Successfully Implementing Activity-Based Costing

(1)

Successfully Implementing Activity-Based Costing

Triggers, Pitfalls and Remedies

Marc Moll

(2)

Title Page

Title: Su ccess fu ll y imp leme nt ing act ivit y-b ased co sting : tr igg ers, p itfalls and remed ies

Au tho r: Marc W. A. Mo ll

Pla ce: Gro ninge n

Da te: Au gu st 8 , 20 05

Supervisors:

(3)

Preface

The six month internship for Heineken in Kinshasa, Democratic Republic of Congo, has truly been an experience that will last a lifetime. The challenges that came with the assignment and the unique environment have enabled me to learn in many different ways.

One of the fundamental dimensions of the assignment was to actually be involved in the dynamics of working for a large international commercial organization for the first time in my life. This has been a valuable professional experience in that it was an opportunity to combine the theory of textbooks with down-to-earth management practice. Another fundamental dimension of the assignment was to design, execute and report an academic management research. In this context I would like to thank my

supervisors at the university of Groningen dr. E.P. Jansen and dr. P.E. Kamminga for their support and guidance.

Apart from these basic aspects, which are closely related to the assignment, the setting in the Democratic Republic of Congo brought along a number of additional challenges and made it all the more memorable. First, living in one of the world’s poorest countries with all its extremes has been an eye-opener and made a significant impact. Second, learning to know about the Congolese people, their habits, food and country has inspired me to see more of Africa and if possible Congo itself. Finally, the struggle to learn French in such a short notice has been a challenge in itself.

I would like to take the opportunity to thank Bralima’s general manager Hans van Mameren for granting me the opportunity to have this experience. Also, our talks about doing business in Africa were very much inspiring. Furthermore I would like to thank my “ABC-partner” Useni and my fellow interns Ton, Jorik and Thijs.

Last but not least I would like to thank my parents for their unconditional support throughout my study curriculum.

Marc Moll

Groningen, September 2005

(4)

Management summary

Background

This research thesis is the result of a six month internship at Bralima. Bralima is one of Heineken’s African brewing groups, which has its home base in Kinshasa, Democratic Republic of Congo. It bottles a number of brands of beer as well as a number of different soft drinks in six different production sites spread around the country.

In the second half of 2003 Bralima implemented an initial activity-based costing system to increase the quality of financial management information. This initiative was induced by a corporate wide effort by Heineken to implement activity-based costing (ABC) in order to “be able to evaluate from different angles and for better comparison between Operating Companies”. However, after a year there are doubts on the future role of ABC within Bralima, how to increase its utilization and whether it should be continued at all. These issues surrounding the implementation of activity-based costing at Bralima formed the starting point for this research.

Problem area

Activity-based costing was introduced in the late 1980’s to overcome the disadvantages of traditional cost accounting systems by providing the means to use additional causes of overhead costs. The general argument of the advocates of ABC is that it provides two types of benefits. First of all, it provides a more accurate method to trace overhead costs to cost objects (e.g. products, customers).

This results in more accurate and specific accounting information with regard to for example product unit costs and profitability. With more accurate output costs, managers can make better decisions about their outputs and the activities that produce these outputs. Second of all, using activities as the basis of resource consumption, ABC provides new insight in how (fixed) costs are generated. By knowing the determinants of these (fixed) costs, specific actions for process improvements can be taken accordingly. Sub-optimal utilization of ABC has a negative effect on the overall company performance due to missed benefits in these two areas. The research question is formulated as follows:

How can the utilization of activity-based costing be increased to its full potential in better supporting strategic decisions and process improvements made by senior and middle management of Bralima?

Research design

Activity-based costing and its implementation have received widespread attention in the management accounting literature; especially in the mid 1990’s. Many researchers emphasize the importance of behavioral and organizational factors next to the usual technological factors as determinants of successful implementation of ABC. In order to achieve a better utilization of ABC at Bralima this

(5)

research examined both these dimensions. The quality of the current ABC system is assessed in terms of structure, functionality and usability. Furthermore, the organizational embedding is viewed as the result of an accounting change process, which is influenced by barriers, advancing forces and procedural characteristics.

Main conclusions

The outcome of this research thesis can be summarized in five main conclusions. The first conclusion is that the implementation of ABC at Bralima is relevant because it has potential value adding capabilities. These can be found in the area of portfolio appraisal, cost reduction / process

improvements (TPM), capital investment decisions and pricing decisions. The second conclusion is that the implementation of ABC at Bralima is motivated by several internal and external conditions.

External motivators are the fiercely competitive market and the unstable economic conditions. Internal motivators are the strategic profitability focus, the advanced production technology and the portfolio management. The third conclusion is that utilization of the potential value adding capabilities of ABC is highly limited. The limitations can be categorized in terms of frequency, functionality, departments and management seniority. The fourth conclusion is that the contribution of the system quality to this limited utilization is mainly caused by the limited frequency of data updates, the inefficient

administration of information, the lack of action-based applications and the lack of linkage to other competitive strategies (e.g. TPM). The fifth conclusion is that the contribution of the organizational embedding to the limited utilization of ABC is mainly caused by ambiguous communication between Heineken corporate (MMO) and Bralima, cultural constraints in the African/Congolese context, the absence of a clear leadership role in the implementation process and the lack of linkage to

performance evaluation and compensation.

Recommendations

In conclusion of this research an implementation model to increase the utilization of ABC is developed. This model is based on six implementation steps:

Step 1: Vision, leadership and momentum. This initial implementation step characterizes the role of top management. This model distinguishes three functions that continue to play an important part throughout the implementation process: (1) creating a vision (2) establishing leadership (3) maintaining momentum.

