The implementation of performance
measurement systems and employee
behavior.
The implementation of performance
measurement systems and employee
behavior.
Author: Bas Eilering
Student number: 1386794
Email: baseilering@hotmail.com
University: University of Groningen (RUG) Faculty: Faculty of Economics and Business
Master: Master of Science in Business Administration Specialization: Organizational & Management Control
Supervision: First supervisor: Dr. B. Qin
Second supervisor: Dr. M.P. van der Steen
3
Preface
Many organizations nowadays use a form of performance measurement. This topic is widely addressed in scientific literature. However until recently the link between the economic-, sociological- and the psychological side of performance measurement was not yet covered. Through a course given by Mr. van der Steen I became aware that this is a logical next step in the scientific research of economics and therefore I thought this would be an interesting topic to write my thesis about.
After almost seven years of study this will be the final work for my master Organizational & Management Control. I am looking back with mixed feelings on my time as a student. I have had a great time and I have grown a lot in both social and professional aspect, however the time has come to leave it all behind and to take the next step. I am looking forward to bring my accumulated knowledge into practice and would like to thank my parents, my friends and the University for providing the platform to develop myself.
Finally I would like to thank my supervisors for helping me in the right direction with this thesis and hope you will enjoy reading it.
Yours sincerely,
Bas Eilering
4
Executive summary
Because of existing goal-incongruence between employers and employees, motivational problems amongst employees and a lack of direction, a form of control is desired by management to structure and guide actions and behavior of employees. One type of control mechanisms are performance measurement systems. These systems check a result or an outcome with a predetermined target and where necessary take action somewhere along the process to achieve established goals.
To implement such a system the company should first determine which system is needed in their specific situation and then implement it. The major barrier for implementation success is the resistance it creates amongst employees. To overcome this, their behavior needs to be influenced and directed in a way favorable by management.
Because different factors play a role in the process of implementing a new performance measurement system, a framework has been developed to guide this process. Within this framework the connection between performance measurement systems, behavior of employees and the creation of networks is explained. This framework can be a plan of action for managers and emphasizes key factors that play a role in making the implementation process a success.
5
Table of contents
Preface 3 Executive summary 4 Table of contents 5 1. Introduction 7 1.1 Relevance 8 1.2 Structure 8 2. Research design 9 2.1 Problem statement 9 2.1.1 Research goal 9 2.1.2 Research questions 92.1.3 Preconditions and restrictions 10
2.2 Research method 10
3. Management control systems 11
3.1 Agent and principal 11
3.2 Management control 12
3.3 Effect of management control systems 13
3.4 Control instruments of management control systems 14
3.4.1 Formal and informal control 14
3.4.2 Types of management control 14
3.4.3 Results, actions, personnel and cultural control 15
3.5 Similarities and differences 17
3.6 Summary 17
4. Performance measurement systems 19
4.1 Performance measurement 19
4.2 Performance measurement systems 19
4.3 Models of performance measurement systems 21
4.4 Implementation of performance measurement systems 23
4.4.1 Barriers to implementation success 23
6
5. Behavior of employees 25
5.1 Integrity from a behavioral perspective 25
5.2 Institutional theory 26
5.2.1 New institutional economics 26
5.2.2 New institutional sociology 27
5.2.3 Old institutional economics 27
5.3 Group behavior 28
5.3.1 Actions and institutions 28
5.3.2 Rules and routines 28
5.3.3 Framework 29
5.4 Individual behavior 30
5.4.1 Schemas and scripts 30
5.4.2 Cues and sense making 32
5.5 Resistance to change 33 5.5.1 Degrees of change 33 5.5.2 Forms of change 33 5.5.3 Resistance 34 5.6 Actor-network theory 35 5.7 Summary 36 6. Analysis 38 7. Conclusion 41
7.1 Conclusion of the research 41
7.2 Limitations of the research 42
7.3 Recommendations for future research 43
8. Framework applied 44
8.1 Case 44
8.2 Case applied to the framework 45
References 46
7
1. Introduction
During almost all my courses in Economy & Management the focus was on the theoretical approach of science. If we would live in a perfect world under perfect circumstances then economic models could predict certain outcomes. This of course is relevant but during my jobs on the side I always noticed there is more than numbers and figures; there are people. Especially when I worked at a gas station which was taken over by a large corporation I saw a lot of struggle and resistance among employees despite of management effort, information meetings, guidebooks, emails etc.
So when I took the course Ontwerp van Management Accounting en Verandering I was pleased to see that this course tried to link economics to sociology and psychology. There is a gap between what economic models tell and predict and what the results on the work floor are. This course showed me that often implementation of models into the organization fails despite of checklists and management confidence in good results.
8
1.1 Relevance
Performance measurement systems started from traditional costing and accounting systems and developed throughout the years. They became more balanced and non-financial measures were added (Bourne et al., 2000). However, it is still a form of measurement that observes actions as they happen or already did occur instead of focusing on why certain things happen. It is necessary to know why employees do what they do to really be able to change and control systems and lead a company. This is a logical next step which is a relative new topic in scientific research. Especially the link between the implementation of performance measurement systems and on how behavior is established has not yet been researched. Therefore this thesis will be useful to give a clear overview and new insights concerning management problems and how to deal with this implementation process. This thesis will give insights in the background of performance measurement systems and on how behavior is established. Finally these two different topics will be presented in a model which could be used by managers as a guideline for implementing new performance measurement models. This should lead to more awareness and guidance when managers want to implement a new system.
1.2 Structure
This thesis will mainly exist out of two topics;
management control and corresponding performance measurement systems, and behavior of employees.
9
2. Research design
Throughout this chapter the research design and methodology used will be explained and clarified. This will be done by looking at the problem statement which exists out of the goal of the research, the main research question and several sub-questions. This is done to give meaning and structure to the research. All this will be done according to the book on methodology of de Leeuw (2001).
2.1 Problem statement
According to de Leeuw (2001) a problem statement should be relevant, effective and researchable and should exist out of a research goal, research questions, and preconditions and restrictions.
