Deze samenvatting is gebaseerd op het studiejaar 2013-2014.
Accounting innovations are often not successfully implemented or spread throughout an organization. One reason is that they are designed to facilitate decision making of top management but then lower level managers have to use the same systems for their decision making. If an accounting system is used for decision control, wealth effects can occur, so that managers who’s wealth is negatively affected resist accounting innovation. Delegation of decision rights can limit that resistance by creating the right environment and by involving subunit managers.
Management accounting systems (MAS) are implemented to improve firm performance and decision making but that does often not happen. Reasons for failure are sought on the organizational level but less attention is paid to attitudes. The authors argue that delegation of decision right determines managers’ acceptance. MAS are used to determine whether the managers use the decision rights optimally so managers will be concerned if new systems are implemented because they can directly influence their evaluation and rewards. So there might be a negative effect of delegation decision rights, but two conditions may reduce that. First, when production managers are involved in designing the MAS. Second, when subunit managers are empowered to implement changes that enables the unit to adapt to new signals provided by accounting information. So these are the mediating variables.
The two functions of MAS are:
Facilitating managerial decision making (of the subunit managers)
Control subunit managers’ behavior by superiors (so support the superior’s performance evaluation decisions)
This study seeks to study acceptance of MAS innovation (new or redesigned MAS). Intervening path models are used.
Decentralization, subunit adaptability, and acceptance
Decentralization enables managers to take actions so that the adaptability of the subunit is enhanced. Changes in the accounting systems provide new information. The superior will use that to assess and reward performance, subunit managers just have to deal with it. If they cannot use the new information to improve performance they are likely to be commensurate effect on their promotion and compensation. Absorptive capacity is the ability to use information. It depends on the ability and willingness of the manager to use the information. Decentralization thus has a positive relation with adaptability of the subunit. That will in turn increase acceptance.
Decentralization, user influence, and acceptance
The user influence on the design of the system will determine the acceptance. Two reasons lie behind this expectation. First, systems should be tailored to the needs of the user. The authors argue that involvement in the design and implementation of MAS leads to the system being better tailored to the needs of the subunit managers. Ex ante uncertainty is reduced and decision making improved. Second, measurement error in the system is reduced. So uncertainty of performance outcomes is reduced and commitment to the system increased.
Consequences of accounting innovation acceptance
Effect on the satisfaction of the user with the information provided
Effect on overall subunit performance
The expectation is that user acceptance improved performance. That is a direct relationship, the possibility for an indirect relationship operating via satisfaction is also considered.
Length of time the system has been implemented into the subunit
Length of time the system has been implemented into the industry
Random sample of production managers of manufacturing business units in Australia
83 firms responded
Measurement instruments from previous research, sometimes adapted/developed
Adaptability of the subunit is positively related to the level of decentralization
Adaptability of the subunit is positively related to managers’ acceptance
Influence on MAS design is positively related to the level of decentralization
Influence on MAS design is positively related to managers’ acceptance
MAS acceptance and decentralization of decision rights have no significant relation
MAS acceptance is positively related to user satisfaction
No significant link between MAS acceptance and performance
User satisfaction is positively related to performance
Only industry affects the results; higher levels of acceptance exist in more complex industries
Influence costs: when the delegation of decision rights enables subordinates to manage their superior’s expectations regarding subunit performance. That did not have an influence.
Discussion, future research, limitations
Individual responses to MAS innovations will be influenced by how the individual perceives the potential wealth redistribution effect of the system. If the subunit manager can adapt as a response to the new information provided by the system he is more likely to accept it.
The study shows that elements of the environment in which the user of the MAS is operating has its influence.
Organizational level factors are not unimportant, but they are more likely to influence individual behavior through their impact on the operating context at lower levels. Further research should integrate those levels.
Co-variation (threatens internal validity)
Potential measurement error
Accountant typically seek to reduce uncertain situations into a framework with well structured tasks and highly measurable outputs. Subordinates can try to distort the information system to their desired ends when they operate outside the framework.
Accounting research evolved from budgetary control to a broader organizational view. This paper deals with behavioral responses to the information system. Three questions:
How is management expected to use accounting?
What are the various inter- and intra-organizational relations affecting the use of accounting?
How do members of the organization attempt to utilize the information system to their own ends?
The first question is viewed normative and there are three relevant activities: score-keeping, attention directing, and problem solving. However, interdependence exists which makes that difficult to use in the organizational context. A solution would be to enact an information system that is capable of resolving the problems and permitting the manager to behave as if the various forms of interdependence do not exist. That is an idealized system:
Suggests that accountants should develop techniques appropriate to the existing task environment
Argues that if we look long and hard enough, analytic solutions to the interdependencies are attainable
Behaviors of managers may include:
Affect the outputs of the information system (select the accounting policy to smooth reported profits)
Select those actions that will reflect their behavior in the best possible manner (gaming the indicator, for instance)
Thompson and Tuden made a model of forms of decision-making. There are two dimensions; the preferences about possible outcomes and the beliefs about causation, 4 outcomes:
Preferences certain and beliefs certain: decision by computation
Preferences uncertain and beliefs certain: decision by compromise
Preferences certain and beliefs uncertain: decision by judgment
Preferences uncertain and beliefs uncertain: decision by inspiration
We often think we know what the task is because we make simplifying assumptions to make situations analyzable. Perrow established a framework on technology variables, using two dimensions: extent to which the task is analyzable and degree to which activities are exceptions.
Unanalyzable, few exceptions: craft
Unanalyzable, many exceptions: non routine
Analyzable, few exceptions: routine
Analyzable, many exceptions: engineering
The two models together suggest a relation between task and control. If the people believe they understand the task and have modeled it and if the goals are known then the control process would have plans based on computation and evaluation based on comparison to standard. If the task has more exceptions, constructing and output measure and controlling the process becomes more difficult. When the task is not analyzable, the control system focuses on the aspects believed to be important (surrogates). The following picture is provided: See fig 3
The systems of control in this model can also be characterized as administrative, self, or social control. Quadrant one will rely on administrative control and the bottom row more on self and social control.
Ouchi’s model is discussed as well, with the important notion that goal incongruence might exist.
Researchers and accountants seek to reduce all decisions to situation which are highly analyzable and consistent, but that is just not possible. Still, in the research the ‘cell 1 characteristics’ are assumed. It assumes output measures to be equal to the task.
Budgetary control process: simple and rational, a superior’s set of standards is communicated to a subordinate and used for performance evaluation and setting the next budget. Higher standards lead to better performance and low standards reduce aspiration level and subsequent performance. But standards should not be too high, that decreases motivation again. Participation was offered as solution for setting the right budget standard.
Arguments supported that participation produces:
Higher job satisfaction and self actualization
Positive attitudes toward the job and company
Motivation to achieve the budget
Evidence for differences between open and lateral and hierarchical participation was not found, as well as improved performance. Some even noted a dysfunctional effect; performance leading to gaming and slack. That is not surprising because participation is a complex variable. Also, pseudo-participation often occurs. But the overall attitude about participation is still positive.
It states that subordinates will expend effort when they believe their actions will result in a certain outcome which will provide them with intrinsic and extrinsic satisfaction. The relevance to the budgeting process is the recognition of varying kinds of rewards that motivate subordinates, their relationship to the internalization of standards, and the conflict that may occur if a subordinate feels its rewards threatened by the budgeting process. Conflicting results were found. Limitations of the expectancy theory lie in identifying and measuring specific parameters.
Some researches indicated possible dysfunctional effects of budgeting systems. This research does not focus on whether they should be used but how. Evaluation styles that use standards loosely resulted in little or no dysfunctional effect. A rigid style of budget evaluation resulted in higher tension, ambiguity, and manipulation of accounting reports. In the long run that will decrease organizational efficiency. However, others say that a rigid style will lead to higher budget accuracy and not to tension and ambiguity. Still, these researches made two important contributions:
Evaluation style as significant variable in the control process
It observed some possible dysfunctional effects but noted that these do not always occur, which supports contingency theories of management accounting
Recent literature has adopted a broader view of control, moving out of cell 1. A contingency view on accounting take into account technology, the environment, managerial style, and structure. Ouchi suggests two variables: knowledge about the transformation process and availability of output measures. However, there is no systematic research on the contingency orientation so there are no concrete conclusions.
Agency theory is also an useful approach. It describes superior-subordinate interaction and addresses control and motivation.
Another new view is the view on organizational rationality. Organizations must be highly adaptable, flexible, and able to create variety to cope with the changing environment. Budgets may be an ex post rationalization of actions rather than an ex ante statement of organizational goals. The ‘new organizational reality’ argues that flexibility in control systems is necessary.
We will now discuss six broad categories of behavior that can arise when a control system in appropriately assumes a cell 1 world
Affect the natural or preplanned flow of data without altering the actual activities. That could result in the information system accelerating a message and sending it now when it actually does not occur until some future date or vice versa.
