2011), it is interesting to study the role and function of Dutch controllers from 2007 to this day and to find out if the role and function of Dutch controllers actually have changed

(8)

4

since 2007. The focus within my research will be on the distinction between the two 'roles' of bean-counter and business-partner. Compared to previous research (Verstegen et al, 2007; De Loo et al, 2011), my study will use an approach that differs from the framework that was developed by Verstegen et al. (2007). Verstegen et al. (2007) grouped controllers based on their work activities, which led to two groups of controllers, and subsequently studied to what extent certain variables, triggers, were significant predictors of this group membership. My research, however, does not relate triggers to groups of controllers, based on their activities, but uses an approach which groups controllers twice, based on different activities and triggers compared to the ones Verstegen et al (2007) used. These groupings result in two classifications: a function classification in which controllers are grouped based on their activities (1) and a role classification in which controllers are grouped based on their role characteristics (i.e. the role-based triggers). These role-based triggers are all characteristics of a bean-counter and a business-partner in which the one extreme of this trigger ('the left-side') applies to the bean-counter role and the other ('the right-side') to the business-partner role. Once

controllers have been grouped into the two groups, both groups will be compared in order to find out if there is a difference between the role classification and the function classification. Within this research, the organization of the controller's position corresponds with the role of the controller and the interpretation with the function of the controller.

In summary, the purpose of my research is to study whether there has to be made a distinction between the role and the function by, in first instance, studying both parts of the controller's position separately and to combine them later to see if there are similarities or differences between the two classifications.

1.3 Structure

Chapter two discusses the theories on the role and function of controllers in further detail. Next, chapter three elaborates the research design, in which it deals with the research structure, the research method and the research population and sample. Subsequently, chapter four describes the results and a discussion is presented in chapter five. Finally, conclusions are drawn and limitations of this research and suggestions for further research are given in chapter six.

(9)

5

2. Theory

This chapter discusses theories on the role and function of the controller, based on literature on the management accounting profession, in greater detail. Subsection 2.1 starts with the positioning of the controller within management accounting. Next, subsection 2.2 aims at studies on the role of the controller and focuses on the book-keeping and the service-aid role of the controller and the transformation towards this service-aid role. The function of the controller is examined in subsection 2.3.

2.1 The position of the controller

No clear cut definition of a controller can be given. Granlund and Lukka (1997) define a controller as a person who is responsible for management accounting

and the wide communication of

management accounting information

around the organization. This description has a lot of similarities with the role and function of management accountants. Therefore both terms are used as synonyms.

Several forms of controllers can be distinguished. Swagerman (2009) speaks of business controllers that are focused on numbers and of controllers that are more focused on the social aspects of the job, the 'social controller'. This dichotomy reflects the distinction between the bean-counter and the business-partner. Bean-counters can best be associated with a head of administration and/or chief accountant (Van der Schroeff, 1953). A chief accountant's most important task is to construct, organize and manage the administration and to produce figures, to

interpret and analyze data. A typical task that distinguishes the controller from the chief accountant is to provide this data to

management and to advise the

management on how to use this data.

Although no clear description of the controller's job can be given, it is generally known that the controller is the person who is responsible for planning and control within an organization. The controller distinguishes itself from other officers, like chief accountants, by the relation it has

with other organization members.

Controllers have, compared to 'general'

accountants, more contact with

management and are often part of the management team. Their main focus lies not on making the numbers, but to interpret and to provide numbers to other members of the management team and to give (financial) advice. The controller is the financial expert within this team, rather than just a person who knows how to record numbers.

This advising role towards the

management, and therefore the

participation in the decision process, is what it makes a controller to be an unique officer within the organization. The controller is a true business partner with a (financial) focus on the entire organization instead of only on the finance and control process (Jarvenpää, 2007).

2.2 The role of the controller

Some of the studies mentioned in the introduction section are here described in further detail. Firstly, several roles of controllers, like the book-keeping role and the service-aid role, and the transformation of those roles are explained. Secondly, some complications in these roles are

(10)

6

described. Thirdly and lastly, research that studies the role of the controller in practice is examined.