Step 2: Cultural change. This second step is needed to tackle some of the cultural constraints within Bralima which seem to form a barrier for an effective use of ABC. Because this is such a complex and slow moving task it also continues throughout the implementation process.

(6)

Step 3: Teams. This third step is characterized by a team-based approach to develop action-based ABC applications. This approach has a number of advantages that should resolve some of the current problems with regard to ABC acceptance and use.

Step 4: ABC system and information administration. The team’s first task is to modify the ABC system to better serve the purpose of the applications for which the system is intended. The team’s second task is to dramatically increase the efficiency of the administration of information.

Step 5: Reporting. In this next step the system’s output is shaped for each specific ABC application.

The ABC reports are a key element of the ABC system and have an internal (Bralima) and external (Heineken) dimension.

Step 6: Performance evaluation and compensation. Because the applications are developed with a value adding intention they could be linked to specific ABC targets or objectives soon after people are satisfied and familiar with the new system. A logical next step is to link these ABC targets to the system of performance evaluation and compensation.

(7)

1 Introduction

This research thesis studies the implementation of activity-based costing at one of Heineken’s African brewing groups called Bralima. The study is based on a six-month internship for Bralima in Kinshasa, Democratic Republic of Congo.

1.1 Heineken

Heineken international is a multinational organization with its roots in Amsterdam, where Gerard Adriaan Heineken purchased a brewery in 1864. In the ensuing decades, under the leadership of three generations of the Heineken family, Heineken has grown into one of the world's leading brewing groups. Besides brewing and selling its own principal international brands (Heineken, Amstel) all over the world, it owns a great deal of local breweries called operating companies (OpCo’s). In a recent effort to optimize corporate performance, every OpCo will implement or has implemented a standardized activity-based costing system. One of Heineken’s African OpCo’s is Bralima, with its home base in Kinshasa, Democratic Republic of Congo.

1.2 Bralima

Founded in 1923 in Kinshasa, Bralima s.a.r.l. is one of Africa’s oldest breweries. Situated in the Democratic Republic of Congo, the Bralima Company produces and bottles a number of brands of beer (e.g. Primus, Mutzig, Turbo King and Maltina) as well as a number of different soft drinks (Coca Cola, Fanta, Sprite and Schweppes). The company owns five different brewing- and bottling facilities spread all over the country (Kinshasa, Boma, Lubumbashi, Bukavu and Kisangani). A sixth facility has recently been (re)opened in the city of Mbandaka and will soon be operational. It also owns its own bottle factory called Boukin, located in Kinshasa. In 2002, Bralima invested heavily in new and modern production equipment. As a result from continuing efforts in mainly technological

improvements, it received the ISO 9001 certification a year later.

1.3 The Market

The Democratic Republic of Congo (DRC), in which Bralima operates, has had a very turbulent history¹. Violent internal disputes, decades of economic mismanagement and a corrupt government

1 See Appendix A for a more detailed historical overview

(11)

has turned DRC into one of today’s poorest countries in the world with an estimated GNI² per capita of $100 (Atlas method³) and a severe debt burden.

However, recent efforts of current president Joseph Kabila and cooperative programs with various aid organizations show a steady recovering of Congo’s economy. Moreover, from the business perspective of Heineken, DRC still has a huge market potential; being the third largest population in sub-Saharan Africa. The beer and soft drink market is currently growing at a high pace. Latest estimate figures show a 30% increase in total sales volume compared to last year.

Within the beer and soft drink market Bralima has only one competitor: Bracongo. With its main brand Skol it still has a marginal market share advantage over Bralima in the beer market, although this advantage is declining. Bralima has managed to recover from a period in which it lost a significant amount of market share. Efforts to improve the brand image of Primus are starting to pay off, although a lot of work is still ahead. Furthermore, Bralima has a strong grip on the premium segment. Bracongo recently limited sales of its only premium brand Castel, whereas Bralima launched Heineken as a super premium brand to further strengthen its beer portfolio.

Soft Drinks Bralima Bracongo Coca Cola American Cola

Fanta Djino

Sprite Schweppes

Vital'O

Table 1-1 Comparison of brand portfolios

With regard to the soft drink market there is less competition at this moment compared to the beer market. With a solid market share of around 70% this remains the more lucrative of the two for Bralima.

1.4 Activity-based costing⁴

From 1925 until the mid 1980’s management accounting practice changed remarkably little (e.g.

Kaplan, 1984; Johnson, 1987). Traditional cost systems of this period use a two stage procedure to

2 Gross National Income; source: World Bank database of World Development Indicators (WDI)

3 The purpose of the Atlas conversion factor is to reduce the impact of exchange rate fluctuations in the cross- Beers

Bralima Bracongo Regular Primus Skol

Dark Turbo King Doppel Premium Mutzig Castel

Special Guinness

Malt Maltina

Super

Premium Heineken

(12)

assign an organization’s indirect and support expenses to outputs. Operating expenses are assigned first to cost pools and second, to the outputs of the production process (see Figure 1-1). These traditional two stage assignment procedures, however, distort reported costs considerably. The traditional systems assign costs from cost pools to outputs using volume drivers such as labor and machine hours, material purchases and units produced. Because many indirect and support resources are not used in proportion to the number of output units produced, these traditional systems provide highly inaccurate measures of the costs of support activities used by individual outputs.