Select from a set of possible messages a signal that is likely to be accepted and most favorable to the manager. That usually occurs when managers have to provide estimates of future events. It is often called ‘creative accounting’ when it is applied to past activities.
When certain aspects of the information set are either highlighted (enhanced) or hidden (degraded). Evaluation based on multiple criteria are more susceptible to enhancing and degrading. It can even go as far as enhancing something that was not even considered when the course of action was selected.
This form of manipulation is most commonly discussed. Smoothing, biasing, and focusing (discussed before) are strategies intended by the sender to manipulate the recipient by affecting the set of data available to the recipient. The message is credible when the recipient does not doubt its validity and desirability. The sender (subordinate) is exploiting aspects of the superior-subordinate ‘game’ to send messages to his own advantage. The on the job behavior has not changed! However, gaming refers to those behaviors where the sender through his job related acts causes the desired message to be sent. The superior sets the rules of the game and the subordinate acts. Gaming thus exists when the superior uses a surrogate measure of performance rather than the principle. This is the ‘moral-hazard’ issue. ‘Hardness’: the performance measure is verifiable, justifiable by an explicit set of rules (few and uniquely defined for the setting).
Data can be filtered so that more desirable elements of the set are communicated. A variant is to delay the report until it is too old to be used. Three other strategies are:
Over-collection: obtaining more data than is needed
Over-presentation: give loads of information to confuse the recipient
Aggregation: aggregate until critical aspects are lost
Violating laws or the organization’s rules.
An example of resource allocation is used to explain which behavior might occur when. In cell 1 some gaming behavior might occur in selecting projects that meet managers’ criteria. In cell 2, a successful submission might be possible but it is more difficult to be sure that they will have the desired effect (the managers know the questions but cannot evaluate the answers). Cell 3 contains the strategies that exploit the manager’s inability to be sure they are asking the right questions. In cell 4 all behaviors may be exhibited. But it applies more to the evaluation of social programs and other public sector resource allocation decisions.
See fig 5 for more illustration
This paper focused upon the role of the accounting information system as a link in the organization. Managing that system is part of the manager’s job because it communicates the history and is involved in the incentive system. Managers try to manage the system in different ways:
Active: influence the message sent without having behavior affected by the system
Responsive: select actions from available set, the trade-off is between how the system represents his actions and their payoff to him
If the situation is not highly analyzable, readily amenable to measurement and if the goals are known, there is slippage between maximizing the indicator and its principle. The manager exploits this slippage to his benefit. So the information system may give power.
This influences resource allocation. If the data are measurable and verifiable and the task analyzable, the traditional approach is relevant. If the data is more subjective, subunits bias their estimates and game the system. If the process is not analyzable, managers focus on the strength of project and try to filter out weaknesses.
The authors study the factors that influence the development, use and perceived benefits of results oriented performance measures in government activities. Measurement development and use are significantly positively influenced by organizational factors like top management commitment to the use of performance information. Technical things like information system problems and problems of selecting and interpreting appropriate performance metrics in activities that are hard to measure influence system implementation and use. Performance measurement and accountability are positively related with greater use of performance information for various purposes. But mandated performance initiatives do not have more perceived benefits if measurement and accountability increase.
Governments are often the first to implement new, more strategic performance measurement systems. Measurement system development and use is positively influenced by organizational factors such as:
Top management commitment to the use of performance information
Extent of decision making delegated to users of performance information
Training in performance measurement techniques
Technical issues that may impede measurement system innovation are:
Difficulties selecting and interpreting appropriate performance metrics in activities that are hard to measure
These technical issues play a bigger role in the implementation of performance measurement systems than in cost system implementation.
Background and hypotheses
Performance measurement initiatives in the US government
The US government promoted a performance based approach to management and accountability of federal activities during the 90s. The goals were:
Increase Congressional oversight and increase accountability for achieving results
Enhance performance based decision making by implementing information systems that supplement traditional input oriented performance measures with measures focused on results and the achievement of strategic objectives.
The Government Performance and Results Act (GPRA) requires government managers to clarify missions and objectives and to measure relevant outputs, outcomes, and service levels so that performance can be evaluated. The thought behind it was that efficiency and effectiveness would be improved by increasing accountability.
Determinants of measurement system implementation and success
Factors that can influence implementation and success of performance measurement initiatives:
Ability of existing information systems to provide required data
Extent to which organizations can define and develop appropriate measures
Decision making authority
These will be discussed now.
Information system capabilities
Organizations that have higher quality information systems might implement new measurement systems more easily because measurement costs are lower. So the higher the current information system capabilities the higher the implementation success. But managers who are satisfied with the information of the existing system might reject investing resources in a new system, leading to a negative relationship. Small-sample field studies in the government sector suggest that information system problems are caused by the need to use data collected by other organizations so that the quality and accuracy cannot be ensured.
H1: Performance measurement development and outcomes are negatively associated with problems obtaining necessary data in a reliable, timely, and cost effective manner.
Selecting and interpreting performance metrics
This issue concerns the ability to define and assess metrics that capture desired actions and outcomes. Problems in the public sector are often the result of:
Complicated interplay of federal, state, and local government activities and objectives
Aim to influence complex systems of phenomena whose outcomes are largely outside government control
Missions that make it difficult to establish measurable outcomes
Difficulties in measuring many dimensions of social welfare or other governmental goals
H2: Performance measurement development and outcomes are negatively associated with difficulties selecting and interpreting appropriate performance measures
Organizational factors have a bigger influence than the above technical factors. Top management support is essential because these managers can focus resources, strategies and goals on objectives they find important, deny resources to innovations they do not support and provide political help. Top management support also influences employees’ perceptions because if they support it then the employees are also more likely to support it. The employees then feel accountable for the results and use the information for decision making.
H3: Performance measurement development and outcomes are positively associated with management commitment to the implementation and use of performance information.
Decision making authority
Next to researches, economic theories support the link between decision making authority and system implementation and results. But the requirement for authority creates a potential threat to increased accountability in governmental organizations where bureaucratic rules, laws, and the separation of powers among different branches of government can place constraints on authority and therefore the extent to which they can be held accountable for results.
H4: Performance measurement development and outcomes are positively associated with the extent to which managers have the authority to make decisions based on the performance information.
The extent to which training and resources are provided to support the implementation influences the implementation and results of performance measurement innovations too. It allows articulation of the link between the new practices and organizational objectives, prevents employees from feeling overwhelmed and pressured, and makes employees understand, accept, and feel comfortable with the innovation. It also shows management support and that the organization is providing adequate resources.
H5: Performance measurement development and outcomes are positively associated with the extent of related training provided to the manager.
This organizational factor is particularly relevant to governmental organizations. It concerns whether or not the innovation is implemented as a responsive to legislative requirements. Institutional theory says that organizations gain legitimacy by conforming to external expectations to appear rational, modern, and efficient. But then they separate their internal activities from the externally focused symbolic systems. External bodies may have the authority to impose practices so that the subordinates’ units remain eligible for funding.
H6: Performance measurement systems that are implemented to comply with the GPRA’s requirements are positively associated with performance measurement development, but are not associated with greater accountability or use of performance data, or with the perceived benefits from GPRA implementation.
Measurement system development and system outcomes
H7: Performance measurement system development has positive direct effects on system outcomes, as well as indirect effects through the level of accountability for the results.
Survey distributed to 1300 middle- and upper level civilian managers
528 used for data analysis
Control variable for differences between senior and lower level managers
Second control variable perceived GPRA benefits
H1 rejected, data limitations are even positively associated with the use of performance information. That may be because managers do not experience problems with information systems and data collection until the information is being used for decision making.
No differences between senior and lower level managers.
Performance measurement development and accountability are hindered by:
H1: measurement, hypothesized + confirmed
H2: accountability, hypothesized + rejected, found - significant
H3: data limitations, hypothesized – rejected, found + significant
H4: metric difficulties, hypothesized – confirmed
H5: commitment, hypothesized + confirmed but not significant
H6: authority, hypothesized + confirmed
H7: training, hypothesized + confirmed
GPRA pilot sites have developed performance measures to a greater extent to meet the requirements but do not make use of the information. That is consistent with the hypothesis about externally mandated control systems. After controlling for the GPRA implementation efforts the use of information was also supported. Greater measurement and accountability are positively associated with the use of performance information for decision making but do not really influence managers’ perceptions of the benefits from complying with the GPRA. Some of the technical and organizational effects factors can have interactive effects on performance measurement system implementation and outcomes.
Not all potential factors included
No information on the target setting process or level of target achievability
The systems may not have been in place long enough to perceive all the benefits
Management control systems (MCS) are an important element in enhancing innovation. This paper examines the adoption of MCS in product development. Seven different systems are examined:
Reports comparing actual progress to plan
Budget for development projects
Project selection process
Product portfolio roadmap
Product concept testing process
Project team composition guidelines
Three research questions:
What are the reasons for adoption of these systems?