Controllers' and/or management

accountants' role in organizations has been differently categorized during the years. Hopper (1980) studied the book-keeper and the service-aid role. He describes, by using Argyris' (1953) description, the keeper role as follows: the

book-keeper is concerned with the

implementation and administration of financial systems to enable superiors to specify and measure the performance of subordinates. In this, budgets were used as performance measures and the accountant reports deviations directly up the

hierarchy. Interaction with

line-management was very rare, conflict-laden and defensive. Such a role is often associated with the centralization of accountants in a functional hierarchy physically separate from the line (i.e. the accounting function is more like a staff-department). The service-aid role, on the other hand, identifies, according to Hopper (1980), lower and middle management as the major clients of accountants, and satisfying such managers' information needs for self control as the principal task of accountants. This role is characterized

by personalized and horizontal

communication, education, identification and analyses of problems with peer and subordinate managers and the provision of a broad range of information. In this role, the controller has more interaction with

management and is no longer a 'servant' of his superiors, but, in fact, part of the ongoing organizational processes and a cooperative partner.

The distinction between the book-keeper role and the service-aid role is also made by Granlund and Lukka (1997) and Sathe (1983). However, the first two researchers define those roles as bean-counters and change-agents and Sathe (1983) defines them into two responsibilities (i.e. the financial reporting and internal control responsibility and the management-service responsibility). Granlund and Lukka (1997) performed a study on the transition of management accountants' role from bean-counters to change agents in Finland. They outlined this transition in a figure in which the management accountant's job description expanded from a historian to a member of the management team/ change agent (see figure 1). During their research, several interviewees gave their opinion about this (changing) role. The roles differ, for example, in the way the controller communicates. In the local part of the line organization the controller is the business oriented member of the managerial team and a financial adviser and in the accounting organization the controller is more like the local guardian, ensuring that the corporate interests are not forgotten. Among the interviewees it was felt that the role of financial monitoring (i.e. the watchdog role) is still important, next to the increased business orientation.

(11)

7

Figure 1: derived from Granlund and Lukka (1997)

The historian and watchdog role (i.e. the bean-counter type) aim to write accurate and correct financial history and thereby attempt to respond to the formal information needs. Communication is typically directed into the accounting function only and communication to other

departments of the organization

concentrates on written reports. The modern business oriented management accountant, or controller, is required to have a good understanding of the business the firm is in, fluent communication skills and an ability to run change projects. 'The orientation of their actions in terms of time is more towards the future than to the past and the role played by the detailed accuracy of the figures they deal with is far from the significance this issue has for the

traditional bean-counter type of

accountants' (Granlund and Lukka, 1997). At last, the controller becomes a change agent when he or she is concerned with bringing his or her special knowledge to

real decision-making situations and feels responsibility for getting through the message based on this knowledge.

This transition, however, as drawn in figure 1, has to be read carefully. It does not mean that one develops from one role to another and abandons the other. Granlund and Lukka (1997) state that certain kind of historiography, or at least watch over, prevails as the basis of all accounting, regardless of what other roles may in practice be built on these.

Besides Granlund and Lukka's (1997) distinction of the controller's role, Sathe (1983) defines the controller's role in management into two responsibilities. Firstly, the responsibility to help the management team in the business

decision-making process. Secondly, the

responsibility to insure that reported financial information pertaining to the relevant organizational unit is accurate and that internal control practice is in

Historian Watchdog Consultant Adviser

Member of the management team Change agent

(12)

8

conformity with the corporate policy and procedures. The former could also be seen as the management-service responsibility (i.e. a business-oriented role) and the latter as the financial reporting and internal control responsibility (i.e. a watchdog role). These two responsibilities are interconnected in a way that making quality information and analyses available gives no assurance that this information is effectively used, or is used at all. Staff executives, like controllers, have to become involved in the business decision-making process to help the operating executives to get the utmost benefit out of the information processed.

The question of Sathe's (1983) paper is:

can the controller carry both

responsibilities effectively? The answer depends, according to Sathe (1983), on whether one is talking about general tendencies or what a particular controller can accomplish. A common view at this point is that both responsibilities are mutually exclusive and emphasis at one responsibility decreases the effective performance of the other one. However, the phenomenon of a 'strong controller' also exists. This person can retain both responsibilities and this ability depends on the personal qualities of the particular individual. Based on the previous, Sathe (1983) distinguishes four ideal types of controllers. Firstly, the involved controller. This functionary's primary emphasis lies on the management-service responsibility and the desired behavior for the controller is to be actively involved in the business decision-making process. The independent controller, secondly, emphasizes on the financial reporting and internal control responsibilities and the desired behavior

for the controller is to retain objectivity and independence in dealing with affiliated management. Thirdly, the split controller. This type of controller only fulfills one of the previously noted responsibility and therefore two individuals have to be appointed to cover both responsibilities. At last, the strong controller. This controller retains both responsibilities and strong emphasis is placed on both.