2.1.1 Research goal
The goal of the research is to build a bridge between the economic and psychological side of the implementation of management control systems. Eventually a model is developed which should help managers to understand the underlying factors influencing employee behavior when dealing with uncertainty is case of the implementation of a new performance measurement system. This model should present guidance for managers who want to implement a new performance measurement system. The model presents a step-by-step path to be taken when management wants to implement a new performance measurement system.
2.1.2 Research questions
The main question to be answered is:
“How can managers influence the behavior of employees to make the implementation of performance measurement systems more effective?”
This will be answered by dividing the main question into sub-questions and by answering these. These questions are:
1. “What is management control and where does it come from?”
2. “What is performance measurement?”
3. “What types of performance measurement systems are available?” 4. “How are these performance measurement systems implemented?”
5. “What is behavior and how is it developed?”
10 2.1.3 Preconditions and restrictions
The preconditions and restrictions make sure that the research is focused on aspects relevant to the research. It has to be sure that the literature used is reliable and valid and is in line with what is expected from the University of Groningen for graduating the master O&MC. The conditions and restrictions for this thesis are:
It has to be written in English.
The literature used should be scientific.
The student should follow the structure presented on the website of the university.1 The thesis has a value of 20 European Credits (EC).
The scientific information should be reliable and valid.
2.2 Research method
De Leeuw (2001) points out that there are five different types of research that can be conducted. These types depend on the degree of involvement of an external client. Because this research has no direct external client or interested party this research is purely scientific. Furthermore this thesis will be partly a desk- and partly an exploratory research. The first part will be a literature study were existing literature is collected and combined to answer the sub-question. This is done through different scientific websites that provide scientific papers. After that the analysis will provide an overview of the literature where a model will be the base of explorative research. This model should provide insights and a deeper understanding on the process of implementing performance measurement systems and the behavior of employees. Below (figure 1) a schematic overview of the paper is presented. Overview Introduction Research design Management control systems Performance measurement systems Behavior of employees Literature Analysis Analysis Conclusion Recommendations Executive summary
Literature study Exploratory study
Descriptive
Figure 1: Overview of the research.
1
See:
11
3.
Management control systems
In this chapter the focus will be on management control systems (MCSs). It will be explained which factors are responsible for introducing MCSs and what different authors say about them. This is necessary to fully understand MCSs, because Performance Measurement Systems (PMSs) are part of MCSs. Therefore the starting point will be the introduction of MCSs to work our way down to finally see the effect on actions and behavior of employees.
3.1 Agent and principal
To understand why MCSs were introduced as a solution to overcome management problems this part will look at the reasons and motivation behind the MCSs. It starts with the relation between „agent and principal‟ which will be explained further on.
In most firms the person owning the business and the one running it are different persons and most of the time these persons have different interests. For example the person leading the company wants as much money being paid for as less work as possible and the person owning the company wants it the other way around. This is called agency theory; the relation between the agent and principal where the principal is the one that depends on the actions of the agent and the problems that arise with it (Shapiro, 2005).
In general terms, agency theory focuses on co-operation of agent and principal in the presence of external effects as well as asymmetric information. The agent is making decisions and thus affecting his own welfare and that of the other individual called principal. The principal is willing to provide a certain reward to the agent in return for a decision/action/effort according to the will of the principal (Spremann, 1987). However most of the times the principal cannot observe the agent‟s actions in full detail (asymmetric information) and therefore control mechanisms are used to align the interest of the agent with those of the principal.
One way of trying to overcome the agency problem is by introducing management control systems. Besides the agency problem Merchant & van der Stede (2007) have three more reasons for using management control systems.
1. Lack of direction (unclear goals);
2. Motivational problems (individuals are self-interested); and
12
3.2 Management control
The last thirty years MCSs are widely studied and three major fields of research are addressed; contingency theory, economic theory and psychological theory. All three theories have different approaches regarding MCSs which will be the topic of this section.
The basics of MCSs are to manage the tension between creative innovation and predictable goal achievement and to balance the basic organizational dilemma between control and flexibility. Also MCSs try to realize goal-congruence; influence behavior of people to reach conformity between self-interest and company-self-interest (Henri, 2004).
This is done by applying a form of control:
“Control is the application of policies and procedures for directing, regulating and coordinating production, administration and other business activities in a way to achieve the objectives of the enterprise” (Otley & Berry, 1980).
The contingency theory states that the design and use of MCSs is being influenced by the context in which the MCS is active (Chenhall, 2003). The right match between a contingency variable and the MCSs makes sure that personal and organization achievements are at their best (Fisher, 1998). The most important factors influencing the design of MCSs are -according to Chenhall (2003)-:
Environment, technology, size, organization structure, strategy and organization culture. Also Fisher
(1998) adds uncertainty as an important contingency variable.
Economical and psychological research has focused on the implementation and use of MCSs. Economical theories mainly focus on the roll that reward structures play in connecting employees to the goals of the organization (Chenhall, 2003). And psychological research has a focus on the individual characteristics of MCSs, for example the personality and learning-curve of an individual. As Ouchi (1979) states there are two ways in which an organization (the principal in this case) can achieve effective control over employees (the agent in this case). They can search for people who exactly fit the needs of the company and act as the company expects them to do, or they can hire people who do not exactly fit the needs and monitor them. This paper focuses on the latter, because it is almost impossible to find and hire people that exactly fit the company‟s needs and act precisely as the principal desires without any form of control.
13
3.3 Effects of management control systems
There are different sorts of management control systems which will be pointed out in this section. Also the effect of MCSs on employees will be addressed.
According to Anthony and Govindarajan (2006) the basic activities of MCSs are: Planning; what should the company do,
Coordination of different activities, Communicate information,
Evaluate information, Controlling actions and,
Influence behavior of employees.
All these activities require a joint effort of managers and employees to achieve a common goal. This goal is to improve the organization by being more effective, efficient, and by satisfying employees (Fisher, 1998). To achieve this, employees need to be influenced in the right way. Chenhall (2003) points out that well adjusted and implemented MCSs lead to job satisfaction and employees who are satisfied with their jobs will do their job better, make better decisions, can identify themselves with organization goals and work more efficient.