Are these reasons associated with differences across companies in the time to adoption?
Are these reasons relevant to performance?
Formal MCS have traditionally been associated with mechanistic organizations. Recent theoretical developments offer various concepts that support the need for formal MCS in uncertain settings. Forward looking efforts typically associated with MCS complement fast reaction to new information to improve how organizations deal with uncertainty. Recent empirical evidence indicates that innovation processes thus may gain from the presence of MCS. Innovation has a pivotal role in the product development process and this paper examines that process. The objective of learning about the adoption of MCS required the early-stage criteria.
MCS are formal systems particular to product development including the seven systems mentioned before. The paper highlights the difference between the reasons for adoption of MCS in innovation processes and the objectives that the systems pursue. The research combines qualitative data in the form of interviews and quantitative data in the form of questionnaires. It triangulated data using three different informants per company. The second research questions examines how the six different reasons for adoption influence the3 speed of adoption of product development systems. Speed of adoption is measured using time elapsed from the start of the company to the time reported for the adoption of the project milestones’ system. It is found that reasons for adoption associated with manager’s background; chaos; legitimize; need to focus; and contract are significantly related with faster adoption compared to companies that rely on informal product development control systems. Also when MCS are adopted because of the manager’s background, product development performance is enhanced.
Six reasons for MCS adoption:
Need to focus
Confirms MCS roles:
Stimulating dialogue and idea creation
Controlling execution through diagnostic systems
Stabilizing an environment that is already rich in opportunities
Manager’s background is associated with the fastest time to adoption
MCS adopted because of manager’s background or code learning is associated with better on time development (which is an important dimension of product development performance)
An important note is that the reasons why systems are adopted are distinct from the roles that the systems ultimately play.
First, MCS were seen as constraining innovation and therefore incompatible with it. But authors now see the benefits of MCS on innovation. It was said that MCS can stifle innovation if not designed to deal with uncertainty. Mintzberg separated planning (rules by formal systems) and managing (informal and decoupled from MCS). Perrow says that certain MCS are most useful when task analyzability is high and the number of exceptions is low. As said, recent empirical evidence says that innovation management appears to benefit from having a balance between tight and loose controls to provide both the support and direction for innovation.
Simons’ typology identifies interactive systems as information-based routines to identify knowledge required to address strategic uncertainties. It allows top management to guide the search stage of the innovation process, without falling into the cybernetic model. Another concept is enabling bureaucracy, which is designed to enhance the users’ capabilities and to leverage their skills and intelligence. So organizations assimilate and exploit the knowledge accumulated in the first stage through flexible, transparent, user-friendly routines. Weick and others described adaptive routines, which are resilient because of their capacity to adapt to unexpected events. It portrays routines as flexible to absorb novelty. Routines provide the background for improvisation to happened and learning to accumulate. These concepts show how MCS can have a positive effect on innovation. MCS are viewed as flexible and dynamic frames adapting and evolving to the unpredictability of innovation, but stable to frame cognitive models, communication patterns, and actions.
Three types of control systems (input, behavior, and output control) enhance radical innovation.
Emergence of MCS
Greiner’s growth model describes a first transition (crisis of control) where MCS are adopted. At the end of the first growth phase, informal management no longer works and MCS are required. It is found that the co-variation between tome to adoption and organizational variables is mediated by events (reasons for adoption) that are absent in such studies. It seems that faster adoption is relevant to organizational performance. But that does not address whether different reasons for adoption are associated with faster adoption and not whether these reasons are related to performance.
Reasons and roles of MCS
Sometimes, the reason for adoption is unrelated to a particular role. Also, the same reason for adoption can lead to the adoption of MCS with different roles across companies. Existing literature identifies the following MCS roles:
Make goals explicit and stable
Code learning from past
Plan the sequence of steps
Promote accountability and facilitate control
Contract with external parties
Symbols to legitimize
Cross-sectional, multi-method multi-case field research design is used. The aim is to gather a large enough variation to probe the research questions, to capture the detail required to answer the questions, and to link contextual variables to the adoption of MCS. Three information sources are used: public data, three questionnaires, and three interviews.
Project milestones are the system most companies have adopted and the fastest to adopt. Time to adoption is significantly different from the other systems. Project selection process is adopted significantly slower than all other systems except product portfolio roadmaps. These systems are likely to be adopted later because they require having various products considered in the development plan.
A time-related measure is most frequently used with 62/69 companies referring to it. Budgets and product functionality related measures are mentioned by 30 and customer-related measures by 24 and quality-related measures by 15.
As said before, there are six reasons to adopt MCS:
Six reasons for MCS adoption:
Legitimize (vis-à-vis external partners, as symbol to enhance the credibility of the company towards external parties)
Contract (when inter-organizational agreements lack the constant interaction required to ground informal management and they need to formalize this interface)
Manager background (mimetic behavior; they emulate practices form other organizations to reduce the cognitive uncertainty)
Need to focus (responds to a particular emerging need)
Chaos (reactive to unexpected events, mistakes, or recurring problems)
Learning (outcome of enactment process; formal systems emerged as the outcome of a learning process)
There were also organizations that had not yet adopted MCS. There, the crisis of control described by Greiner had not yet happened. These organizations stated the following reasons for maintaining an informal approach:
Team has worked together for a long time and informal interactions are well-understood but not coded.
Management team believes that formal systems would kill creativity
Not large enough to grant MCS
Management team does not have knowledge to implement MCS
No one-to-one relations between the reason for adoption and the role of the system was found. Only the reason legitimize was related to the role symbols to legitimize and the reason contract to contract with external parties. But these roles were also related with other reasons.
Results for time to adoption
All the reasons for adoptions are typically triggered by an event, so adoption depends on whether that event happens and there is no clear directional expectation. Some predictions can be made:
Companies with an informal approach will report adopting systems later than the rest of the companies in the sample.
Certain events happen early in the life of companies (learning likely because of no event external or proactive internal that triggered MCS, so later adoption may be associated with the learning adoption reason).
Some results that were found:
Manager background, chaos, legitimize, need to focus, and contract are associated with significantly faster adoption than the reference group with an informal approach.
Different reasons for adoption do not lead to faster adoption (except when compared to the informal approach) which may be due to the fact that events that lead to adoption happen randomly. Or it may be that 69 companies is not enough to unveil differences
Only learning is not significantly associated with faster adoption of the project milestones’ MCS.
Results for reasons for adoption
If the reason is manager background the company performs significantly better (on-time development) than companies that adopt because of chaos or need to focus.
Learning by doing companies perform better than contract and chaos.
Need to focus companies perform worse than informal management firms and not significantly better than any other reason.
Informal management firms perform better than contract, need to focus, and chaos.
Need to focus
Only background and learning have a positive coefficient. So informal management is not necessarily bad an my outperform formalized systems. The question is whether that is really the case or that the underperformance reflects the limitations of the informal approach that led to the adoption of these MCS and that those companies are still learning how to use them. When systems are adopted as a reaction to crisis and problems, performance is worse than other categories.
With regard to the manager background, the import-in notion is important. Key individuals may perceive a gap and hire a manager who brings in (imports-in) a new MCS. That is more likely with early-stage companies than established companies as they have the full set of functional capabilities from day zero.
Limitations and extensions
Assumption of causality (perception of management)
Reasons for adoption emerged from data, possibly other reasons too
Sample in two main industries
No longitudinal observation
Adoption of MCS is treated as an event while they may be growth and refinement
Focus on product development, nothing on marketing, sales, HR, finance
Future research may research whether these reasons also exist in established companies to evolve their systems
The companies that do not survive are an interesting research sample
Further research on the consequences of not adopting MCS is a useful complement
Agency theory offers unique insight into information systems, outcome uncertainty, incentives and risk. Also, it is an empirically valid perspective, particularly when coupled with complementary perspectives.
Four questions are addressed in the paper:
What is agency theory?
What does agency theory contribute to organizational society?
Is agency theory empirically valid?
What topics and contexts are fruitful for organizational researchers who use agency theory?
Origins of agency theory
Agency theory developed out of risk sharing. It occurs when cooperating parties have different goals and division of labor. It is directed at the agency relationship, where the principal delegates work to the agent. Two problems can arise:
Agency problem: goals/desires conflict and it is difficult/expensive to verify what the agent is doing
Risk-sharing: principal and agent have different attitudes towards risk.
The ‘contract’ governing the principal-agent relationship makes assumptions about people, organizations, and information. The question is whether a behavior-oriented contract is better than an outcome-oriented contract. See table 1
It has developed in two lines, originating from information economics: positivist and principal-agent. They have a common unit of analysis and assumptions but differ in mathematical rigor, dependent variable and style.