According to Sathe's (1983) classification, a controller is, by means of a strong controller, able to effectively fulfill both responsibilities. Granlund and Lukka (1997), however, state that both types of controllers (i.e. the bean-counter and the business-oriented) fundamentally differ. In fact, both roles differ to such an extent that it is somewhat difficult to imagine that somebody would manage to act in both roles, at least not simultaneously. In their study, Maas and Matějka (2009) also distinguish two responsibilities of business

unit controllers. The functional

responsibility means that controllers have to report fairly and objectively on the economic situation of their unit to facilitate corporate control. The local responsibility means that controllers are part of local management teams and have to provide the

business unit management with

information to facilitate strategic and operational decision-making (i.e. the distinction between the financial reporting and internal control and the management-service responsibility according to Sathe (1983)). Maas and Matějka (2009) note that role conflict and role ambiguity arises due to simultaneous occurrence of two or more sets of pressures (i.e. the two responsibilities) and, as such, both responsibilities cannot be accommodated

(13)

9

at one person. Hopper (1980) also note that putting effort into one responsibility is at the expense of the other responsibility.

To conclude the previous, controllers usually perform two roles. The first role is the role of a book-keeper whose focus lies on financial reporting and internal control. The second role is that of a business-oriented controller. His/her focus lies on management support in decision making processes.

These aforementioned studies on the two different roles were mainly theoretical in nature. Burns has, in cooperation with other researchers, performed several case studies in the field of the changing role of

controllers in which the two

responsibilities and/or roles were the central objects of research (Burns, Scapens and Turley, 1996; Burns and Vaivio, 2001 and Burns and Baldvinsdottir, 2005). Burns and Baldvinsdottir (2005) studied a manufacturing division of a multinational pharmaceuticals organization. In this organization they witnessed a changing role of (management) accountants towards a more supportive and advising role. The term 'hybrid accountants' was used to define the new role of accountants. Hybrid accountants arose because business managers were performing most clerical accounting and, in this case study, operating managers were more empowered in terms of designing and managing their own budgets and forecasts. Previously, those activities were largely done by management accountants, but in this new business the 'new' hybrid accountants have a more supportive and advising role. A few finance managers in this case study underlined this changing responsibility.

They said that the finance function has to change whenever the business structure changes and the accountant should interact more heavily with operation managers about their specific information needs. If the operation manager says 'this is what I want', the accountant will answer 'Right, this is how we can get it from the system'. As such, the accountant has a very

important supporting role towards

operating managers and this relationship is valued by both parties, as evidenced by this case study. Another typical citation in Burns' and Baldvinsdottirs (2005) paper is the comment of one finance manager about the managers talking about commercial and financial implications of any decision. 'This was not the case four or five years ago', according to the finance manager.

So, in the case study described above, management accountants and operating managers experienced a change in their role within an organization. Management accountants' focus on actual financial issues is declining and their management-support role is increasing. Burns and Vaivio (2001) also noticed this change. Routine accounting tasks (e.g. transaction processing and statutory reporting) will continue to demand a small (and declining) number of 'specialist' accountants who can rely on the technology to do the bulk of the work. Other management accountants are more likely to take on a role as internal 'business consultants' (i.e. a supportive and advising accountant). Evidence of this increased focus on the supportive and advising role is also found in case- and field studies conducted by Burns, Scapens and Turley (1996). In the firms studied, the accounting function was fulfilling a more supportive and advising role in providing

(14)

10

the information business managers need to

make their decisions. The actual

accessibility to information is, due to the

impact of information technology,

decentralized to workers, like business or operating managers, at all levels. Managers have on-line access to the information they need to control all aspects, including the costs, of their operations. Because of this increased access, managers are able to produce their own budgets and cost reports. Management accountants, who were previously located in a central accounting function and performing tasks like producing those budgets, are now dispersed to the operating levels to give direct assistance and advice to managers. As operating managers are more involved in producing accounting information, accountants need to understand the processes of operations management and to interact more closely with the day-to-day activities of the firm. They are providing an essential link between the operating level and, for instance, the management board.