Williamson (1985) states that control instruments influence opportunistic behavior, meaning that employees strive to achieve personal goals. According to Williamson (1985) employees make a choice between personal gains and losses and MCSs can reduce the negative behavior by increasing the costs for this negative behavior to the employees; i.e. firing, no promotion etc. However, Ghosal & Moran (1996) turn this around and state that control instruments can in fact encourage employees to achieve their own goals because MCSs take away negative feelings towards the organization by more communication.
In the literature different negative aspects of MCSs are addressed. According to Merchant & van der Stede (2007) there are two different sorts of costs involved with MCSs; direct and indirect costs. Direct costs are those costs that have to be paid directly by the firm for the control instruments; for example, rewards and training.
14 MCSs can slow down activities, because activities need to be checked afterwards (Merchant & van der Stede, 2007). And finally MCSs can add a negative attitude to employees by reducing the freedom they experience and so they feel their autonomy has reduced. This can lead to stress and induce a lower productivity (Merchant & van der Stede, 2007).
Ghosal and Moran (1996) add one more negative effect of MCSs, being that MCSs can reduce the intrinsic motivation of employees by increasing their extrinsic reward. This means that the will of employees to do the right thing all by themselves is reduced by increasing the external pressure on employees.
3.4 Control instruments of management control systems
MCSs exist out of different elements of control which is categorized in different ways by different authors. In this section these different elements will be explained and an overview of categories used by Anthony (1989), Ouchi (1979) and Merchant & van der Stede (2007) will be given.
3.4.1 Formal and informal control
Anthony (1989) has categorized forms of control in two categories; formal and informal control. Formal control instruments are rules, procedures and budget systems, which are specifically designed. Informal control is less visible; these instruments are not specifically designed but are developed over time such as organization culture, which are the collective believes and values within an organization.
3.4.2 Types of management control systems
Ouchi (1979) made a categorization of forms of control in „market‟, „bureaucracy‟ and „clan‟. Markets try to overcome problems of control by measuring effort of employees and by putting a „price‟ on their effort. This is the most efficient form of control when an employer can observe the employee and has access to all the information he/she needs about the employee (Ouchi, 1979). The focus here is on the generated output of the employee.
15 Clans focus on the social aspects of control by removing „goal incongruence‟. There are no prices or rules to control employees, because a clan tries to align values and believes within the group. Clans rely on similar ideas within the group about desired behavior (Ouchi, 1979).
In theory these mechanism are strictly separated, however in practice there can be an overlap between different forms of control. Which form is the most appropriate one depends on the ability to measure outputs and the knowledge of the transformation process (figure 2). The ability to measure outputs means that outputs can directly be traced to an individual. The principal can monitor the effort of individual employees. Whereas the knowledge of transformation process means knowing in detail how outputs are achieved.
Knowledge of the tranformation process
Ability to measure outputs High Low Perfect Imperfect Behavior or output measurement Output measurement Behavior measurement Ritual and ceremony, “Clan” control
Figure 2: Conditions determining the measurement of behavior and of outputs (Ouchi, 1979).
When the ability to measure outputs is high and the knowledge of the transformation process is perfect -for example someone picking strawberries- they can be monitored by looking either at their behavior (picking strawberries) or at the output (amount of strawberries picked). However, when the knowledge of the transformation process is imperfect (i.e. stock traders) then output measurement is needed. On the other hand a low ability to measure outputs and a perfect knowledge of the process call for some sort of behavior measurement.
16 3.4.3 Result, action, personnel and cultural control.
Also Merchant & van der Stede (2007) derive four types of management control types: 1. Results control (Through results)
2. Action control (Through actions)
3. Personnel control (Through recruiting and selection) 4. Cultural control (Through group values)
With results control the focus is on rewarding the most talented and hard working employees (Merchant & van der Stede, 2007). This combination of rewards and results should lead to the desired behavior of employees. According to Merchant & van der Stede (2007) the implementation of results control desires four steps: Defining the right performance dimensions, measure performance, setting targets and provide rewards to encourage desired behavior. For results control to be effective it is important that three conditions are met (Merchant & van der Stede, 2007).
1. Organizations should be able to determine the desired results;
2. Employees who are controlled should have significant influence on the results they are held accountable for; and
3. Organizations should be able to measure the results effectively.
Action control is the most direct form of control because the emphasis lies on the actions of employees and therefore on their behavior. This type of control can take four basic forms (Merchant & van der Stede, 2007).
Behavioral constraints make it impossible for employees to conduct certain actions, i.e. locks on doors, passwords etc. Another form is by preaction reviews where before actions are taken, the actions are discussed, modified and planned. The third form is action accountability where employees are held accountable for the actions they take and the final form is a form of redundancy where more employees are assigned to a task then is strictly necessary.
Again Merchant & van der Stede (2007) determine two conditions for action controls to be effective. The organization should be able to determine which actions are (un)desirable, and organizations should be able to ensure that these (un)desirable actions (do not) occur.
Personnel controls focus on self-control and social control, so employees “control” themselves. These forms of control help the employees to understand what the company wants to achieve and secondly help the employees to do a good job. This can be achieved by selection, training, job design, and provision of necessary resources (Merchant & van der Stede, 2007).
17 and values. It is believed that these forms of control are most useful when members of the group have an emotional connection.
As can be seen these types of control overlap the ones mentioned by Ouchi. Results control focuses on output, action control focuses on behavior and personnel/cultural control on “clan”.
3.5 Similarities and differences
Anthony (1989) is the one that categorizes control in to two forms; formal and informal, whereas Ouchi (1979) and Merchant (2007) define control into more specific fields. Formal control has rules, procedures and routine as pillars to control employees and informal control makes use of a social form of control where shared values and believes make this sort of control effective. Agent and principals have the same believes and strive to achieve the same goals.
The formal form of control can be compared to the „bureaucracy‟ of Ouchi (1979) and the „action control‟ of Merchant & van der Stede (2007). These forms of control focus on rules, procedures and routines to direct employees. This is a form of extrinsic motivation because not following the rules and/or procedures might lead to being punished, i.e. being fired. In accordance, „market control‟ of Ouchi (1979) and „results control‟ of Merchant & van der Stede also have a formal approach to control. The emphasis is on rewards and therefore also an extrinsic form of control.