Figure 1-1 The traditional two stage procedure

Activity-based cost systems differ from traditional systems by modeling the usage of all organizational resources on the activities performed by these resources and then linking the cost of these activities to outputs such as products, services, customers and projects (see Figure 1-2). In particular activity-based systems measure more accurately the cost of activities not proportional to the volume of output produced. Thus, activity-based systems differ from traditional systems in two ways. First, cost pools are defined as activities rather than as production cost centres. Second, the cost drivers used to assign activity costs to outputs are structurally different from those used in traditional cost systems. These modifications to the two-stage procedure allow well-designed activity-based cost systems to report more accurate costs than a traditional system because they identify clearly the costs of the different activities being performed in the organization and they assign the costs of these activities to outputs using measures that represent the types of demands that individual outputs make on those activities.

Figure 1-2 The activity-based two stage procedure

This introduction to the basic concepts of activity-based costing is extended in chapter 2.2, which provides a more detailed overview of the theoretical concepts related to this subject.

(13)

2 Research definition

2.1 Problem area

2.1.1 Activity-based costing at Bralima

Heineken has had experience in using activity-based costing (ABC) for several years now, but comparison of ABC figures between operating companies was not always possible because not all OpCo’s had adopted the system. Not too long ago Heineken Corporate decided to let all OpCo’s implement the activity-based costing system in order to “be able to evaluate from different angles and for better comparison between Operating Companies”.

Bralima, being one of those OpCo’s, is relatively late in adopting ABC methods. In the last half of 2003 a system was developed using Metify software. Yet, many activities and related cost drivers are standardized⁵, so the degrees of freedom in designing the system were limited. This is a big advantage if you consider the corporate objective. However, for Bralima this has limited the possibilities to adjust the system to local needs to a certain degree. Despite this disadvantage, Bralima still sees opportunities to improve their system within the boundaries set by Heineken Corporate. Some difficulties arose in determining cost centre activities and getting the right information for calculating some cost drivers. On top of this, not all sites could be included in the system in time. Most of these problems will be fixed in the 2004 update.

Currently the system has only been used to calculate end-of-year ABC results of 2003. This information has been used to give senior management insight in overall company performance, product profitability (per site), activity costs and profitability of the two main customer groups. One action result is that these figures have supported some strategic decisions (e.g. focus on marketing cost control, product pricing). It is not clear, however, if these decisions are directly initiated by ABC figures.

Although these first output figures of the ABC system have had a positive contribution to strategic decision making and have given senior management more accurate insight in product profitability, the utilization of all potential ABC information is limited in two ways.

First, its use is mainly limited to senior management. Middle managers hardly ever use ABC information for decision making or process improvements. The ABC results have not been clearly

(14)

communicated to middle managers or at least not specifically relevant for each cost centre. Whereas these figures could potentially be of great value for middle management when pursuing for example cost control or lean processes. This means that the insight that ABC can give, is not being used to its full potential.

Second, its use is limited to the functional area of finance. The financial department is the ‘owner’ of the system; meaning that non-financial managers do not have direct access to specific ABC data.

Furthermore, non-financial managers are not aware that ABC potentially holds the type of information they are looking for. This implies that the system is either not providing the information needed for middle management decisions (e.g. not enough detail or outdated) or the system is not (yet) truly embedded in the organization (i.e. regularly used in daily routines and decision making). Another issue is that the data in the system is currently being updated only once a year. The financial department reports the ABC results as a consequence also only once a year. Therefore, many non- financial managers view it as an interesting new financial toy, but don’t consider it an important tool to guide their own decisions. Some managers though, have had experience with ABC in other business units and do see the possible advantages. However, as the year comes closer to its end, the financial department is not able to provide up-to-date information (e.g. product profitability). This results in very little use of ABC information in daily practice.

2.1.2 How utilization of activity-based costing affects company performance

Although some researchers caution against new cost management techniques like ABC (e.g. Dolinsky and Vollmann, 1991; Johnson, 1992), the general consensus among most researchers is that activity- based costing is a more effective management accounting tool than traditional methods⁶ (e.g. Cooper, 1988; Cooper, 1994; Innes and Mitchell, 1995; Swenson, 1995). The general argument of these ABC advocates is that it provides two types of benefits. First of all, it provides a more accurate method to trace overhead costs to cost objects (e.g. products, customers). This results in more accurate and specific accounting information with regard to for example product unit costs and profitability. With more accurate output costs, managers can make better decisions about their outputs and the activities that produce these outputs (Cooper et al., 1992). Second of all, using activities as the basis of resource consumption, ABC provides new insight in how (fixed) costs are generated. By knowing the

determinants of these (fixed) costs, specific actions for process improvements can be taken accordingly⁷.

6 This argument is sometimes limited to specific business environments. Bralima’s business environment qualifies for the greater part to these limitations, thus it is assumed that this claim holds for Bralima.

7 Some researchers and practitioners refer to this broader application of ABC as ‘Activity-Based Management’

(ABM). This thesis, however, defines activity-based costing more broadly; including ABM applications.

(15)

Like many researchers and practitioners, the assumption in this thesis will be that activity-based costing potentially provides more accurate management accounting information and gives more insight in determinants of (fixed) costs within Bralima than the conventional volume-based method.

Two series of problems can be identified within Bralima which result from missing these two distinct advantages of activity-based costing. The first series of problems is presented in the figure below and is based on using less accurate conventional management accounting information (hl-based) instead of ABC information.