Besides research in foreign countries, a couple of studies on the role and function of Dutch controllers were performed (Riedijk et al., 2001; Verstegen et al., 2007; De Loo et al. 2011 and Van Veen-Dirks and De Loo, 2011). Riedijk et al. (2001) quoted some research that studied this role and function in practice. Two studies explored the requirements of controllers based on job advertisements (Van Helden, 1998 and Van Helden and Rutten, 2000). Those studies showed that the traditional work activities, like managing the planning and control

function and providing management

information, were still important and they

concluded that there was no need for a wide controller, or a strong controller according to Sathe (1983), (yet). Another study quoted by Riedijk et al. (2001) was an interview-based research by Van der Meer-Kooistra (1998). She concluded that the role and function of controllers differed because of organizational and personal characteristics. Despite of those studies, the role and function of Dutch controllers in practice still did not have had a lot of attention in research, according to Riedijk et al. (2001). In order to start research within a new terrain, Riedijk et al.'s (2001) study was a first attempt of practical research to this role and function. They solely focused on the role of the controller in the process of planning and control and they studied the difference of this role among different kinds of organization types. They used Mintzbergs (2009) theory on organization structures to define a theoretical based expectation of the expected role of the controller. From this starting point, they interviewed ten controllers of ten different Dutch organizations, based on four structures of Mintzberg (2009), to study what role these controllers have in practice. They found different roles for controllers in organizations with a simple, a machine bureaucracy, a professional bureaucracy and a division structure. In their concluding remarks, Riedijk et al. (2001) finished with a call for more research on the role and function of Dutch controllers. Verstegen et al. (2007) answered this call by introducing a new comprehensive framework in order to study the role and function of controllers in a more complete and coherent way. They found, based on their classification, two groups of controllers. The one group of controllers

(15)

11

was concerned with traditional activities or beancounter-type of activities (45% of the questioned controllers) and the other group (55%) with the business-oriented activities. Next to this, they found six triggers that significantly predicted group membership. Examples of triggers are, amongst others, years of experience and the size of an organization. Van Veen-Dirks and De Loo (2011) and De Loo et al. (2011) rerun this study by using the same framework and therefore the same activities and triggers. It was important to do so, otherwise no comparison during time was possible. De Loo et al. (2011) studied the year of 2007 and found some differences compared to the research of 2004 by Verstegen et al. (2007). De Loo et al. (2011) also distinguished two groups of controllers based on a cluster analysis, but those were different from those found by Verstegen et al. (2007) in a way that both groups merged somewhat. The triggers that predicted group membership also changed partly compared to the earlier study. The ICT-development of an organization was not seen as a significant predictor anymore, but new predictors were, amongst others, the financial status of a organization and whether someone is an executor or a thinker. Subsequently, Van Veen-Dirks

and De Loo (2011) studied the

development of the role and function of Dutch controllers for 2010 under the title 'Is de 'strong controller' in aantocht?' ('Is the 'strong controller' coming'?). They did not find clear results to be able to find an answer to this question, but they did find an increased focus on traditional management accounting techniques and a decreased importance of strategic oriented activities because of the influence of the financial crisis. The financial crisis seems

to have had influence on the importance of the different activities within the daily function of the controller.

2.3 The function of the controller

Willson and Colford (1991) divide the basic functional responsibilities and activities into five categories: planning, control, reporting, accounting and other primary responsibilities. Firstly, planning. A controller should formulate and maintain an integrated plan of operation consistent with the company's goals and objectives. This plan has to be accepted and

communicated to all levels of

management. In this process, the controller is counselor and coordinator. The second activity, control, focuses on the development and revision of standards which are used to measure performance. The controller should provide functional executives with information he or she is expected to use to achieve the required performance. Thirdly, the reporting responsibility consists of the preparation, analysis and interpretation of financial results for utilization by management in

the decision-making process, the

evaluation of data with reference to the company's objectives and the production of external reports for example regulatory bodies. The fourth activity is that of accounting. The controller designs, establishes and maintains general and cost accounting systems at all company levels to properly record all financial transactions in accordance with accounting principles with adequate internal control. Finally, the other primary responsibilities comprise of the supervision of the tax function, maintain an appropriate relationship with the independent auditor and for example

(16)

12

risk management, financial public relations and responsibility for information systems.

Willson and Colfords (1991) performed a theoretically based study. Malmi, Seppälä and Rantanen (2001) conducted a practical research on the function of management accountants in Finland. The aim of their study was to ascertain what managerial accounting tasks were most important among the respondents. In this, they asked their sample to estimate their most common working activities, if the working activities have changed since five years ago and whether it will change in the next three years. Besides this, the respondents were asked to determine the importance of various skills and abilities. The results of this study show that the most common work activities consist of, among others, budgeting and annual planning, financial reporting, development and planning of

accounting information systems,

development of accounting and

management systems and the preparation of ad hoc financial analyses. Four of these five common activities were also seen as the most important activities, where financial reporting and budgeting and annual planning are the most important ones. Financial reporting was seen as the activity that required the most extra effort compared to five years ago. Budgetary activities were seen as requiring less effort. These budgetary activities are also expected to decrease in importance in the next three years. Use and development of balanced scorecards and development of accounting and management systems are

expected to increase in importance in the next three years.