On the other hand according to Abernethy & Brownell (1997) informal control can be compared to „clan control‟ of Ouchi (1979) and, „personnel‟ and „cultural control‟ of Merchant & van der Stede (2007). These types of control have a focus on shared values, shared believes and therefore have a more intrinsic way of directing employees. This is because employees have a strong social commitment to the rest of the group/company and thus want to perform the best they can on one‟s own accord.
3.6 Summary
In this chapter the focus was on management control systems. In its most basic form MCSs were introduced to overcome the problem between the principal and the agent, between the employer and the employee. This is because there is existing goal incongruence between the employer and the employee. Most of the time the employee does not automatically do what is in the best interest of the employer and a form of control is needed.
18 focus mainly on the roll that reward structures play in connecting employees to the organization. And research that focuses on the psychological aspect of control focuses on the individual characteristics of MCSs.
Also literature shows that MCSs can have negative and positive effects. MCSs can make employees behave in the way the company wants, but wrongly implemented MCSs can slow down activities and increase costs. Employees might feel too much pressure and perform below average or try to find a way around the MCSs by fraud and/or gamesmanship.
19
4. Performance measurement systems
Performance measurement systems (PMSs) are a form of management control systems. First it was relevant to know the covering assumptions of MCSs before PMSs can be introduced. In this chapter the focus will be on performance measurement; what it is, which systems are involved and which models can be derived? Also the implementation of PMSs will be covered, because it is an important factor for a new system to work.
4.1 Performance measurement
One part of MCSs is to measure performance; you need to know where you are and where you want to go as a company and to check progress towards the established goals. Neely et al. (1995) define performance measurement as the process of quantifying actions, where measurement is the process of quantification and action correlates with performance. This is adopted by Tangen (2004) who states that:
Performance measurement is defined as the process of quantifying the efficiency and effectiveness of actions; and
A performance measure is defined as a metric used to quantify the efficiency and/or effectiveness of an actions.
Traditionally the focus of PM was on accounting and financial measures, however throughout the years PM became more important and systems where developed around it as will be shown in §4.3.
4.2 Performance measurement systems
Kanji (2002) states that: “effective management depends on the effective measurement of performance
and results”. He claims that the first condition to improve business excellence is to develop and
implement a system for performance measurement.
In their most basic form systems in a company use inputs which will be translated into outputs, and are then move back into a feedback loop (Burney & Matherly, 2007).
Input Process Output
20 In PMSs, from time-to-time, data (input) is collected and the computed values (output) are used to assess organizational and/or individual performance. These values are checked against predetermined expected values and when necessary adjustments to the system are made (feedback loop).
Because there is so much research done on PMS, a clear concept of what a performance measurement system is, needs to be established. Ferreira & Otley (2009) say that “PMSs are the evolving formal and
informal mechanisms, processes, systems, and networks used by organizations for conveying the key objectives and goals elicited by management, for assisting the strategic process and ongoing management through analysis, planning, measurement, control, rewarding, and broadly managing performance, and for supporting and facilitating organizational learning and change”.
In the case of performance measurement different systems where developed and analyzed throughout the years that all have a different focus on part(s) of the system mentioned in figure 4. Some have a focus on the input side of the company, others on the output etc. Every company is different and there is no unique and best PMS. It goes beyond the scope of this paper to address the wide variety of all these different PMSs, so for an overview see Tangen (2004). It is sufficient to say that there are basic elements from which all PMSs can be derived and these will be addressed later on (§4.3).
To see the influence of PMSs on the behavior/actions of employees, it is important to distinguish different levels of measurement.
According to Neely et al. (2005) a PMS can be examined at three different levels. 1. The individual performance measures;
2. The set of performance measures – the performance measurement system as an entity, and; 3. The relationship between the performance measurement system and the environment in which
it operates. The Environment Performance measurement system Individual measures Individual measures
Individual measures Individual measures
Individual measures
21 Every PMS exists out of performance measures taken on an individual level. These are the actions and behavior of individual employees which are measured and which all together make the PMSs. Eventually we need to look at the effect of these individual measures on the behavior of employees.
4.3 Models of performance measurement systems
There are many different types of PMSs designed to fit specific companies. As with any field of research the attitude towards PMSs has changed over the years. Bourne et al. (2000) claim that since the late 1970s and 1980s, traditional PMSs have been criticized due to their shortcomings. There was a need for a more balanced approach which depends on a wide range of financial and non-financial measures (figure 4). So a shift from „traditional‟ to „contemporary‟ PMSs was made in the field of research (Burgess et al., 2007).
Traditional financial-based
Items PMSs Contemporary PMSs
Basis of system Accounting standard Company strategy
Types of measures Financial Financial and non-financial
Focus of measures Internal, historical Internal and external, future-oriented
Audience Middle and top managers All employees
Shop floor relevance Ignored Used
Frequency Lagging Real-time
Maintenance Expensive Relevant and easy
Integration Ignored Integration exists
Linkage with reality Indirect, misleading Simple, accurate, direct
Local-global relevance Static, non-varying Dynamic, situation structure dependent
Stability Static, non-changing Dynamic, situation timing dependent
Format Fixed Flexible/variable
Purpose Monitoring Improvement
Function Allocate blame Encourage creative and learning
Decision making Structured Unstructured
Effect on continuous Impedes Supports/stimulates
improvement
Linked to strategy No/less link to strategy Derived from strategy
Figure 5: Comparison between traditional and contemporary PMSs (Burgess et al., 2007).
In spite of the shift from traditional to more balanced PMSs, the main models of PMSs found in the literature can still be directed to five typologies (De Toni and Tonchia, 2001):
1. Models that are strictly hierarchical, 2. Models that are balanced scorecard,
22 And these typologies can be further deducted to three different structures; vertical, balanced and horizontal or by process (figure 6).
Architecture vertical strictly hierarchical models frustrum models Architecture balanced “balanced scorecard” models models with internal-external performances Architecture horizontal models related to value chain Figure 6: Classification of the PMSs (De Toni & Tonchia, 2001).
These three different architectural structures make up for the total field of PMSs and the measures involved with these PMSs and can be categorized in cost and non-cost measures (De Toni and Tonchia, 2001) which can be split up in several other factors (figure 7).