Figure 2-1 Problem series part I: using less accurate accounting methods

Use of less accurate accounting methods

Within the Heineken Company in general and Bralima in particular, a commonly used variable for financial planning and control purposes are the costs and proceeds per hectoliter (hl). This makes it easier to deal with for example pack types differences between brands or pack type changes over time of the same brand. By using financial data on a hectoliter basis, Bralima can evaluate and compare the financial performance of products, customers and the company as a whole. In order to include all costs in these evaluations, both variable and fixed costs⁸ are considered on a hectoliter basis. However, this volume-based approach might not always be the true determinant of all costs, especially overhead expenses. While the conventional hl-based approach has the advantage of simplicity it will result in significant miscostings where overheads are not volume driven (Innes and Mitchell, 1995; Cooper et al., 1992). Within the brewing and packaging process of Bralima many activities can be distinguished of which the amount of resources they consume is not proportional to the number of hectoliters brewed. Examples of these activities include: filtration, cooling, fermenting, lagering and conditioning⁹.

Little insight in product and customer profitability

The use of the less accurate conventional accounting methods can lead to less insight in product and customer profitability due to significant miscostings (Dearman and Shields, 2001). In order to overcome the disadvantages of the hl-method, the overhead costs are at the moment sometimes even disregarded in financial evaluations. This was especially noticeable in calculations for return on investment figures for certain projects (see Example 2.1). Although the outcome of the final decision

8 I use both the term fixed costs and overhead to refer to those expenses that are not directly linked to a product

(16)

might not have been affected by this inaccuracy, it clearly shows the difficulty Bralima has with allocating overhead costs to specific products in order to determine its unit costs.

Example 2.1 Maltina margins

Ineffective foundation of strategic and tactical decisions

The consequence of having little insight in product and customer profitability is that it mounts to ineffective financial foundations of certain strategic and tactical decisions. Within Bralima several examples of strategic decisions can be identified for which insight in product unit costs or profitability is relevant for a solid financial foundation. They categorically concern product pricing, sourcing and product portfolio decisions (see Example 2.2). The product unit costs become increasingly important in pricing decisions because of the pressure that high inflation and exchange rate risks put on the margins¹⁰. An example of a tactical decision is the weekly production planning. On many occasions production priorities have to be set on the basis of product profitability. In the month of December the highest sales volume is generated due to the many holidays like Christmas. This puts a lot of strain on the limited production capacity. To also make it the most profitable month of the year a good

knowledge of each product’s profitability is essential in some cases.

Ineffective financial foundations for important decisions do not necessarily imply that the decision itself is ineffective¹¹, but eventually it could well lead to irrational decisions caused by distorted views on financial implications of these decisions. Thus, it can be generally assumed that surely on the long term, ineffective financial foundation of (strategic) decisions has a negative effect on the overall company performance in terms of profit maximization.

10 The exchange rate of Congolese francs to US dollars increased from 395 FC/USD in September 2004 to over 470 FC/USD in February 2005 (19% in 6 months!). This puts a lot of pressure on the margins as Bralima has significant expenses in US dollars and practically all income is in FC

11 Some decisions can be so evident that even common sense can effectively validate the outcome.

(17)

Example 2.2 Introducing Mützig 65cl

The second series of problems is presented in the figure below, which depicts those problems that can be addressed by using broader applications of activity-based costing (sometimes referred to as activity-based management).

Figure 2-2 Problem series part II: little insight in determinants of fixed costs

Little insight in determinants of fixed costs

Within Bralima three main groups can be categorized as being the large consumers of resources regarded as overhead costs: production, marketing and staff¹². Each group faces different challenges in

(18)

trying to keep fixed costs at a minimum. The two groups that are best helped by ABC information in this context are production and marketing (Lewis, 1993).

Traditionally, the production department is very cost-oriented. The main objective for Bralima’s production managers is to brew and pack high quality beer at the lowest possible costs. In practice this means that they are most of the time occupied with process improvements, line efficiencies and production planning in order to keep non-value adding activities and quality disruptions at a minimum.

In order to achieve this Bralima uses a number of so called Key Performance Indicators (KPI’s), which are standardized throughout Heineken, to monitor line efficiencies and possibilities for improvement. This results in a relatively high insight in determinants of fixed costs. However, it is often not clear which KPI is more determinant than the other. In other words: it is not always clear what the dollar impact is of a certain KPI; for example the costs of ten minutes downtime at the packaging line. In general production-line manufacturing industries are ideal for a highly structured, standardized cost accounting system (Lewis, 1993).

Example 2.3 Improving logistics with Celpay

For the production departments at Bralima, activity-based costing could contribute to the insight in fixed costs determinants; especially by adding dollar impacts to existing KPI’s. Another example of

(19)

the potential advantage of ABC at Bralima in providing insight in dollar impacts is the distribution efficiency, which is discussed in Example 2.3.

The marketing department is not traditionally cost oriented. It emphasizes more on sales volumes and market share and requires decision-making techniques that are flexible and timely. Because of this focus on sales, the marketing department is far more occupied with how to spend money than with how to save money. The difficulty in general here is that there is no clear relation between the amounts of money spent on marketing on the one hand and sales volume on the other. This makes it difficult to find a balance between the two. These and other factors have made Bralima struggle with controlling marketing costs for the last two years. Activity-based costing can provide significant insight in which activities are not efficient and where to search for process improvements.

Ineffective cost management

For the production department it is sometimes difficult to prioritize KPI’s as it is not always clear which KPI variations have the highest impact on costs. This may well lead to ineffective cost management decisions. ABC can support KPI priority decisions by determining the dollar impact of KPI variations.