Although Malmi et al. (2001) were hoping to find a shift in activities, especially from bean counters to business partners, a clear shift was not found. However, they did discover some signs of the growing importance of business partnering.

De Loo et al. (2011) studied, as mentioned earlier, the role change of Dutch management accountants. They have clustered several work activities, performed by controllers, in groups which has ultimately led to different kinds of controllers, performing their own and

shared sets of activities. They

characterized the two groups of controllers as 'reporting business analysts' and 'business system analysts' in which both groups have a shared focus on internal analysis and risk management and their own focus on internal reporting and external reporting (reporting business analysts) and maintaining management control systems (business system analysts). Their research also shows that the reporting and scorekeeping activities are still very much alive, in contradiction to the opinions in literature that tell the controller/ management accountant is evolving into a business-oriented or strategy-focused one (Järvenpää, 2007; Russell et al., 1999). De Loo et al.'s (2011) results show that such a shift has not yet been completed. Van Veen-Dirks and De loo (2011) also found evidence that traditional activities are still important in today's controllers' function.

(17)

13

3. Research design

The research design is divided into three parts. Subsection 3.1 discusses the research structure, in which questions are posed and an overall structure is given. Next, the research method(s) is the focus of subsection two. The final subsection describes the population and sample of this research.

3.1 Research structure

Several studies on the role and function of controllers/ management accountants have been discussed in chapter two. A distinction was made between studies on the role and function of Dutch controllers and those that studied this role and function in foreign countries. Both groups of studies are used as a basis for the research structure.

The framework developed by Verstegen et al. (2007) forms the basis of my research. They have chosen several activities that represented the function of the controller. So called triggers were derived from literature and used as possible predictors of group membership of controllers. In my research the role and function of controllers is researched in a slightly different way. Other triggers are selected from the literature. These triggers are divided in three groups (see Appendix 1). The first group consists of business-related triggers, the second group of person-related triggers and the third group consists of triggers based on the distinction between the bean-counter role and the business-partner role (i.e. the role-based triggers). The business-related and person-related triggers are mainly selected in order

to describe the population and sample. The role-based triggers are all formulated in questions on a five-answer-scale with on both sides one extreme. The one extreme shows the characteristic of a bean-counter and the other extreme that of a business-partner. Those characteristics are based on several articles on these two roles. Besides other triggers, my research also uses different activities (see Appendix 2) to study the function of controllers. Those activities are mostly derived from Malmi et al.'s (2001) study on the role and function of Finnish controllers and for a minor part from IMA (1996) and Russell et al. (1999).

Both, triggers and activities, are selected using the distinction between the bean-counter and the business-partner role. These two roles therefore play a central role in my research. The ultimate goal is to either or not find a distinction between the role and the function of Dutch controllers (i.e. the role differs from the function). The role will be studied by grouping controllers based on the role-based triggers and the function by grouping them based on the activities. The main research questions can be described as:

1. Based on the triggers related to the bean-counter and the business-partner role, can Dutch controllers be seen as bean-counters or as business-partners or do they show characteristics of both roles?

2. Based on the activities related to the bean-counter and the business-partner function, to what extent can Dutch controllers be seen as bean-counters or as business-partners or do they perform in activities that

(18)

14

correspond with both functions?

The focus is on coherent

combinations of activities in order to be able to cluster them into groups of activities and to explore how the function of Dutch controllers can be described.

3. Do observed triggers and activities of bean-counters and business-partners correspond with each other or is there a distinction between the role and the actual function of the controller?

Analyzable data in order to draw conclusions on research question 2 is derived by examining the importance of the several activities in the controllers day-to-day's work. This question can be divided in three sub-questions:

1. Which work activities of Dutch controllers are most (or least important) at present? Results on this sub-question will be used to form groups of activities in which controllers perform.

2. Which work activities of Dutch

controllers increased (or

decreased) in importance compared to four years ago? This question is of importance to be able to compare the results with previous research.

3. Which work activities of Dutch controllers will increase (or decrease) in importance in the next three to five years? This question is included to explore a probable future change.