Identifying situations where conflicting goals may exist and then describe governance mechanisms that limit agent’s self-serving behavior. It is less mathematical and the research normally focuses on large, public corporations. Two important propositions arise. First,
“When the contract between the principal and the agent is outcome based, the agent is more likely to behave in the interests of the principal”
That is because the rewards depend on the same actions. Increasing firm-ownership would be an example. Second,
“When the principal has information to verify agent behavior, the agent is more likely to behave in the interests of the principal”
That is because the agent will then realize that he cannot deceive the principal.
Positivist agency theory enriches economics by offering a more complex view of organizations. But it has been criticized as minimalists, tautological and lacking rigor.
This is a general theory of the principal-agent relationship. It involves careful specification of assumptions, followed by logical deduction and mathematical proof. It has a broader focus and includes more testable implications. The two streams are complementary. Positivist theory identifies various contract alternatives and principal-agent theory identifies which one is most efficient under varying levels of outcome uncertainty, risk aversion, information, etc.
The simple model assumes goal conflict between principal and agent, an easily measured outcome and an agent who is more risk averse than the principal (because agents cannot diversify their employment and principals can diversify their investments. Several cases are possible:
Complete information, the principal knows what the agent has done: behavior-based contract most efficient because an outcome-based contract would transfer risk to the agent
Principal does not know what agent has done: moral hazard (lack of effort agent) and adverse selection (misrepresentation of ability by agent) my occur. The principal can discover the behavior by investing in information systems or contract on the outcomes (but outcomes are not only result of the agent’s behavior). So it is a trade-off between the costs of measuring the behavior and the costs of measuring the outcome and transferring risk to the agent. The propositions:
“Information systems are positively related to behavior-based contracts and negatively related to outcome-based contracts”
“Outcome uncertainty is positively related to behavior based contracts and negatively related to outcome-based contracts”
“The risk aversion of the agent is positively related to behavior-based contracts and negatively related to outcome-based contracts”
“The risk aversion of the principal is negatively related to behavior-based contracts and positively related to outcome-based contracts”
“The goal conflict between principal and agent is negatively related to behavior-based contracts and positively related to outcome-based contracts”
The task also matters. Programmability is the degree to which appropriate behavior by the agent can be specified in advance. So:
“Task programmability is positively related to behavior-based contracts and negatively related to outcome-based contracts”
“Outcome measurability is negatively related to behavior-based contracts and positively related to outcome-based contracts”
“The length of the agency relationship is positively related to behavior-based contracts and negatively related to outcome-based contracts”
Agency theory and the organizational literature See table 2
Agency theory has several links to mainstream organizational theories and political models of organizations (see table 2). Differences agency theory and:
Political models: goal conflict is solved through negotiation, bargaining, and coalition (power mechanism of political science) and in agency theory through co alignment of incentives (price mechanism of economics).
Contingency theory: focuses on optimal structuring of reporting relationships and decision making responsibilities and agency theory on optimal structuring of control relationship resulting from these.
Organization control: risk implications of principal, risk aversion of agent, outcome uncertainty.
Transaction cost: arose from different tradition in economics, concerned with organizational boundaries. Agency theory focuses on the contract, regardless of boundaries. Also they have unique independent variables: asset specificity and small numbers bargaining (transaction cost) and risk attitudes, outcome uncertainty, and information systems (agency theory).
Contributions of agency theory
Importance of self-interest and incentives in organizational thinking
Importance of a common problem structure
Contribution to organizational thinking:
Information as a commodity (has cost, can be purchased) (board of directors is important information system; the richer their information the less likely compensation based on firm performance).
Uncertainty viewed as risk/reward trade-off.
Conglomerate mergers: not in interest of stockholders because they can diversify directly through stock portfolio but attractive to managers to diversify their own risk. Manager-controlled firms indeed engaged in significantly more conglomerate acquisitions and were more diversified.
Takeover bids: not in managers’ interest because they may lose their job. Managers who have substantial equity positions (outcome-based contract) are less likely to resist takeover bids.
Market discipline: long-run reputation effects of the market co aligned the short-run behaviors of the general partner (agent) with the limited partner’s (principal) welfare.
Boards and greenmail: boards that resisted greenmail had a higher proportion of outside directors and a higher proportion of outside directors executives (paying greenmail is not in the stockholders’ interest).
Executive holdings: executive security holdings were related to acquisition and financing decisions that were more consistent with stockholder interests. So executive stock holdings co align managerial and stockholders’ preferences.
Golden parachutes: use to co align executive and stockholders’ interests in takeover situations and they are seen as alternative outcome-based contract to executive stock ownership.
Cost of equity capital: employee stock ownership (outcome-based contract) co aligns interests of employees and stockholders.
So agency problems arise in situations where the interests of shareholders and top executives diverge. They can be solved through outcome-based contracts (golden parachutes and executives shareholders) or information systems (boards, efficient markets).
Vertical integration: difficulty of measuring outcomes positively related to using a corporate sales force (behavior-based contract) and not to manufacturer’s representative.
Commission (outcome-based) vs salary (behavior-based) compensation for salespeople in retailing: depends on task programmability, information systems and outcome uncertainty.
Replication former: information systems negatively related to performance-contingent (outcome-based) pay.
Transfer pricing: choice between cost (behavior-based) and market (outcome-based) transfer pricing mechanisms depends on decentralization (measure of task programmability).
So there is support for the link between contract form and information systems, outcome uncertainty, outcome measurability, and task programmability.
Recommendations for agency theory research
Agency makes contribution to organization theory, is testable and has empirical support. Five specific recommendations:
Focus on information systems, outcome uncertainty and risk: most unique contributions but very little empirical attention.
Key on theory relevant contexts: so situations in which contracting problems are difficult;
Goal conflict such that agent opportunism is likely
Outcome uncertainty to trigger the risk implications of the theory
Unprogrammed or team-oriented jobs in which evaluation of behavior is difficult
Expand to richer contexts: organizational behavior topics that relate to information asymmetry in cooperative situations and beyond pure behavior and outcome contracts.
Use multiple theories.
Look beyond economics: more mainstream empirical work.
Two extreme positions were taken: that agency theory is revolutionary and a powerful foundation; and that the theory addresses no clear problem, is narrow, lacks testable implications, and is dangerous. The conclusion lies in the middle. Agency theory provides a unique, testable perspective on problems of cooperative effort.
This paper establishes the performance management systems framework as a research tool for describing the structure and operation of performance management systems (PMSs) taking a more holistic view. The extended framework may be useful for those who want to study the design and operation of PMSs. It provides a template to help describe the key aspects of such systems.
There has been a tendency in research to focus only on specific aspects of control systems, rather than taking a more integrated and comprehensive approach. So the authors think that research would benefit from a framework that provides such a view on the key aspects of a MCS with a holistic overview but still efficient.
Management control systems
PMSs are seen as including all aspects of organizational control, including those included under the heading of MCSs. The classical view divides control between strategic planning, management control, and operational control. Management control is “the process by which managers assure that resources are obtained and used effectively and efficiently in the accomplishment of the organization’s objectives”. But there was no link between MCS and strategic planning and between MCS and operational control. Also, it encouraged a narrow view of MCSs because it concentrated on formal controls without setting them in their wider context. The authors define PMSs as “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”. So it consists of formal mechanisms and informal controls so it supports a broad range of managerial activities. A PMS can even support emergent strategies through its change and learning facilitation role.
Otley proposed a framework for MCSs before, which highlights five essential areas to develop a coherent structure for PMSs:
Identify key organizational objectives and the methods involved in assessing the extent to which these are achieved.
Formulating and implementing strategies and plans and the measurement and evaluation of this implementation.
Relating the process of setting performance targets and the levels at which such targets are sets.
Drawing attention to rewards systems used and to the implications of (not) achieving performance targets.
Types of information flows required to provide adequate monitoring of performance and to support learning.
Otley’s framework has some strengths:
Provides a helpful structure for analyzing MCS by focusing on 5 key areas and it can be used for profit and non-profit organizations.
It is general and therefore other frameworks can be used as a complement
Its application is straightforward, clear, and unambiguous and its questions are meaningful at different levels of management.
It facilitates the process of dealing with data
Otley’s framework also has some weaknesses:
It does not consider the role of vision and mission in MCSs where these may be key elements.
It focuses on diagnostic control systems, but all four levers of control should be considered.
It does not stress how accounting and control information is used by organizations, as against the existence of formal control mechanisms.
It takes a static perspective; ignoring the dynamics of control system development and change.
The interconnection between different parts of the performance management system are not explicitly addressed.
Simons developed a levers of control (LOC) framework, which is an action-oriented theory of control. There are four key concepts to his LOC, which are all directly controlled by a particular system or LOC:
Core values, controlled by the beliefs system (guides creative process of exploring new opportunities and instills widely shared beliefs).
Risks to be avoided, controlled by the boundary system (plays the negative, limiting role of circumscribing the domain where the company seeks new opportunities).
Critical performance variables, controlled by the diagnostic control system (monitors, assesses, and rewards achievement on key areas of performance).