Within the marketing department of Bralima, little insight in the determinants of fixed costs combined with a low priority has led to a significant overspending of budgets in the last two years. Efforts have been made to improve cost control within the department, but it has been found difficult to find a solid financial base to redesign processes and activities.

Inaccurate budgets

As stated before, the marketing department faces significant overspending of its budget. The last paragraph touched upon some cost management issues in this context. The accuracy and validity of the marketing budget also has a big influence on the effectiveness of cost management. At the moment budgets are set on very subjective and ad hoc grounds by the head of the marketing department without consulting many middle managers. This makes the budget very unreliable and not transparent.

The lack of transparency and reliability often causes significant difficulties in controlling and judging costs. An application of ABC, called activity-based budgeting, could be of high value in providing more accurate budgets to support cost management efforts.

High fixed cost burden

In 2003, gross sales proceeds increased by 8%, whereas the total fixed costs increased by 13%. This has induced that more than half of the potential extra profit was consumed by extra overhead. Overall,

(20)

importance of and the need for an effective cost management system. Otherwise, much of the extra sales volume, created in this period of economic growth, will not result in a relative equal increase of profit. The high fixed cost burden clearly affects Bralima’s overall company performance.

With regard to fixed costs, marketing accounted for relatively the highest increase by spending 33%

more than a year before. Like mentioned earlier, gross sales proceeds only increased by 8%. Most effort in increasing effectiveness of cost management systems is therefore needed in this department.

Although Bralima took some action to tackle this problem, the situation is still far from perfect.

2.1.3 Conclusion

The previous paragraphs have given an overview of problems faced by Bralima that can be linked to activity-based costing. The link is based on two types of missed benefits of ABC by not using its full potential. One missed benefit is that ABC provides more accurate management accounting

information. Ineffective foundations of strategic and tactical decisions, which result from less accurate information, ultimately lead to sub-optimal company performance. The other missed benefit is that ABC provides important new insight in determinants of fixed costs. This is very important for designing a more effective cost management system to solve the high fixed costs burden. In the end, sub-optimal utilization of activity-based costing has a negative effect on the overall company

performance. This makes it a relevant functional problem for Bralima. Figure 2-3 summarizes the web of problems related to this issue.

Figure 2-3 Web of problems

(21)

2.2 Theoretical concepts

2.2.1 Theoretical framework

From 1925 until the mid 1980’s management accounting practice changed remarkably little (e.g.

Kaplan, 1984; Johnson, 1987). Traditional accounting systems of this period tend to use highly aggregated cost information that is allocated to products by simplistic and arbitrary methods. These systems attach overhead costs to each unit of output in proportion to a time-based work measure such as labor or machine hours. The underlying assumption of this approach is that overhead resources are consumed on a time basis. This means that the more time is spent on an item the more overhead costs should be allocated to it. As labor hours will vary closely with output levels, this approach can be viewed as volume-based (Cooper, 1994). Although this approach has the advantage of simplicity, it will lead to significant miscostings when overhead costs are not volume driven (Innes and Mitchell, 1995). The allocations are often unrelated to demands made by the products on organizational resources (Swenson, 1995). Specifically, such cost systems systematically overcost high-volume, low- complexity and undercost low-volume, high-complexity (Dearman and Shields, 2001). Activity-based costing¹³ (ABC) was introduced in the late 1980’s to overcome the disadvantages of traditional cost accounting systems by providing the means to use additional causes of overhead costs.

For more than a decade ABC has received widespread attention in the management accounting literature; especially in the 1990’s (see Bjørnenak and Mitchell, 2002). The ABC approach measures the costs of objects by first assigning resource costs to the activities performed by the organization, and then using causal cost drivers to assign activity costs to products, services or customers that benefit from or create demand for these activities (Cooper, 1988). In particular, activity based systems measure more accurately the costs of activities not proportional to the level of outputs produced. In manufacturing processes four levels of activities can be identified:

1. unit-level (volume-driven) activities performed every time a unit of product is produced 2. batch-level activities performed every time a batch of products is produced

3. product-sustaining activities performed to support the manufacture of a given product 4. facility-sustaining activities performed to sustain the production of products in general

For example, ABC systems use cost drivers that capture the characteristics of the product batch, such as the number of inspections or the number of production orders, to assign the costs of batch-level activities to products. The costs of product-sustaining activities are assigned to products using cost drivers that capture the characteristics of the product itself, such as the number of components it

13 Some researchers prefer to use the term activity-based management (ABM) for applications of ABC which

(22)

contains or the number of engineering change orders it required. ABC advocates claim that this approach captures the economics of all of the processes in the value chain more closely than traditional unit-based cost systems, thereby providing more accurate cost data (e.g., Cooper and Kaplan, 1991). They argue that ABC provides the sophisticated cost data needed to make appropriate product mix, pricing, process improvement and other key decisions. These claims have prompted a growing number of firms to implement ABC systems. Yet, many firms struggle with the fact that their ABC systems are not being used to the full potential of these applications.

Intentions and objectives

During the 1990’s a growing number of firms gained interests in the possible benefits of activity-based costing applications. Most firms initially adopted ABC to improve strategic decision making such as product costing and customer profitability analysis. But as ABC systems evolved, more emphasis was placed on supporting operational improvements (Swenson, 1995; Innes and Mitchell, 1995).

According to Swenson (1995) this trend reflects the perception of practitioners that developing a better product costing methodology is a one time fix. An operational improvement, on the other hand, is a continuous process.