The main focus of my research will be on the present role and function of Dutch controllers. The three main research questions can be summarized in the research structure as shown in figure 2. The additional parts of my research consists of the earlier mentioned business-related and person-business-related triggers and the sub-questions 2 and 3 of the second research question. Those parts are not included in the research structure and are further discussed in the results and later on in the discussion in order to describe the population and sample and to be able to make a comparison with previous research.

Figure 2: Research structure

In advance, two hypotheses on the expected classifications are drawn based on literature. For the role classification the

expectation is that 'the Dutch controller is

a business-partner'. Since 2007, and the

financial crisis, the world became more

Bean-counter

/

(19)

15

complex and uncertain. One would say that an increased complexity and uncertainty claims a controller to interact more intensively with other organization members, especially managers, in order to observe challenges and treats adequately and to respond to them on a timely basis. Besides, a slight shift towards a business-partner role was already observed by Verstegen et al. (2007). On the function, Van Veen-Dirks and De Loo (2011)

concluded that Dutch controllers

performed more in traditional then in strategic-related activities compared to the 2007-research of De Loo et al. (2011). They have described this shift as an effect of the financial crisis. Although different activities are used in my research, the same results are expected to be found. So, the expectation to the function classification is that 'at present, traditional (bean-counter

like) activities are more important than business-partner like activities'. In conclusion, the following hypotheses are stated:

1. The Dutch controller is a business-partner (role classification).

2. Bean-counter like activities are more important than business-partner like activities (function classification).

3.2 Research method

Overview

Previous research on the role and function of Dutch controllers used different approaches. The first structured research method was developed by Verstegen et al. (2007) and was later on used in similar research conducted by De Loo et al. (2011) and Van Veen-Dirks and De Loo (2011).

In these studies factor and cluster analyses were used to group activities and to group controllers into those groups of activities. A regression analysis, in which triggers

were connected to the groups of

controllers, was used to predict the significance of triggers in estimating group membership. My research is largely based on this research method. It will, however, use a different approach to study the role of the controllers. As mentioned in previous sections, this research introduces a different view in the distinction between the role and the function by making a clear distinction between the role and the function of a controller. This distinction also leads to a clear distinction in the research methods that will be used to study both parts. The function part is, in accordance with previous research, also seen as the group of activities in which controllers perform. The role part, however, is in first instance seen as an independent research topic which will also be studied by using triggers. These triggers are not used to predict group membership, as was the case in previous research, but to actually examine the role of the controller and to make a distinction in the bean-counter role and the business-partner role. Later on, both results (i.e. the group of controllers based on activities and the group of controllers based on role-based triggers) are intertwined to explore if those two groupings correspond with each other or if there really is a distinction between the role and the function of a controller (see research question 3).

The function part will be studied by using the same methods as in the Verstegen et al. (2007) study. A factor analysis groups the different activities and reduces the total

(20)

16

amount of activities into a smaller amount of factors. A cluster analysis will group the respondents based on the correlations of each respondent with the different factors. The role part consists of questions on subjects that are directly linked with the bean-counter and the business-partner role. As mentioned earlier, each question addresses a different subject and uses a five-answer-scale with two extreme answers. Those extreme answers are typically characteristics of both roles. Differently from the function part, the role part is already structured in a way that makes a clear distinction between the roles. As a consequence, there is no need for any prior grouping by using a factor analyses. The factors are already known (i.e. the bean-counter and the business-partner). However, the respondents do have to be grouped. Since every question uses a 1-5 scale (excluding two questions that use a 1-2 scale) in which the lower or the more left

answer always corresponds with a

characteristic of a bean-counter and the higher or the more right answer with a

characteristic of a business-partner, the total amount of the score of the different answers compared to the minimum and maximum score can be used to group respondents. After these two classifications (i.e. the role classification and the function

classification) are made, both

classifications will be compared to explore if there is coherence between the two groupings. This ultimately should lead to an answer to the third research question.

The mentioned methods all correspond with the underlying main research questions. An overview of the research method is shown in figure 3. The next subsections are devoted to a wider description of the three research methods.

Figure 3: Research method

Factor analysis

The function part of the research, which will ultimately lead to a function classification as shown in figure 3, starts with a list of activities derived from the literature (see appendix 2). The activities are selected based on previous research on the distinction between the bean-counter and business-partner. Therefore, this list of

activities consists of bean-counter like activities and business-partner like activities. Following the instructions and guidelines of Hair et al. (1992), a factor analysis will be used to group those activities into factors. An oblique rotation technique will be used, because of the

presupposed interrelations between

activities. Appendix 4 shows more statistical details on the factor analysis.