Strategic uncertainties, controlled by the interactive control system (encourage organizational learning and the process of development of new ideas and strategies).
All four levers should be used in a good combination to implement strategy successfully.
It is a useful and broad framework, but a limitation is that the same control mechanisms may be part of more than one lever of control. The differences then come from the emphasis that is given in the use of the control mechanisms. Diagnostic use of MCS follows the mechanistic, repressive, traditional control approach. Interactive use of MCS takes an organic, constructive, learning-oriented control approach.
Strengths of Simons’ framework:
Focuses strongly on strategic issues and on its implications for the control systems.
Offers a broad perspective of the control system by looking at the range of control employed and how they are used by companies (that is very important because the way controls are used is key to establishing whether all four LOC are employed and to assess the balance between positive and negative controls).
Weaknesses of Simons’ framework:
No sufficient emphasis to socio-ideological controls.
Focused on top level of management.
Unlikely that it adequately explains the operation of the whole control system.
Meanings of the concepts embedded are diffuse.
Ambiguity in the definition of ‘interactive controls’, may be split into two components: interactive use of controls, and strategic validity controls.
Not susceptible to universal applicability.
The PMS framework
This paper tries to extend Otley’s framework. The PMSs framework contains 12 questions:
What is the vision and mission of the firm and how are they brought to the attention of managers and employees? What mechanisms, processes, and networks are used to spread the purposes and objectives?
What are the key factors that are believed to be central to the firm’s overall future success and how are they brought to the attention of managers and employees?
What is the organizational structure and what impact does it have on the design and use of PMSs? How does it influence and how is it influenced by the strategic management process?
What strategies and plans has the organization adopted and what are the processes and activities that it has decided will be required for it to ensure its success? How are strategies and plans adapted, generated and communicated to managers and employees?
What are the organization’s key performance measures arising from its objectives, key success factors (KSFs) and strategies and plans? How are they specified and communicated and what role do they play in performance evaluation? Are there significant omissions?
What level of performance does the organization need to achieve for each of its key performance measures, how does it go about setting appropriate performance targets for them, and how challenging are those performance target?
What processes, if any, does the organization follow for evaluating individual, group, and organizational performance are performance evaluations primarily objective, subjective or mixed and how important formal and informal information and controls in these processes?
What rewards (financial and non-financial) will be gained by achieving performance targets or other assessed aspects of performance?
What specific information flows (feedback and feed forward) systems and networks has the organization in place to support the operation of its PMSs?
What type of use is made of information and of the various control mechanisms in place? Can these uses be characterized in terms of various typologies in the literature? How do controls and their uses differ at different hierarchical levels?
How have the PMSs changed through change dynamics in the organization and its environment? Have the changes in PMSs design or use been made in a proactive or reactive manner?
How strong and coherent are the links between the components of PMSs and the ways in which they are used?
This framework facilitates the description of PMSs design and use in practice, without prior assumptions on good or bad. Two aspects of PMSs are not explicitly addressed: contextual factors and organizational culture. That is because they are viewed as contingent variables that might explain why certain patterns of control are more or less effective, rather than characteristics of the control systems that need to be incorporated into a description.
Performance management begins with purposes and objectives. The mission outlines the overriding purpose of the firm in line with their values or expectation of stakeholders. The vision sets out the desired future state, so the aspiration of the organization. So the focus of the first question is to elicit information on how organizational values and purposes are established and communicated as a means of influencing the behavior of organizational participants.
The KSFs are those activities, attributes, competencies, and capabilities that are seen as critical pre-requisites for the success of an organization in its industry at a certain point of time. It concerns the elements perceived to be important by management. They include major factors on different timescales.
The classic view of organizations says that they exist to perform activities that could be performed by markets to increase efficiency through the reduction of transaction costs. An organizational structure is then the means to establish roles and tasks. Configuration consists of the structures, processes and relationships through which the organization operates. Structures include the functional, the multidivisional, the holding company, the matrix, the transnational, the team-based, and the project based. Processes include supervision, planning, and market processes. Relationships consist of the internal relationships and external relationships.
Many control processes operate horizontally but literature concentrates on vertical control. Organizations are sometimes part of networks that impact upon their control arrangements. Controls are sometimes built into the physical structure. Structure decisions are linked to KSFs and to strategic decisions. The KSFs require assessment of the fit of structure. The relationship between strategy and structure is bidirectional. Corporate and business strategy affect structure, and organization structure affects operating strategy. The relationship also depends on whether the firm experiences an evolutionary (strategy follows structure) or revolutionary stage (structure follows strategy) of development.
Strategy is the direction the organization chooses to pursue over the long term as the means of achieving organizational objectives. Many strategy typologies have been proposed (see page 270). The focus of the fourth question is on the actions that will probably achieve outcomes (so the relationship between means and ends). It is about communicating intended strategies, not of developing emergent strategies. The nature of the strategic management process is also assessed, it can be top-down or bottom up. The differences with Otley’s question are that this question adds the generation and communication of strategies and plans and that the issue of performance measures is separated and discussed in the next question.
Key performance measures are the financial or non-financial measures used at different levels in the organization to evaluate success in achieving objectives, KSFs, strategies and plans to satisfy expectations of different stakeholders. This question related to Simons’ critical performance variables and his interactive use of control systems. Both the measures that are actually in use and the ones that are absent need to be observed. The number of key measures is also relevant since managers have a limited attention span so that the more measures, the less impacts. Articulation of measures between organizational levels is also important, as well as the explicit development of causal relationship between measures in some causal model.
Target setting and using that for evaluation has been an issue forever, but field research cannot provide a solution. Research found that target levels have effect on performance; moderately difficult goals enhance group performance. Benchmarking legitimizes targets and has been advocated a lot.
Success in areas that senior management finds important determines status and promotion. One should distinguish between performance evaluation routines and those actually operated by senior managers. Perception is often more important than the formal situation. Team- and organizational orientation works better than individual orientation. Performance evaluation can be subjective (can correct for flaws in performance measurement but may be costly in time and can be biased) or objective (no ambiguity but no adjustments; acceptable when the input-output relation is clear, the performance is controllable or when it is accepted as part of institutionalized practice. Increased attention is being paid on relative performance evaluations.
Rewards are the outcome of performance evaluations. Rewards range from recognition to pay to promotion. There are positive and negative (punishments) control items. The relationships between rewards, motivation and performance is complex. The general assumption is that rewards motivate individuals to align goals and that behaviors not rewarded are neglected but there are also other results. Bonner and Sprinkle found that the combination of target setting and monetary incentives had a bigger effect on performance than monetary incentives alone. There are some challenges to group rewards: freeriders, detachment from group, lack of equity, difficult to relate individual performance to group performance. But it has been supported by many.
Information flows, systems and networks keep the PMS together. Feedback information is information used to enable the undertaking of corrective and or adaptive action. Feed forward information is information used to enable the organization to learn from its experience to generate new ideas and to recreate strategies and plans. Single loop learning is associated with feedback information and double loop learning with feed forward information. Systems organize accounting and other control information. They are part of the information system (IS) and information technology (IT) infrastructure. PMSs often revolve around budgeting systems but increasingly, broader PMSs are used. Issues related to the characteristics of PMSs are for instance the information scope, timeliness, aggregation and integration. Also the level of detail, relevance, selectivity, and orientation are important. Networks are another layer in the IT/IS infrastructure. Information networks go beyond formal mechanisms, because informal networks of individuals can also play a key role in spreading information.
The use of control information can be more significant than the formal design of the control system. But the concept of ‘use’ is unclear. Simons made a good attempt by distinguishing the four LOC and the concept of interactive use, but even the latter can be split up in five elements. Simons’ concepts of diagnostic and interactive use have commonalities with other concepts. Feedback flows are essential to diagnostic use because they enable single loop learning. Feed forward information with its double loop function can provide a check for strategic validity. Strategic dissonance (difference between strategic intent and strategic action) can become strategic key to development. Diagnostic and interactive use of control systems are key components of organizational learning processes, but the use of strategic validity controls identifies the failure of intended strategies and the rise of emergent strategies.
As environments and organizations change, PMSs need to change too. The design and the way the information is used should change. The causes and consequences of change in the PMSs are important. The extent to which strategy has changed is an issue of interest for understanding the functioning of the PMSs.
A PMS is greater than the sum of its parts so there needs to be alignment and coordination between the different components. A key issue is the extent to which key performance measures link back to strategies and how strategies link back to key success factors and the organizational objectives. It is not assumed that an PMS will be coherent. Some parts are coherent but others have tensions and conflict, that is only natural.
A normative position is not taken in the framework. Also, no consistency is expected between the practices adopted from one part of an organization to another. Rational and unplanned differences will occur. It is a useful research tool to enable practices to be documented and correlated with other variables, such as in traditional contingency studies.