Utilization

Cooper et al. (1992) found that although some companies claim to have successfully implemented ABC, its functional use is very limited. Many companies had performed extensive activity analysis, but this was followed by very little action. According to these researchers ABC information by itself does not invoke actions and decisions leading to improved profits and operating performance.

Management must institute a conscious process of organizational change and implementation if the organization is to receive benefits from the improved insights resulting from an ABC analysis. A number of studies have tried to measure the success of ABC and its determinants. These studies usually focus either on the potential applications or on the implementation of an ABC system.

System functionality and quality

As stated earlier, initial benefits are mainly based on new insight in product and customer costs (cost object costs). However, the benefits of ABC have gradually been extended as organizations have become experienced in its use. Innes and Mitchell (1995) list a number of specific applications of ABC that assist in cost control and reduction or identify the cost implications of decisions. Swenson (1995) provides specific examples of how managers use the ABC information to support strategic and operating decisions. Strategic decisions include: sourcing, product/pricing mix, customer profitability.

Operating decisions include: process improvement, product/process designs, and performance measures.

(23)

Organizational embedding

In the late 1980’s and early 1990’s, when more firms started to design and implement ABC systems, most effort concentrated around the architectural and software design of ABC without giving sufficient attention to behavioral and organizational issues involved (e.g. Cooper, 1992). Many researchers emphasized the importance of these behavioral factors in determining the success of ABC implementation (e.g. Shields and Young, 1989; Young and Selto, 1991; Argyris and Kaplan, 1994;

Shields, 1995; Anderson 1995; Swenson, 1995; Foster and Swenson, 1997). Anderson (1995) found similarities in the information technology implementation literature, which developed from a rational view of technological factors to a more complex view of interacting socio-technological factors.

Meanwhile significant progress was being made in the accounting change literature, of which ABC implementation is merely an example. While the researches of ABC implementation focused more on the procedural factors influencing the success of the implementation process, other researches tried to grasp the structural factors, which either support or hinder accounting change in a more general sense.

In this latter effort some have tried to identify the advancing forces of change in this context (e.g.

Innes and Mitchell, 1990; Cobbs et al., 1995). Many others have discussed various barriers to change (e.g. Markus and Pfeffer, 1983; Brooks and Bate, 1994; Roberts and Silvester, 1996; Strebel, 1996;

Hardy, 1996; Kasurinen, 2002). In this thesis ‘structure’ is related to both concrete structures (e.g.

existing reporting systems) and abstract structures (e.g. organizational culture).

Implementation process

According to the management accounting literature many companies struggle with the implementation of an effective ABC system and are sometimes even not successful. An important reason for this is that many companies have focused the introduction of ABC on the architectural and software design, without giving sufficient attention to the behavioral and organizational aspects involved (Cooper et al., 1992; Hankinson and Lloyd, 1994; Argyris and Kaplan, 1994). Shields and Young (1989) argued that the key to successfully implementing ABC is effectively dealing with these behavioral and

organizational variables. As a result many ABC researches aimed at identifying the key determinants affecting the ABC implementation process. Shields and Young (1989) developed a comprehensive theoretical model about the implementation of cost management systems that is applicable to ABC, in which they identify seven behavioral and organizational variables as being important to the

implementation process. Shields (1995) narrowed these seven variables down to five, in an empirical follow-up research:

1. Top management support

2. Linkage to competitive strategy (mainly JIT and quality systems) 3.

(24)

4. Training in implementing cost management systems

5. Non-accounting ownership (this refers to the involvement of other

departments, alongside accounting and finance, in ABC development and use) Argyris and Kaplan (1994) describe two processes required to implement a new practice for managing an organization in general, and ABC in particular. During the first process, called education and sponsorship, change advocates articulate the benefits of the new innovation to persuade key individuals and authoritative managers in the organization of the value of the proposed approach.

During the second process, called creation of internal commitment, barriers resulting from defensive routines have to be overcome. These defensive routines usually aim to avoid the experience of embarrassment or threat by bypassing situations that may trigger these responses¹⁴. The internal commitment process must overcome these barriers by motivating individuals to implement the new ideas and to take action based on their implications.

Advancing forces of change

While the specific researches on ABC implementation focused more on the procedural factors influencing the success of the ABC implementation process, the accounting change literature aimed more to map the structural factors which either advance or hinder accounting change in a more general sense. Innes and Mitchell (1990) divided the structural advancing forces associated with change into three categories on the basis of nature and timing of their influence. Motivators are related to change in a general manner (such as a competitive market or production technology). Facilitators are

necessary factors but not sufficient, per se, to result in change (such as accounting computer resources and the degree of autonomy from the parent company). Catalysts can be directly associated with the change (such as poor financial performance and loss of market share).

Cobb et al. (1995) further developed the accounting change model by adding leaders and momentum¹⁵ as important internal advancing factors of change, whereas the categories of Innes and Mitchell remained as the external factors. On top of this they emphasized the importance of factors hindering, delaying or even preventing change; these were referred to as barriers. The accounting change model of Cobb et al. is presented in Figure 2-4.