(21)

17

Cluster analysis

After the activities are grouped into factors, the respondents will be grouped into different clusters by using a cluster analysis. Following Verstegen et al. (1997) and the suggestions of Hair et al. (1992), a hierarchical clustering algorithm called 'Ward's method' will be used. Appendix 4 shows more statistical details on the cluster analysis.

Total score analysis

The role-based triggers are derived from the literature and are related with characteristics of a bean-counter and a business-partner. Each question, excluding two questions, consists of a five-answer-scale in which the outer left (five-answer-scaled 1) corresponds with a bean-counter and the outer right (scaled 5) with a business-partner. The other two questions consists of a two-answer-scale and a 1 corresponds with a bean-counter and a 2 with a

business-partner. Because of this

consistency in the development of the questions, respondents can be grouped directly into a bean-counter group and a business-partner group. The higher the total score on the triggers, the more the respondent is seen as a business-partner. The lower the score, the more the respondent shows signs of a bean-counter. As shown in Appendix 1, fifteen triggers will be used to study the distinction between a bean-counter and a business-partner. The maximum score is therefore 69 (complete business-partner) and the minimum score 15 (complete bean-counter). The average score is 42. Based on the possible scores and the given maximum, minimum and average score, four groups are distinguished:

1. Group 1, ranging from 15 to 28

points: predominantly

bean-counter.

2. Group 2, ranging from 29 to 42 points: more bean-counter than business-partner.

3. Group 3, ranging from 43 to 56 points: more business-partner than bean-counter.

4. Group 4, ranging from 57 to 69 points: predominantly business-partner.

Respondents are grouped based on their total score on the fifteen triggers.

Comparison analysis

Respondents are grouped twice, once in the role classification and once in the function classification. Both groupings will be compared in the comparison analysis. It is tested if, for example, more respondents are a bean-counter in the role classification and a business-partner in the function classification. This analyses will help to answer the third research question and to draw conclusions on the posed distinction between the organization (i.e. the role) and the interpretation (i.e. the function) of the controller job.

3.3 Research population and sample

Dutch controllers are connected to the 'Vereniging van Registercontrollers (VRC) (see also www.vrc.nl), an association for registercontrollers. 'Registercontrollers (RC's)' are chartered controllers that all have fulfilled an adequate and similar education. The accessibility of this group of controllers is quite manageable and these controllers are therefore chosen as

(22)

18

the population of this research. A senior co-worker of this association was contacted by a formal letter. Soon after the approval of cooperation, an email was send to all members of this association. In total, 3.000 controllers were contacted and they form the population for my research.

4. Results

Before the results on the research questions are described, the sample and general results on this sample and comparison with the population are described. Then, the results on the first research question, addressing the role classification, are examined. The next part contains the description of the results on the function classification (i.e. the second research question) and is divided in the results on the factor and the cluster analysis. Subsequently, both results are compared in the next part of this chapter, addressing the third and last research question. Additional results on the future and past importance of the activities (i.e. sub-questions 2 and 3) are defined in the for last section. The chapter ends with a concluding figure and description of the most important results.

4.1 Sample description

The respondents were asked to fill in a questionnaire, which consisted of two parts. The first part focused on general questions, concerning the business-related, the person-related and the role-related triggers, and the second part asked questions about the importance of the included activities. Both parts were completed differently, as to how many respondents per part. The first part was completed by 166 respondents and 118

respondents fully completed the

questionnaire. Both figures are already corrected for incomplete responses and thus represent the usable responses. Therefore, using the mentioned population of 3.000 controllers in chapter 3, the response rate for the first part is approximately 5.5% and for the second part and the total questionnaire 4%.

4.2 General results

The following results are based on the 166 respondents that completed the first part of the questionnaire. This part consisted of several general questions aimed at

collecting information about the

organization the respondent works for and more person-related questions. The average organization of the respondents is quite complex (3.87 on a scale of 1-5) which operates in a dynamic market (3.83 on a scale of 1-5). The organizations were rather big with 66.3% of the respondents working for an organization with 500 full time equivalents (fte) or more. The core business of the respondents organizations was categorized in a mixed way. The most

important core businesses of the

respondents' organizations were

manufacturing/ industrial (28.3%), service (other than financials) (24.1%), financial services (22.3%) and public sector/ government (18.1%). Trade (6.0%) and agricultural and extractive (1.2%) were the least represented categories of industries. Nearly half of the organizations have a divisionalized structure (46.4%). Other most prominent structures were the professional bureaucracy (24.1%) and the

machine bureaucracy (13.3%).