1. Cost measures, including production costs and the productivity.
These measures have a direct link to the final results of the firm in net income and profitability.
2. Non-cost measures, regarding time, flexibility and quality.
These measures have an impact on economic and financial performance but their impact is harder to determine. For example better quality is better for the company but it is hard to point out the direct influence on profitability.
Eventually all PMSs have a focus on different aspects pointed out in figure 7 and therefore this figure can be seen as a complete overview of measures of performance measurement systems.
Performance Measures Non-cost Cost Production costs Productivity Time Flexibility Quality
Materials & Labour Machinery Total Specific Capital Production Internal External Produced quality Perceived quality In-bound quality Quality costs
Run & Set-up times Wait & Move times
System times
Delivery speed & reliability Time-to-market
23
4.4 Implementation of performance measurement systems
For a PMS to be used properly it should be designed, implemented and used. The design part may vary from company to company although most authors agree that measures should be derived from strategy (Bourne et al., 2000). The importance of developing performance measures is finding measures that encourage behavior that will support strategy. However, the most important and difficult part of PMSs to be successful is the implementation of it. This paragraph will focus on the implementation side of PMSs and shed some light on barriers to overcome.
As Cheng et al. (2007) point out the most difficult part of the implementation process is to overcome the barriers to change. “Research has shown that only 20 percent of employees in organizations are
prepared to take positive action in response to change initiatives” (Cheng et al., 2007). Eventually the
success of a PMSs lies in the acceptance of the system by those being “managed”. Cheng et al. (2007) have found that there are several reasons why the implementation of PMSs fails.
4.4.1 Barriers to implementation success
Lack of senior management commitment.
This might be because of a lack of understanding or a perceived threat by the new system. Also managers‟ were not able to see the direct benefits of the new system and search for alternatives.
Employee resistance to change.
This is a crucial part of the implementation because the employees are the ones actually performing the actions. Reasons for this resistance are: Fear of losing jobs, negative experience with previous implementations, change to their internal status, or the stressful conditions because of the change. All these factors have to do with the fact that uncertainty is created.
Inadequate training and support.
Many times when a new PMS is introduced the managers have a clear picture of where to go. However this is not translated down to the employees so they don‟t know what is expected from them. The study done by Cheng et al. (2007) revealed that for a system to be successful the processes involved in change are more important than the actual quality of the system. Therefore the focus of this paper is on how to direct behavior to overcome this resistance to change of employees. Of course it is also important that management supports the system and that enough training and support is present, but for this paper I assume that management wants to introduce a new PMS and therefore their support is in place. The main issue is the uncertainty created amongst employees and because uncertainty may exist on two different levels -behavioral and cognitive2- this uncertainty will be further addressed later on in the chapter about behavior of employees (§5.5.3).
24
4.5 Summary
Performance measurement systems are part of management control systems. It is used by companies to control employees by measuring performance, which is quantifying efficiency and effectiveness of action. This can be done by implementing a performance measurement system. This system needs to establish where you are now, where you want to be and how you are getting there (see figure 3). Like any system a PMS can be examined at different levels. To see whether or not a system is effective you can look at individual measures, the system itself and how the system interrelates with the environment in which it operates. For managers it is important to distinguish between these different levels, because it helps them to see how a system exists out of small parts which all interrelate. They should know where to look when a system is not working as intended and be aware that a relative small adjustment on an individual level might be able to solve the problem experienced on a higher level.
Different systems have a different focus and approach of what to measure and how to measure it. This is different for every company and there is no best system for every company. However there has been a shift from traditional financial-based PMSs to PMSs in line with strategy (§4.3). Also the focus of PMSs became more balanced meaning that the focus shifted from purely financial to covering more „non-cost‟ fields within the organization. In figure 7 there is an overview of different performance measures that might be important for a specific company to implement into a performance measurement system.
25
5. Behavior of employees
The previous chapters covered what management control is and how this is translated in to different performance measurement systems. As discussed before it is necessary for the principal that the agents behave the way the principal wants according to the best interest of the company. The principal tries to control and influence this by implementing PMSs. However, to understand what the effect of PMSs on the behavior of employees is, we have to look in to what behavior is and how it is established. In this chapter we will first look at integrity to rule out behavior that deliberately goes against management instructions. After that institutional theories will be explained to describe the different global approaches to behavioral theories at macro level after which we can define group behavior through routines and individual behavior through scripts.
5.1 Integrity from a behavioral perspective
Before behavior, and how behavior is established amongst employees, is introduced it is important to shed light on integrity. Management tries to direct and control behavior and in this paper the assumption is that this can be done by introducing PMSs. However it might be that all instruments and factors are in place and the system still fails, because employees are not integer. Employees might be involved in fraud, theft etc; all things they know are wrong. If this happens to often a system will never be a success. Therefore in this paper the assumption is that at least the behavior of employees is integer. It may not be the desired behavior by management, but the employees think they are doing the right thing. This is called behavioral integrity.
Parry and Proctor-Thomson (2002) defined a concept called behavioral integrity (BI) where the focus lies on the actual represented values in the behavior of people. Simons (2002) describes behavioral integrity as “the perceived alignment between an actor‟s words and deeds”, meaning that integrity is the key to trust. People may have values, believes and so on, but if these are not followed in their behavior, they are useless. Parry and Proctor-Thomson (2002) use the definition of Becker (1998) to say that integrity is: “commitment in action to a morally justifiable set of principles and values.” This means that there are standards that go further that laws and rules and that people should behave in a way that is concerned with “ought” and “ought not” instead of just “must” and “must not” (Minkes et al., 1999). To understand integrity it is difficult to separate personality from behavior. The difference lies in word and deed and when those two are aligned we can speak of behavioral integrity.
26
5.2
Institutional theory
Institutions are defined by Hodgson (2006) as “systems of established and prevalent social rules that
structure social interactions.” So institutions give structure to social interactions and therefore
influence behavior.3
The beginning of „new institutional theory‟ was founded through research in prisons where it showed that informal procedures on inmates were more of use than guard-enforced formal procedures.