The PMSs framework represents a considerably improved tool to Otley’s for describing aspects of PMSs design and use. Empirical evidence from case studies is needed. The tool allows the speedy and comprehensive description of many aspects of PMSs design and use. Full use of the framework requires the questions to be asked at the various hierarchical levels. Collier used the framework and said it was useful in a rational-instrumental sense but has been limited to accommodating only two of Simons’ control systems (diagnostic and interactive controls). The authors reply by saying that the framework explicitly considers vision, mission, key success factors, strategies and plans, and structure. They are part of or influence belief systems, boundary systems or both. Also, the strength and coherence aspect related to the idea of balance between the positive and negative controls. Some more critics and replies on those critics are mentioned on page 277 of the paper.
Chapter 7: Jansen, E. P., K. A. Merchant, et al. (2009). 'National differences in incentive compensation practices: The differing roles of financial performance measurement in the United States and the Netherlands
This paper focuses on incentive compensation practices of firm in the automobile retailing industry. The authors wanted to find out the extent to which these practices and their effects were similar across countries. Theory is ambiguous about whether international practices have a situational best fit or global best practices. It is found that Dutch firms are much less likely to provide their managers with incentive compensation in any form. If Dutch firms do offer incentive compensation, the pays are smaller and the bonuses are less likely to be based on profit measures of performance. But where Dutch firms use incentive compensation, their performance/functions are more complex. Also, the effects of incentive compensation on net profit and pay satisfaction are negative in Dutch firms.
Something that works in one country will not necessarily work in another. Incentive compensation systems are very important because they provide the primary means by which organizations elicit and reinforce desired behaviors. The unit of analysis in this research are the systems of incentives. The automobile retailing industry can be used to generate clean and powerful tests for cross-national effects. The one-industry setting is also good for the research design. Almost all firms are privately owned and operate very local, with local owners. So they are less subject to the possible homogenizing effects of operating multinationally. The design of the study is descriptive and the following research questions were posed:
Are the US and Dutch dealerships’ incentive compensation practices largely the same?
If not, how and why do they differ?
Do incentives have the same effects in the two countries?
For the results see the first paragraph.
In the US firms, almost everyone seemed to believe strongly in the power of incentives to influence employees’ behaviors. In the Dutch firms, people believed that non-monetary incentives were more effective than incentive compensation in motivating employees.
Studies have shown that nearly all US firms rely on incentive compensation. Also, two major categories of differences exist between the US and the Netherlands: cultural and institutional differences. These differences may result in differences in incentive practices. Cultural factors include sets of norms, values and beliefs that are programmed into the minds, and therefore, are deeply engrained in individuals in a particular group. Institutional factors provide external forces that either coerce or motivate certain types of behaviors.
Three cultural aspects are involved in the study. First, the beliefs about the purpose of corporations. US managers have a shareholder perspective and Dutch managers a stakeholder perspective. So the Americans are focused on productivity and financial performance indicators and employees are viewed as resources. Incentive contracts help align the employees’ and the organizations’ interests. The Dutch see corporations as having a relatively complex objective function. Customers, employees, suppliers and trade unions are important stakeholders and each can be involved in decision making. Also, they have labor unions and employer unions who negotiate collective labor agreements. Payment of bonuses is allowed but has been traditionally small. The second cultural aspect is masculinity (Hofstede). US scores are significantly higher.
Masculinity concerns preferences for competitiveness, achievement, and material success rather than relationships and the quality of life. People high on this dimension prefer basing rewards on performance and those low in masculinity prefer allocations based on need. In the US, empowerment is often used to improve the quality of employees’ work life. Dutch firms make group work more rewarding by allowing groups to function as self0-contained social units and by supporting cooperation among group members. Third, long-term versus short-term orientation. US scores are significantly lower. High scores reflect the presence of values oriented toward the future, such as saving, respect for tradition, perseverance, and fulfilling social obligations. Dutch are prudent, economical, and never reckless. People high in long-term orientation have a preference for more stable fixed income rather than bonuses.
Three institutional factors are included. First, formalization of terms of employment. In US firms, there is a relatively high emphasis on formalization. They tend to sue contracts to regulate relatively unambiguously the relationships between employees and managers. Appraisals are based on measurable aspects of performance to get visible fairness. Also, US managers can punish or reward employees within the boundaries of a fair contract. In the Netherlands, the focus is on continuous communication and then consensus. The high ‘right to manage’ in the US may be due to:
A private enterprise culture
A low level of state involvement
Relatively high management antagonism toward trade unions
Second, tax rates. The Dutch are more prepared to pay for an expensive social security system and to accept a high tax burden to fund it. So tax rates in income are much higher in the Netherlands. So monetary expensive are more expensive to use in the Netherlands, so less use of them is expected. Third, experience with incentive systems. Most Dutch managers have less experience with incentive compensation than US managers. But there are signs of change.
There is some evidence that suggests an international convergence of incentive practices. There are some management universalities that are invariant of national differences. Standard economic theory (which assumes classical forms of economic behavior of agents) would predict similar incentive practices across countries. This research provides a powerful test of cross-national differences because the industry setting and the economics of the firms studied are nearly identical.
A survey was used. Payments consisted of the following elements:
Spiffs (miscellaneous rewards)
Small percentage of Dutch firms provide their managers with performance dependent bonuses
Sizes of the Dutch formula bonuses are much smaller than in the US
Discretionary bonuses and spiffs are given less often to Dutch managers and when given, they are smaller
Merit raises are more common in the Dutch firms than are bonuses
If formula bonuses are used, US firms are significantly more likely to base those bonus awards on profit measures (particularly net profit)
Dutch firms are more likely to set a threshold and a cap (consistent with preference for greater compensation equality or ‘leveling’
The following control variables were included:
General manger span of control
General manager experience
General manager delegation of decision rights
Dealership environmental uncertainty
Dealership customer service orientation
Dealership differentiation strategy
Bonus recipient experience
There are statistically significant differences between the firms in the two countries for all of these variables. It was found that multicollinearity is not a threat to the interpretation of the multivariate results. The effects of the control variables are generally as expected:
Use of incentives is greater where dealerships are larger (1)
Use of incentives is greater where managers have a larger span of control (2)
Use of incentives is greater where dealerships face greater competition (5)
Use of incentives is greater where dealerships are pursuing a differentiation strategy (8)
Although incentives are much less prevalent in the Dutch sample, when incentives are used they appear to be chosen in ways that theory predicts. Where incentives have become general practice, the effects of these contextual variables become muted. Additional control variables included are matched entity size, sales growth, and changes in employment. There are no significant profit effects of incentives for the US sample. A significant negative effect is indicated in the Dutch sample. The use of incentives enhance pay satisfaction in the US and weaken pay satisfaction in the Dutch firms. The total effect of incentives on pay satisfaction is significantly negative in the Netherlands, for each of the incentive elements except for spiffs. So there is some aversion towards incentive pay in the Netherlands. Reverse causality may be a threat, so it might be that dealerships with performance or employee productivity problems are more likely to use incentives to try and remedy their performance problems.
Field research follow-up
Interviews were done because the differences were so extreme. One Dutch firm was chosen because it was typically Dutch and one Dutch firm because it was different in that it used incentive compensation. Also one US firm was interviewed. The US manager said that motivating employees was the key to success and that was achieved through the incentive compensation packages offered. But his system was not perfect because not all the right behaviors were motivated. For instance, the salespeople were not consistently effective in following up with potential customers. They would also manipulate some of the measures on which rewards were based.
The CEO of the ‘typical Dutch’ firm had some interest in incentive payments based on group performance but not on individual performance because he wanted to motivate the team as a whole. Individual bonuses stimulate competition amongst colleagues which is bad. Mutual trust and cooperation should be stimulated. Performance is measured and feedback provided in formal and informal performance reviews. Implicit incentives are provided (promotion possibilities, training opportunities).
The Dutch ‘outlier’ introduced bonus plans for dealership general and department managers. The bonus payments are small relative to those paid in US firms. Bonuses were introduced to make managers conscious that something had changed (the CEO succeeded his father). The managers were given more decision-making authority but also had a new responsibility to achieve a certain performance.
There was widespread disbelief throughout the organization that monetary incentives provided much motivation. That lack of enthusiasm might cause the firm from stop offering the incentives.
So the field study shows that the US managers believed strongly in the power of incentives to influence behavior. Incentive contracts are also used by them to attract and retain good people. Dutch managers focus on non-monetary motivators such as recognition (feminine). Independence are given to employees to a certain degree by the Dutch managers also take care of their employees and emphasize cooperation and equality. Some Dutch managers argue that money only has a short-0term effect and that most people prefer the security of a fixed income.
Discussion and conclusion
See first paragraph for results. One direction is that better measures of or other controls for the many other possibly relevant independent variables are needed. Limitations of the study are that there was a three-year time lag between the collection of the US data and collection of the Dutch data; the study focused only on monetary incentives; findings and extensions should be studied in other industry settings to see if they can be generalized.