Management accountants and/or

controllers in these organizations were mostly located in the accounting & finance

(23)

19

department (71.7%) and only occasionally situated at other departments (10.2%).

Of all respondents 86.7% are male and 13.3% female. The average age is around 41 years and the average experience in a financial function is around 16 years. Most of the respondents are located at the accounting & finance department (62.7%) and only a minor part in a different department (13.3%). The respondents were asked to indicate how they see their own role and how other managers, other than financials, see their role. Respondents see themselves more as a member of the management team (65.7% of the times) than other, non-financial managers do (58.4% of the times).

For generalization purposes, a comparison of the sample to the population can be made on a few perspectives. Of the population 17% is female and 83% is male, compared to 13% and 87% for the sample. The average age of all male controllers is 42.5 years and 39.4 years for female controllers, while this is respectively 41.5 years and 40.5 years for the sample. The average experience of the population is approximately 10.5 years for men and 10 years for women, while the average experience of the sample is approximately 16 years for men and 14 years for women. Based on this comparison, the sample

fairly represents the population.

Unfortunately, comparison with other data (e.g the core business of the organizations and the size in fte) was not possible, because this information was not available for the population.

4.3 Results on the role classification

A total of 166 respondents were grouped into the four groups by using a total-score analysis on the fifteen triggers that were used to categorize the respondents' role. After the analysis, just one respondent can be grouped in group 1 (i.e. predominantly bean-counter), 28 respondents in group 2 (i.e. more bean-counter than business-partner), 116 in group 3 (i.e. more business-partner than bean-counter) and 21 can be seen, based on their total scores, as business-partners (i.e. are members of group 4). Based on this total score analysis on the role of the respondents, 82.6% of

the respondents (137 out of 166

respondents) can be categorized as business-partners and 17.4% (29 out of 166 respondents) as bean-counters.

In order to be able to make a comparison of this role classification with the upcoming function classification, the same sample should be used. 118 respondents also completed the function part of the

questionnaire and of those 118

respondents, one respondent was grouped in group 1 and 23, 81 and 13 respondents were grouped in respectively group 2, 3 and 4. So, 79.7% of the respondents (94 out of 118 respondents) can be categorized as business-partners and 20.3% (24 out of 118 respondents) as bean-counters. Compared to the classification of the total 166 respondents, the group of 118 respondents contains relatively more bean-counters (i.e. 20.3% compared to 17.4%).

4.4 Results on the function classification

For statistical details of the derivation of the following results, the interested reader

(24)

20

is referred to Appendix 4. The factor analysis that was performed, based on the second research question, to group statistical related activities, resulted in a distinction of nine factors (see Appendix 5). Each factor received a label that, to the opinion of the author, best described the underlying activities that formed the factor. As a consequence, the following factors were distinguished:

1. Managing and planning the finance & control function.

2. Financial reporting and budgeting. 3. Development of accounting

information systems.

4. Advanced financial reporting. 5. Internal support and control. 6. Internal auditing

7. External accounting, auditing and control

8. Strategic planning and risk management.

9. Corporate restructuring

Factors 1, 2, 3, 5 and 6 are more related to bean-counter like activities, while factors

4, 7, 8 and 9 represent business-partner like activities. The cluster analysis performed using each factor as a variable, resulted in two clusters that consist of the two mentioned groups of factors. Grouping respondents into the two resulting clusters has led to 68 members of cluster 1 (factors 4, 7, 8 and 9) and 50 members of cluster 2 (factors 1, 2, 3, 5 and 6). Based on a factor and cluster analysis on the working activities of controllers (i.e. the function of the controller), 57.5% of the sample is a business-partner and 42.5% a bean-counter, according to their function classification.

4.5 Results on the comparison analysis

The role and function classification of the 118 respondents that completed both parts of the questionnaire, were compared to determine the percentages of bean-counters and business-partners. Figure 4 shows a crosstabulation with the function groupings as the columns and the role groupings as the rows. Comparison Analysis (n = 118) Function classification Total Bean-counter (1) Business-partner (2) Role classification Bean-counter (1) 1 0 1 Bean-counter (2) 20 3 23 Business-partner (3) 43 38 81 Business-partner (4) 4 9 13 Total 68 50 118

Figure 4: Crosstabulation of the function and role classification

As mentioned in the previous sections, function classification 1 corresponds with factors that relate to bean-counter like

activities, while function classification 2 corresponds with business-partner like activities. Role classifications 1 and 2

The (Changing) Role and Function of Dutch Controllers: a Survey Study