“This was generalized to the notion that explanation of human behavior within institutions had to
encompass not only the old institutional focus on formal rules ("by the book" procedures) and value-maximizing rationality, but also a new focus on the role of norms, symbols, myths, belief systems, and informal arrangements forming the culture of the organization” (Campbell & Pedesen, 2001).
This can be applied to organizations and institutions in a broader sense and there are different approaches to institutional theory. These different approaches try to explain how organizations and the people involved in these organizations are influenced by their environment (Abdul-Aziz et al., 2010). There are three approaches to institutional theory that will be discussed: the New Institutional
Economics (NIE), the New Institutional Sociology (NIS) and the Old Institutional Economics (OIE).
These theories all have a different approach to which factors influence social behavior and these will be explained below.
5.2.1 New Institutional Economics
The most important aspect of NIE is that it uses an economic approach to try to understand behavior. This theory states that people behave in a rational and efficient way over viewing all possibilities and alternatives available for him/her. As Simon (1986) states: “If we accept values as given and constant,
if we postulate an objective description of the world as it really is, and if we assume that the decisionmaker's computational powers are unlimited then two important consequences follow. First we do not need to distinguish between the real world and the decisionmaker's perception of it: he or she perceives the world as it really is. Second we can predict the choices that will be made by a rational decisionmaker entirely from our knowledge of the real world and without knowledge of the decisionmaker's perceptions or modes of calculation.”
In this light ideas and ideologies have no meaning, because complete efficient markets characterize economies. In this perfect and efficient market there are no transaction costs, i.e. search costs, negotiation costs etc., because all information is known. This of course has some pitfalls, because it is hard to assume that people have unlimited mental capacity to see all possibilities and alternatives.
3
27 The transaction cost theory (TCT) has its founding in NIE and believes people are rational, but are limited in their mental capabilities. This theory states that there are always some transaction costs and that companies exist because they try to minimize these costs (Roberts & Greenwood, 1997). TCT assumes that people try to act rational, but suffer from „bounded rationality‟; meaning that they lack the cognitive ability to choose the best alternative available (Williamson, 1975). Williamson (1975) also states that with TCT, opportunism is involved, because people try to act in their best interest and fraud may be present. Therefore behavior of people and effort from management to control this behavior is influenced by economic factors.
5.2.2 New Institutional Sociology
This approach tries to explain behavior of companies and people by placing them in a social setting. This means the actors are not completely rational as with NIE and NIS claims that the environment in which the organization operates influences the form and behavior of the company and its employees. DiMaggio & Powell (1983) researched this by looking at the reasons why companies in similar fields look and act the same, a process they call „isomorphism‟. They claim there are two forms of isomorphism; competitive and institutional. Competitive isomorphism emphasizes market competition where companies compete over resources and customers and is most relevant in areas of open and free competition. Institutional isomorphism goes even further and adds the competition for political power and institutional legitimacy (DiMaggio & Powell, 1983).
Institutional isomorphism describes three processes to explain why companies look-a-like. „Coercive isomorphism‟ results from formal and informal pressures practiced by other organizations and by cultural expectations of society in which the company operates (DiMaggio & Powell, 1983). „Mimetic isomorphism‟ derives from uncertainty; companies look at other companies to find their way in a field that is poorly understood (DiMaggio & Powell, 1983). And the last on „normative isomorphism‟ stems from professionalization. Here isomorphism occurs because professionals of different companies influence each other (DiMaggio & Powell, 1983).
Both NIE and NIS are focused on stability instead of change; they try to rationalize and explain environmental factors to adapt to institutional pressures which are given and cannot be influenced (Roberts & Greenwood, 1997). To explain change „Old Institutional Economics‟ was introduced.
5.2.3 Old Institutional Economics
28 OIE makes the link between actions, rules, routines and institutions and in contrast to NIE where institutions are given and set, OIE assumes they all are interrelated. So rules and routines can influence institutions and the other way around.
5.3 Group behavior
The institutional theory described before shows the different approaches to explain behavior. These approaches are at a high macro level and to describe how individual employees behave this need to be worked down to meso- and finally micro- level. The OIE is the theory that provides the most room for deducting this further down, because it describes the link between institutions, rules and actions. Especially Burns & Scapens (2000) have tried to develop a framework to link these factors. In this chapter the focus will be on explaining rules, routines and actions and on the framework of Burns & Scapens (2000).
5.3.1 Actions and institutions
In the beginning of this chapter institutions were already mentioned and Burns & Scapens (2000) describe institutions as: “A way of thought or action of some prevalence and permanence, which is
embedded in the habits of a group or the customs of people.” They state that institutions bring form
and social coherence upon human activity formed through routinization of human activity. Therefore there is a duality between action and institutions, because when certain actions are performed many times over time by a group of people, these can be institutionalized and institutions give structure and direction to behavior. For example when an employee is new to a company, he or she will probably behave in line with the rest of the employees. However this person also has his/her own thoughts and values and influences his/her environment, therefore institutions are not fixed, but are influenced over time by the actions of the individuals or group.
5.3.2 Rules and routines
In every company there are rules on how to act and to do certain things. These rules can be written down in manuals, procedures etc and are meant to guide the actions of employees the way management wants it. Rules are defined as “the formally recognized way in which „things should be
done‟” (Burns & Scapens, 2000).
The way these rules are actually executed by people are the routines. Routines are described by Burns & Scapens (2000) as “the patterns of thought and action which are habitually adopted by groups of
individuals”. The rules make that employees conduct certain behavior and when this is repeated
29 from the rules. Also routines can influence rules and with an example this will be clearer. When there is a rule that employees can have a break at 12:00 o‟clock and employees take their lunch at that time (routine), this is desired behavior. Also they can take it at another time and it is still a routine but not in line with desired behavior and rules. Furthermore it can be that there is no rule about lunch at all and because most people lunch at 12:00 o‟clock management decides to make this a formal written rule. Therefore there is a duality between rules, routines, actions and institutions and this is captured in a framework by Burns & Scapens (2000).