Chapter 8: Jordan,S and Messner,M , Enabling control and the problem of incomplete performance indicators
This paper studies operational managers’ attitudes towards incompleteness of performance indicators. It turns out that managers perceive performance indicators as enabling if they facilitate their actions without unduly constraining them. That is even the case if the performance indicators are incomplete, as long as managers can flexibly use them as means rather than ends. But once top management considers the indicators as more important, the flexible use becomes more difficult and incompleteness becomes a concern. The authors explain the way top management can engage in sensegiving through which they tighten the control (see the indicators as more important) and how the managers then see the indicators as coercive rather than enabling.
Performance indicators have been discussed extensively in literature, but now how managers look at them; whether they care about their qualities of performance indicators and performance measurement systems. It seems that the qualities are not of primary concern of the managers. They only explain a part of the complexity of organizational life and therefore managers do not blindly rely on such measures, but they complement and contextualize it. So the measures become ‘subject to moderation’. So even broad or incomplete financial indicators can result in concrete operational action because they will be completed with other measures and information.
Performance indicators are there to facilitate managerial action and to control. Too much flexibility would make them inappropriate for control. There is a tension between the two roles of performance indicators. Flexibility in dealing with them makes managers feel better enabled to do their work but a more focused attention on them by top management makes them feel coerced into a control system. The research question is:
How may operational managers’ attitudes towards performance indicators change over time and in response to a change in top management control?
A longitudinal case study of a manufacturing firm is used. Data was gathered through direct observations, interviews and study of documents. The framework of enabling and coercive control is used to explain changes in managers’ attitudes, combined with a process view on the interaction between top management and operational managers.
The results show that incompleteness does not really matter if the managers can use the performance indicators in a flexible way. Such flexibility is promoted when a new performance measurement system builds upon existing issues and concerns, so that continuity is allowed. But when top management signals (through their sense giving) an increased importance of performance indicators, flexibility becomes more difficult to sustain. Evaluation pressure is an outcome of such sense giving activities. The results contribute to the literature on enabling control and incompleteness of performance indicators.
Enabling and coercive control
Adler and Borys developed the framework of enabling and coercive control to understand reactions to control systems. Formalization will be perceived positively if managers feel like the system better enables them to master their tasks. However, if they feel that top managers want to coerce them , formalization will be perceived negatively. What the perception is depends on the design of the control system and its implementation. Four design characteristics of enabling control systems:
Formal system can be repaired in case of problems (permission to modify)
Internal transparency so that managers understand the logic of the system
Global transparency so that managers understand the up- and downstream implications of their work
Allows for flexibility in how they are used (adjustable guidelines)
These must be seen as the outcome of on-going interaction between top management and operational managers. So control is a dynamic process. How top management uses control systems to influence subordinates’ actions is part of sense giving.
Incompleteness and enabling control
Performance indicators are often incomplete. Some suggest that management involvement in the design and development process of a control system helps to improve and accept the system. Two of the design features just mentioned also help to reduce incompleteness:
Flexibility because it allows managers to use other measures than accounting information
Repair because it allows managers to adjust the system to improve it
The other two elements of transparency can be seen as conditions to recognize incompleteness. As said, there may be friction between top management wanting to use the performance indicators to control and operational managers wanting to enable them better.
The development process of such control systems should also be enabling, so user involvement and fit into the organization are key. Most studies ignore the decision-facilitating role of accounting, they only focus on the control role.
Research setting and design
Qualitative field study in a single organization. The company, ‘LeanOrg’, had a new COO and introduced four new key performance indicators:
Six Sigma quality
The implementation was delegated to project groups.
Implementing Lean Six Sigma with the help of indicators
The Lean Six Sigma strategy included continuous improvement, one-piece flow, customer orientation and employee empowerment. Productivity, efficiency and quality were key and had been before this strategy came. So there was some continuity with the past. Project groups were formed to implement the four new performance indicators, they regularly had meetings.
Reliance on the indicators became visible:
They shaped the agenda
They created new responsibilities
They motivated new actions and activities to turn the firm into a Lean factory
In the past, there were also indicators but these were operational and driven by middle managers in a bottom-up way. Now there was a higher workload and a top-down approach. This made the managers more sensitive to the indicators and to how they would implement the lean strategy. During meetings, managers assessed the indicators as incomplete representations of performance. For instance, they thought that the indicator ‘productivity’ did not represent what shareholders wanted, whereas the COO thought it did.
Incompleteness can be narrow, usually solved by adjusting definitions, or broader, solved by adding indicators or reducing the relative attention paid to the indicator. If incompleteness was found to be the case, the managers did sometimes not regard that as a problem. To say something is a ‘problem’ depends on its importance regarding its influence on how people act and experience the world, now and in the future. Problems of incompleteness in literature evolve around dissatisfaction, job-related tension and dysfunctional behavior.
A pragmatic view on incompleteness
The view that incomplete performance indicators are not seen as problematic is a pragmatic view. Instead of improving the representational qualities, the managers analyzed the drivers behind them and initiated improvement activities. So doing something was more important than measuring it. Indicators were used as means, as points of orientation. Why did this attitude exist? First there is the notion that performance indicators can never be complete. It also had to do with the existing managerial culture. Two main conditions allowed for the pragmatic attitude.
Building upon the past
The topics and relating activities concerned with the Lean Six Sigma strategy were not new to the firm, although there were many new initiatives. Some existing concerns were just given more attention and legitimacy. Arguments other than those deduced from concern for the indicator were given space. Managers often think in terms of finality rather than cause-effect. So they are satisfied with believing in a causal relationship rather than being able to prove or measure it. The authors even say that the unclarity of the cause-effect relationship is not seen as problematic because the end is not seen as the only or ultimate end. So the flexibility in handling the indicators allows a pragmatic attitude.
“Visions”, not targets
Top management adopted a sense giving approach in communicating the four targets and their indicators. Challenging targets were set, often seen as unrealistic. The COO first communicated talking about targets. Top and middle managers soon replaced that by the word visions, seeing it as a general direction. Top management was partly responsible because they did not set deadlines. That the targets were unrealistic was not seen as a problem because it was so obvious that they could not be attained.
Heightened concern for incompleteness
Over time, the incompleteness of the indicators became to be seen as problematic, two conditions caused that: the use of indicators as evaluation devices and identification for possibilities of improving the indicators.
Example: quality is an inadequate measure of product development performance because it conflicts with innovation. Even though bonuses were not really affected by whether targets were achieved or not, employees were concerned because they just wanted to do a good job. The objectives were first seen as visions but then replaced by shorter term goals so that they became more important. The sense giving of top management changed because they intensified their dialogue about the indicators by specifying targets and deadlines. Also they compared the plant to other plants in the division.
Cause and effect
In the beginning there was a pragmatic attitude and the targets were seen as visions. It was easy to make some improvements. Later on, it became more difficult to make improvements as the easy thins were already done. Careful selection was needed to ensure that the changes had effect on the performance indicators. Internal transparency about the definition of the indicators became therefore more important and managers were more concerned about their completeness. The ‘doing’ was still more important but once things were done, people became concerned about changes to the measurement system. One can argue that it has to do with evaluation concerns too, but as long as improvement activities are fundamental and can be expected to impact the indicator anyway, improvement of the indicator is not necessary, even if evaluation concerns are high. Sometimes the indicators were found to be inappropriate because actions that were deemed appropriate did not influence the indicator (enough).
Perceived incompleteness is not always regarded as a problem, especially at the beginning of the implementation. A pragmatic attitude towards the performance indicators existed. The reason for this could be that the indicators were seen as an enabling form of control. They were more seen as means than ends. Two conditions allowed for this flexibility; the way in which top management communicated the role and relevance of the indicators (visions, no pressure) and the way in which the performance measurement system built upon existing practices and concerns.
Contribution to previous research: incomplete calculations are not always problematic. It does not depend on the representative quality but on how indicators are related to the world of action. It also adds to the literature on enabling and coercive forms of control.
Evaluation pressure, transparency and incompleteness
An increased control focus occurred due to several mechanisms. Top management discussed more intensively about the performance of the indicators. Also, the indicators took a more central relevance when more realistic targets were set. There also was a reduction in time horizon attached to the targets. Lastly, more frequent comparisons were made. So taken together, there was a higher evaluation pressure. Cause-effect thinking increased when performance on the indicators were already high combined with high pressures. The high transparency helped managers to see issues concerning the choice and definition of the indicators. Internal transparency allowed the managers to recognize narrow forms of incompleteness (definition and measurement) and global transparency was created by linking the indicators to the overall strategy and objectives. So transparency can help managers accept the control system by making them understand it, but it also allows managers to see incompleteness.
Once managers see incompleteness as a problem, possibilities for repairing help them to see the system as still enabling, but could not completely remove the concerns for incompleteness. That is because repair can only solve narrow forms of incompleteness.