14 Argyris and Kaplan refer to this behavior as ‘Model I behavior’.

15 Cobbs et al. (1995) roughly define momentum in this context as ‘the expectation of continuing change’.

(25)

Motivators Catalysts Facilities

Potential for change

Leaders Momentum

for change

Barriers to change Change

Figure 2-4 Accounting change model (Cobb et al., 1995)

Barriers to change

The accounting model of Cobbs et al. (1995) has succeeded in combining the forces advancing and hindering change. However, it seems to be limited in the sense that it has only a general category of barriers, whereas the advancing forces are divided into subcategories (motivators, facilitators, catalysts, leaders and momentum). Kasurinen (2002) tried to overcome this limitation by revising the model. He divided the barriers to change into three categories. Confusers tend to ‘disrupt’ the change process; for example uncertainty about the project’s future role in the organization or different views on change. Frustraters tend to ‘suppress’ the change attempt; for example existing reporting systems and organizational culture. Delayers are often directly related to the new management technology in question (e.g. ABC). The revised accounting change model of Kasurinen is presented in Figure 2-5.

(26)

2.2.2 Conceptual model

The previously covered theoretical concepts imply that if an activity-based costing system with the right characteristics is effectively implemented within a certain set of advancing and hindering factors, it holds the potential for being a structurally more effective management accounting system than the traditional volume-based cost system. This study therefore defines an ABC reporting system as being successful when firms achieve optimal utilization of its full potential. Utilization in this context is the result of a combination of the technical system characteristics (such as functionality, structure and usability) and the organizational embedding. The conceptual model used in this thesis is presented in Figure 2-6.

Utilization ABC System of

Quality

ABC System of Organizational

Embedding

Barriers to

Change Implementation

Process

Advancing Forces to Change Structure

Functionality Usability

Figure 2-6 Conceptual model

The term ‘organizational embedding’ is used as the outcome of the accounting change process

initiated by the introduction of the ABC system. This change process is analyzed on both the structural and procedural dimension. The influencing forces on accounting change presented in the model of Kasurinen (2002) form the basis to analyze the structural factors (i.e. the organizational context) which cover both the advancing forces as well as the barriers to change. The ABC implementation literature provides relevant tools to analyze the procedural factors concerning the organization of the

implementation process itself (e.g. the model of Shields [1995]).

(27)

The system quality is analyzed on three dimensions. First, the structure of activities, cost drivers and cost objects is broadly described and the output is compared to a volume-based output. Second, the system’s functionality is mapped by cross-linking potential analysis tools (system output) and the action intents of its users. Potential tools and intents come from both theory (e.g. Cooper et al., 1992;

Innes and Mitchell, 1995; Swenson, 1995) and practice (e.g. interview results). Finally, the system usability is assessed in terms of for example accessibility and frequency of data updates. These dimensions emerged from interviews and active involvement in operational work on ABC within Bralima. They are considered to be the most relevant and most influencing factors of an effective use of activity-based cost information in this specific case. Cooper et al. (1992) used similar dimensions to analyse the ABC-models in their study of eight case sites.

The conceptual model clearly shows that although a system’s quality for example can be relatively high, a relatively low success of implementation (which is influenced to a certain degree by the contextual factors) can still result in little utilization. In other words, in that case much of the design effort would be lost due to unsuccessful implementation; of course it is also possible that

implementation efforts are lost due to design errors or by not taking contextual factors sufficiently into account. These implications have been made explicit, because they are important when analyzing the utilization of ABC’s potential at Bralima.

The quality of the ABC system design combined with the degree to which this system is embedded in the organization provides important insight into how well Bralima currently utilizes ABC’s potential.

Next, it provides important clues to direct the centre of attention when trying to improve ABC utilization. Finally, it can assist in setting up concrete change steps for that improvement, because ultimately, action is needed to get things done and getting things done is in this case in the best interest of Heineken as a whole and Bralima in particular.

2.3 Problem statement

The problem statement has two functions: harmonisation with the client, Bralima, and internal direction of the research (De Leeuw, 2001). It has three elements: research objective, research question and boundary conditions.

2.3.1 Research objective

The research objective determines recipients, deliverables (output) and the relevance of the research.

Based on the problem description above, the research objective for this thesis is formulated as follows:

(28)

2.3.2 Research question

The research question formulates the main question which relates to the research objective. This main question will later be divided into sub questions, which in turn form the basis of the research. Based on the problem description above and the research objective, the research question is formulated as follows:

2.3.3 Research sub-questions

The main research question will be divided into sub-questions based on the conceptual model. The sub-questions structure the way in which the main research question is answered. After the sub- questions are formulated below, a structural research model is presented which links the sub-questions to the following chapters. These chapters further operationalize the concepts covered by the sub- questions.

System quality

1. What is the quality of Bralima’s ABC system in terms of structure, functionality and usability?

2. How can the quality of Bralima’s ABC system be improved in terms of structure, functionality and usability in order to increase its utilization?

Organizational embedding

3. Which barriers to change have limited the ABC system to be fully embedded in the organization of Bralima?

4. How can the barriers to change be overcome in order to increase the utilization of Bralima’s ABC system?

5. Which advancing forces to change can support the ABC system to become fully embedded in the organization of Bralima?

6. How can the advancing forces to change be better exploited in order to increase the utilization of Bralima’s ABC system?

(29)

7. To what extent has the implementation process contributed to the organizational embedding of Bralima’s ABC system?

8. Which implementation efforts are needed to let the ABC system become embedded in the organization of Bralima?

Each sub-question will be answered in one of the following chapters. Figure 2-7 presents the structural research model which links the research question, research objective and sub-questions on the one hand to the chapters on the other.