5.3.3 Framework
Burns & Scapens came up with a framework to explain the links between institutions, routines, rules and actions (figure 7). Here it is clear that rules influence routines and institutions and vice versa. Arrow A shows the translation from institutions to routines and rules, whereas arrow B translates routines and rules in to actually actions. These actions are repeated and reproduced and slightly change over time as shows arrow C. Eventually it might be that there is a shift and new rules, actions, routines and institutions are formed as shown by the dotted line C.
Realm of action
Institutional Realm
a
b c
Key: a=encoding, b=enacting, c=reproduction, d=institutionalisation b c c b Routines Rules d a b c b c c b Routines Rules d Time
30 This framework focuses on the fundamental characteristics of the process of change and is relevant for this research, because it gives information and insights on how actions are linked to stability and change. Because the implementation of a PMS (new rules) is a form of change, this framework is important to understand how this change might influence routines and eventually institutions, because eventually managers want to see their new PMS to become institutionalized. However, this framework tells not how changes should be implemented and what the reaction of employees is. Although Burns & Scapens (2000) do mention that: “a change that is consistent with the existing routines and
institutions will be easier to achieve than change which challenges those routines and institutions”.
Also the focus of this framework is still on groups instead of individuals. So to understand how an individual person reacts to change we need to look at how actions are established at an individual level. This will be done in the next paragraph.
5.4 Individual behavior
The framework showed before (figure 8) has its focus on groups of people. However it might be that certain individuals behave differently than the group. So to see the effect of the implementation of PMSs on behavior it is necessary to see how actions and behavior are established among individuals. This will be done by looking at scripts, schemas, cues and sense making. The main article that is used will be that of Gioia & Poole (1984).
5.4.1 Schemas and scripts
Gioia & Poole (1984) describe a schema as: “a generalized cognitive framework that an individual
uses to impose structure upon, and impart meaning to, social information or social situations in order to facilitate understanding”. This looks much like the description for institutions but on an individual
level. Furthermore “a scripts is a schema held in memory that describes events or behaviors
appropriate for a particular context” (Gioia & Poole, 1984). So scripts are schemas that are actually
performed and where actions of an individual can be seen. These scripts can be acquired in a direct and indirect way. Direct script acquisition is by experience with other people, events or situations and indirect acquisition by means of communication or media (Gioia & Poole, 1984).
31 For scripts to be actually turned into action several preconditions must be present (Gioia & Poole, 1984).
1. A person must have a cognitive representation of a particular script (i.e. a person must have cycled before).
2. A context or situation eliciting the script must be experienced (i.e. the person has a bike and wants to use it).
3. The person must perform the script (i.e. the person actually takes the bicycle for a ride).
In the situation described above there is a strong script present. However, it could be that a situation is new to a person and he/she does not know how to act. Then there is no script present that can be recalled and the person is aware of the situation he/she is in and should think about what to do instead of automatically recalling a script and deal with the situation automatically. Figure 9 describes the development of scripts. The more frequent a typical situation occurs the stronger the script is developed and recalled.
Controlled Processing
Low High
Typicallity of the situation Novel Situation Infrequent Situation Conventional Situation
Partially Stereotypical
Situation Stereotypical Situation
Automatic Processing
Unscripted behavior and action Script development cued Conscious performance of expected behavior Weak protoscript modified and performed
Strong protoscript performed mindlessly
Figure 9: A continuum of script development (Gioia & Poole, 1984).
32
Scripts at T1 Scripts at T2 Scripts at T3
T1 T2 T3 Realm of action Institutional Realm a b c d a b c d a b c d
Key: a=encode, b=enact, c=replicate or revise, d=externalize and objectify Figure 10: Studying the links between action and institution (Barley & Tolbert, 1997).
Routines can thus be seen as a collection of individually performed scripts. Most likely a group of people within a company is influenced by the same institutions and individually they produce their own scripts accordingly. When all individuals within this group have developed the same scripts they make up for the routines of the group. So again there is a duality and relation between institutions and actions through scripts.
5.4.2 Cues and sense making
For a script to be used it first has to be triggered; there has to be a situation that calls for a script. For example you can hear a phone ring (trigger) and you bring up your „phone-script‟. This situations that trigger scripts are called cues (Becker, 2004). However cues can also be involved in a situation where no scripts are available; an alarm sound goes off (trigger/cue) and you have never heard it (no script). And because there is no script to fall back on, this creates uncertainty and makes that people are very aware of what is happening.
33
5.5 Resistance to change
When implementing a new PMS it can be a totally new system demanding employees to completely change their way of working, or a system that is in line with existing routines. Therefore before resistance is addressed it is necessary to look at the degree of change, because the degree of resistance is linked to this. Also change may come in many forms and through different channels and this will also be addressed. Bartunek & Moch (1994) and Burns & Scapens (2000) have classified the process of change into different forms and orders. It is important for management to know what the impact of the implementation of a new PMS has on existing rules, scripts, routines and institutions. They have to classify their implementation process to factors presented below to realize what they can expect with regard to resistance.
5.5.1 Degrees of change
Bartunek & Moch (1994) have categorized these degrees of change in three orders. The first degree of change is incremental and in line with existing schemas and scripts. Small changes are made to overcome problems, but these changes have no or only a small effect on existing routines. The second degree of change is more radical and involves changing existing cognitive frameworks and thus existing schemas and scripts. Finally the third degree of change involves making employees aware of their personal limitations and schematic behavior. By doing so employees are made aware of why they are doing things a certain way and together they try to evaluate and where necessary change schemas and scripts.
When management wants to implement a new PMS they need to look at the degree of change. When a new PMS is in line with existing routines, resistance will be less than when this PMS radically changes the way employees need to work and thus changes their existing routines and institutions (Burns & Scapens, 2000). The best way of implementing change is by using the third degree. Because by doing so employees are made aware of the necessary changes, see their own limitations and understand why they need to adapt. However this is difficult to achieve, because is it hard to name schemas since they are not tangible (Bartunek & Moch, 1994).
5.5.2 Forms of change
Change can come in different forms and Burns & Scapens (2000) classify three different types of change. They make a distinction between formal and informal change, revolutionary and evolutionary change, and regressive and progressive change.
“Formal change involves the introduction of new rules and/or through the actions of a powerful individual or group (p.18).” This can require that new scripts/routines have to be developed. Whereas
34 conditions. If the formal process is ahead of the informal process this might lead to resistance and failure of the implementation process.