(30)

Chapter 3

Analyze and improve the quality of the ABC system

Sub-questions 1 & 2

Chapter 4

Analyze and limit the barriers to change

Sub-questions 3 & 4

Chapter 5

Analyze and exploit the advancing forces to change

Sub-questions 5 & 6

Chapter 6

Analyze and improve the ABC implementation process

Sub-questions 7 & 8 Research Objective

The research objective is to develop a model to increase the utilization of activity-based costing to its full potential in better supporting strategic decisions and process improvements made by the senior and middle management of Bralima.

Research Question

How can utilization of activity-based costing be increased to its full potential in better supporting strategic decisions and process improvements made by senior and middle management of Bralima?

Chapter 7

Assemble conclusions and recommendations

Research question

Figure 2-7 Structural research model

2.4 Boundary conditions

This thesis will mainly be conducted at the head office and main production facility of Bralima s.a.r.l.

in Kinshasa, Democratic Republic of Congo. The greater part of research activities is done during a six month internship. Although all five production sites in Congo fall within the scope, all research

(31)

activities take place in Kinshasa. This aspect of scope is limited by the time limit and traveling opportunities within the country¹⁶.

Besides all findings at Bralima, corporate policy and strategy form an important element of this thesis.

It initiated ABC implementation and limits system design changes (e.g. standardization of cost drivers, activities, etc.). Within Bralima three departments are included in the scope of this research: Finance, Production and Marketing. These departments are thought to be most relevant for this thesis since ABC is most applicable to these business functions (Lewis, 1993). Other departments are therefore excluded in this research with regard to ABC applications.

2.5 Sources of data

The data for this thesis comes from three sources: (1) business documents and systems, (2) own observations during operational work, (3) interviews with managers of Bralima and Heineken.

2.5.1 Business documents and systems

The most basic source of data are the available business documents and systems. These include organizational charts, strategic plans, (financial) reports, formalized processes, presentations, market research reports and software user manuals. These documents will be used mainly to get general insight in the organizational processes and strategic intent. Some of these documents are directly available as a result of operational work for the financial department of Bralima. Other documents were requested upon or provided as a result of the interviews. The guidelines used for the analysis of this type of data are listed in appendix D2.

2.5.2 Own observations

The second source of data are own observations during operational work on technical ABC system improvements¹⁷ and the update of the 2004 ABC figures. This is an important source of information for this research. Being directly involved in the development and update of Bralima’s ABC model has enabled me to gain first hand experience on this process. This knowledge would otherwise have been difficult to obtain through interviews; partly because I would have had to rely on second hand information and partly because the process is fairly complex. Especially the insight that this

experience brought in the procedures and the technical characteristics of the ABC system is considered to be a valuable asset in this research.

16 There is a security policy which limits Interns to travel within the Democratic Republic of Congo to a certain degree.

(32)

2.5.3 Interviews

The third source of data are the interviews with managers of both Bralima and Heineken. They can be classified into three categories¹⁸. The first category consists of corporate-level managers who were responsible for implementing ABC at more than one operating company. These individuals can be characterized as facilitators of the ABC implementation effort, but not as leaders of the

implementation team. The second category consists of financial managers of Bralima. They are generally responsible for the implementation of ABC and preparing the ABC reports, but they also use ABC information to support decision making. The third category consists of non-financial managers.

They are potentially the primary users of ABC information.

Table 2-1 and Table 2-2 list the interviews used in this research. The interviewees are categorized according to the above discribed categories. The table further illustrates the subjects that were discussed in each interview. The specific interview formats used for each subject can be found in appendix D1. The subjects are directly linked to the research sub-questions found in chapter 2.3. The subjects were formulated in a preliminary stage of the research (based on both literature and own experiences) and in that context were thought to be essential in answering the different research sub- questions. However, these subjects are never thought to cover every important aspect since it would be virtually impossible to know all these beforehand. Therefore, it might occur that the enlisted subjects and interview questions do not entirely cover the content of the actual thesis; additional questions and issues could have cum up during the interviews themselves.

17 This includes a global review of activities and cost drivers based on the 2003 system. A thorough review would be beyond the scope of this research, regarding the time limit.

18 Swenson (1995) used a similar classification in his research.

(33)

ABC applications Use of ABC figures ABC implementation process Cost centre structure Activity structure Cost drivers Treatment of external costs Frequency of data updates Data collection Software Future role of ABC Reporting Corporate relations Cultural issues and HRM Market and economic conditions Corporate policies Corporate use of ABC Linkage to competitive strategy Training Performance evaluation and compensation Related research questions _all _all _all _1/2 _1/2 _1/2 _1/2 _1/2 _1/2 _1/2 _3/4 _3/4 _3/4 _3/4 _5/6 _5/6 _5/6 _7/8 _7/8 _7/8 Interviewee Department Place Date Subjects

General manager Africa Heineken Africa Kinshasa December 2004 x x x x x x x x x x x x x

Financial manager Africa Heineken Africa Amsterdam July 2005 x x x x x x x x x x x x x x x x x

Business analyst Africa Heineken Africa Kinshasa November 2004 x x x x x x x x x

Kinshasa January 2005 x x x x x x x

Category 1

Amsterdam April 2005 x x x x x x x x x

Financial director Finance Kinshasa February 2005 x x x x x x x x x x x x x

Director planning and control Finance Kinshasa October 2004 x x x x x x x x

November 2004 x x x x x x

January 2005 x x x x x

February 2005 x x x x x x x x

ABC project manager Finance Kinshasa October 2004 x x x x x x x x

Category 2

January 2005 x x x x x x x x x x

Table 2-1 Interview table: category 1 and 2

Successfully Implementing Activity-Based Costing