Furthermore with revolutionary change there is a fundamental change in existing routines and institutions as with evolutionary change there is only a small change in these routines and institutions. Finally there is the distinction between regressive and progressive change. Regressive change is used to describe behavior that strengthens ceremonial dominance. There is an existing structure of dominance and changing the rules does not change this structure. On the other hand progressive change involves changing ceremonial behavior by instrumental behavior. New rules, best available knowledge and technology are applied to problems and try to intensify relationships; it tries to overcome and restructure existing structures of dominance.
5.5.3 Resistance
As mentioned before in the topic about the implementation of PMSs (§4.4.1) the main reason for the implementation process to fail is due to resistance to change by employees. “Resistance to change is
considered as a behavior to protect an individual from the effects of real or imagined change, it serves to maintain a status quo, seeks to challenge, disrupt, or invert prevailing assumptions, discourses and power relations” (Fiedler, 2010). In the case that a new PMS conflicts with existing routines and
institutions it is most likely that resistance is strong and Burns & Scapens (2000) provide three reasons for this resistance:
Formal and overt resistance due to competing interests;
Resistance due to a lack of capability to cope with such change; and
Resistance due to a „mental allegiance‟ to established ways of thinking and doing, embodied in existing routines and institutions.
To manage this resistance to change there is a lot of literature with checklists and recommendations. However as Fiedler (2010) mentions, most of this literature have little systematic structure where the focus is on content-related activities instead of management activities. Fiedler (2010) distinguishes different phases that should be kept in mind by management to overcome resistance:
Identification and evaluation of resistance potentials, Planning of managing resistance,
Avoidance/promotion of and preparation for resistance, Resolution of resistance,
35 Also Fiedler (2010) mentions that the process of overcoming resistance is subject to contingency factors. For every company and every implementation process different factors play a crucial role in the degree of resistance. One way to address and assess these degrees of resistance is to form a group of people from different parts of the company to guide the implementation process and to manage resistance. This will be the focus of the next chapter where the influence of networks within the company will be discussed.
5.6 Actor-network theory
We have seen that to design and implement a PMS it is important to overcome resistance created amongst employees. This can be done by creating a network around the system and its process of implementation. As Harty (2010) states: “The diffusion of innovations (a new PMS) argues that it is
networks of individuals and organizations which facilitate or inhibit adoption, through sharing or blocking know-how about and experiences of an innovation”. One theory concerning this creation of
networks is the actor-network theory (ANT).
ANT states that within networks both human actors and technological actors are situated in specific socio-technical settings (Harty, 2010). These actors interrelate and this interrelationship gives form to the network. Also “the process of translating an innovation implies that interactions are created
between actors who make alliances in order to pursue some goals rather than others in the change process” (Alcouffe et al., 2008). For a network to be successful it is important that the network grows
and becomes stronger. This is especially true for an implementation process of a new PMS. Many actors need to be involved to make it successful. To create and make the network grow Alcouffe et al. (2008) determine four processes that interact with each other:
Problematization; actors within the network try to convince others of joining them by giving solutions and explanations why they have the best solution.
Interessement; different interests of stakeholders are determined and the links between these different interests are strengthened.
Enrolment; roles are defined and divided amongst the stakeholders to structure and expand the network.
Mobilization; monitoring the different interests and links to keep the network stable and to try to mobilize other actors within the company.
36
5.7 Summary
The second part of knowing how to successfully implement a PMS is to understand how employees behave and react to the process. It should be clear how behavior is established within groups and individuals before this behavior can be directed and guided.
At a macro level behavior is influenced by institutions which are systems of social rules that structure social interactions (Hodgsons, 2006). There are three different approaches on how these institutions are developed and influenced (§5.1). The New Institutional Economics states that the behavior of people is rational and efficient and that they conduct the sort of behavior that is the best alternative out of all possible possibilities. This has its pitfall because we as humans do not have the mental capability to oversee all alternatives and possibilities. New Institutional Sociology says that behavior is influenced by the environment in which the people operate. People act in a way that is desired by actors within their environment. The last one –Old Institutional Economics– focuses on both rational behavior and the environment. It states that behavior and actions are influenced by interpreting signals from the environment. These signals can be (written) rules within a company to which employees in a company act. And the actually execution of these rules form the routines of the group of employees. In §5.3.3 a framework is presented which gives an overview of interactions between these rules, routines and institutions.
The only problem is that these approaches focus on groups of people instead of individuals. To understand how individuals develop behavior; scripts, cues and sense making are introduced. Scripts (§5.4.1) are cognitive frameworks held in memory that describes events or behaviors appropriate for a particular context (Gioia & Poole, 1984). These scripts can vary from a very weak to a very strong script depending on how often the script is executed. A strong scripts leads to automatic processing of the current situation/problem and is an efficient way to deal with our limited mental capabilities. Before these scripts come in action they have to be triggered by a cue. A cue is a specific signal in a specific situation that triggers a certain script to be executed. However it might be that with a specific signal there is no script present to deal with the situation. This is called sense making, because the individual cannot deal with the situation automatically through a script and tries to overcome his/her uncertainty on how to deal with the situation.
38
6. Analysis
To understand the process of the implementation of a new PMS and to come to the end result of desired behavior from employees, literature described before needs to be linked together. This will be done by examining a framework (figure 11) in which the connections between the different parts of the literature will be explained. This framework will be a tool of guidance for managers who want to implement a new PMS or are already involved in an implementation process. It should help to provide insights on what the path of an implementation process is and what factors play a crucial role in making the process a success. This framework will be described and explained in the remaining part.
6.1 Framework
Agent-Principal problem Develop a new PMS New cues Network implements Establish a network Prinicpal Network Performance Measurement System Scripts and Routines Institutions Awareness New rules Lead to New scripts Strong scripts Create Become Routines of the group Institutions of the group New institutions Institutionalized PMS Lead to Become Influence Are adjusted toOvercome resistance throughout the process
Time Cues
Desired behavior Leads to
Stabilize and expand the network
Time
Leads to
Provide information
Figure 11: The process of implementing a new PMS.