This paper examines how control mechanisms and trust are used to achieve control in an outsourcing relationship. A case study of an electricity company that outsourced its IT was used and it was found that they used a trust based pattern of control.
Outsourcing is a form of strategic alliance. There is a high failure rate, partly due to high risk. Management control systems (MCS) and development of trust may be used to decrease risk and failure. Outsourcing is the “the contracting of any service or activity to a third party”. There are many frameworks for studying the design of MCS but only in one organization. Van der Meer-Kooistra and Vosselman developed a model of management control based on transaction cost economics (TCE) but integrated with the role of trust.
It has been argued that the complexity of partnerships precludes specification of detailed contracts. Also, they need flexibility and adaptation so that there is less formal control. But sometimes a formal mechanisms may enable greater control and transparency.
Within firm control systems
Within formal controls we can distinguish between outcome control and behavior controls. Clan or social controls are always present to a certain extent. Ouchi discussed the relation between the mode of control and the information characteristics of the task (degree of output measurability and task programmability). Outcome controls are suitable if output measurability is high and task programmability low. If both are low, social controls emerge.
Trust and control systems
The role of trust in governance relationships has been researched. Processes of learning and adaptation may result in trust. Trust makes the relationship more durable and reduces opportunistic behavior. Trust is having confidence that one’s expectations will be realized. Trust is especially important inter-firm because it is only important when there is risk. Three types of trust:
Contractual trust: assumption that other party will honor the agreement (the higher, the less the need to gather information to reduce opportunistic behavior)
Competence trust: perceptions of ability and expertise
Goodwill trust: perceptions of a partner’s intentions to perform in accordance with those agreements (integrity, dependability, responsibility)
TCE and control
TCE is based on the notion that firms choose efficient organizational forms of governance structures based on transactional issues (firm-specific investments and internal and external uncertainty). Governance structures can be:
Markets (free competition to ensure control)
Hybrids (including strategic alliances) (authority to ensure control)
Hierarchies (long-term contracts to ensure control)
The mode of governance is determined by:
Frequency of the transaction
Asset specificity (degree to which asset can be redeployed to alternative use without sacrifice of productive value; high level creates dependency):
Physical assets specificity
Human assets specificity
Brand name capital
TDE models do not adequately consider the social context. The latter influences the design and the relationship and the behavior of the parties involved. Trust is an important way of reducing opportunism.
TCE-based models of management control
Speklé developed a TCE theory of control to explain nice archetypes of control. Three dimensions are included:
Ex post information impactedness (output measurability)
Two control archetypes are specified for outsourcing relationships:
Hybrid arms-length control
Hybrid exploratory control
Van der Meer-Kooistra and Vosselman developed a model of control in inter-firm relationships that integrates TCE concepts and trust. Three management control patterns are identified:
Market based pattern (similar to archetype of market control, but Speklé does not see it as suitable for outsourcing relationships)
Bureaucracy based pattern (similar to hybrid arms-length control)
Trust based pattern (similarities to hybrid exploratory control)
The market based pattern fits transactions characterized by high task programmability, high measurability of output, low asset specificity and high task repetition. There are many suppliers and prices are linked to quality of output. Detailed contracts or other forms of explicit control are not required. The uncertainty of the transaction environment is low. Trust is not relevant in achieving control.
The bureaucracy based pattern fits transactions characterized by high task programmability, high measurability of output, moderate asset specificity and low to mediate task repetition. The uncertainty of the transaction environment is relatively low and the future is pretty predictable. Controls are prescriptive and include detailed rules of behavior and rigid performance targets. Detailed contracts monitor performance. To ensure compliance, hostage arrangements may be made. Arbitration is used to resolve disputes and to counter opportunistic behavior. Both behavior and outcome controls are used. Trust plays a limited role but is important in the early stages. Competence trust and contractual trust determine selection of the outsourcer and the precedence with the contract.
The trust based pattern fits transactions characterized by low task programmability, low measurability of output, high asset specificity and mediate task repetition. The environment is very uncertain and risky. Trust becomes the dominant mechanism for control. The selection of the outsources is based on perceptions of competence trust, contractual trust and goodwill trust. Contracts are merely broad frameworks that develop over time. The institutional environment can help develop competence trust and contractual trust. Information asymmetry can be overcome through goodwill trust. Regular personal contact and commitment are used to develop goodwill trust. Behavior controls are not suitable and outcome controls and social controls develop over time. Trust is very important in achieving control.
Hybrid exploratory control is similar to trust based control except that asset specificity is moderate. Speklé and TCE in general say that in outsourcing relations, high asset specificity cannot be tolerated because it increases the potential for opportunistic behavior. But firms still do that! The trust based patterns shows that this may be because the development of goodwill trust and contractual trust mitigates opportunistic behavior. Hybrid exploratory control says that exclusive contracts with suppliers are not acceptable because they increase asset specificity, dependence and risk. But goodwill trust and contractual trust will counter opportunistic behavior.
The Australian company Central Energy was chosen because it outsources its IT and telecommunications functions and has well-established outsourcing contracts. Interviews were done and documents reviewed. Cost effectiveness and strategic issues were the main broad reasons for outsourcing. The specific reasons were:
Need to improve cost management of IT&T
Solving the problem of merging two very different IT&T cultures
Need to bring more discipline and control to IT&T
Greater access to technological knowledge and expertise
Criteria used in outsourcing decision
There were bids from two large global companies and one in-house bid. The criteria:
Clear cost reduction paths
Costs of providing the service
Plans of managing the contract
Expertise, skills base and competence
Staff transition issues
Nature of billing arrangements
Approach to strategic planning of IT
Global was chosen and the existing IT&T staff was given three options (leave, do the same at Global or do something different at Central).
Managing the outsourcing relationship
The first year was difficult and after 18 months the positives clearly emerged. Some key issues to manage were:
Agreeing on the baseline level of service included in the contract (interpretation issues)
Verifying the costs of the baseline services
Managing differences between the cultures
Managing false expectations of staff
Achieving cost reduction
Implementation of the risk-reward system
Improvements emerged slowly due to:
The most valuable staff had left the company
Different objectives (good return versus managing costs)
Managers were charged the full costs (previously direct access and fewer charges)
First there was one contact points, later more who were better able to respond to different needs of the four businesses of Central. Open communication and timely addressing issues became key. Trust was developed over time. The charge for IT applications was a source of mistrust since some managers felt they were ripped off. But the fact was that they were just charged directly for that for the first time. In the initial contract, there were no performance measures. 18 months later a risk-reward scheme was devised to monitor Global. The performance measures included cost, time, and quality.
Benefits of outsourcing
After the first 18 months the following benefits could be identified:
Increased accountability and cost consciousness
Access to Global’s expertise and enhanced services (creative solutions, international network)
Greater discipline in IT planning
Analysis and discussion
First, the characteristics of the transaction and the transaction environment and the parties are considered. Then, it is assessed whether the type of control in place and the role of trust in achieving control matches one of the three patterns.
The inter-firm control model
Task programmability not high
Output measurability first low, increased over time
Asset specificity high:
High human asset specificity
High site specificity
Transaction frequency: mix of one-off projects and repetitive transactions leads to uncertainty and thus to low output measurability and low task programmability
So first there was a more trust based pattern but when task programmability and output measurability became higher there was a move towards the bureaucratic pattern.
Therefore there was a reliance on social embeddedness, through the development of goodwill trust and competence trust as social controls. So the characteristics of the environment match with the trust based pattern.
Central wanted a good reputation, a high level of competence and extensive experience as an outsourcer. There was a high level of risk and uncertainty so they were not sure of getting such characteristics. Risk sharing and low information asymmetry were not there initially, but developed later. So the latter matches the trust based pattern.
Trust based patterns of control have social and outcome controls that develop over time. Reliance on formal outcomes is first low, but increases.
Lack of precision of provisions in the contracts
Performance measures and targets developed later
Complexity and uncertainty makes it impossible to specify all details
Conflict and negotiations with global may have helped to develop shared understandings and objectives
Developed through discussions and meetings
Participatory, interactive process to share information:
Regular formal meetings
First the IT&T outsourcing manager handled the contact, later more managers
Formal communication protocols
Building high levels of trust
Trust is essential in the trust based pattern but less important in the bureaucratic pattern. Contractual trust and competence trust are necessary in both. Competence trust will develop further in a trust based pattern. Goodwill trust is difficult but essential in the trust based pattern.
Proactive information collection: strong reputation, competence trust was not considered a concern with Global
Goodwill trust: first low, because it started with an incomplete contract and conflicting priorities, different cultures. Due to the high level of dependence and the high risk, issues were resolved and goodwill trust was developed through:
Establishing mutual interests (meetings, development performance indicators; the latter as a forum of interaction, an efficient way to communicate expectations, and sharing in rewards through the risk-reward system)
Building individual and team level trust
Joint dispute in resolution (to develop a strong understanding of perspectives and approaches)
Contributions to literature:
Draws on the three patterns of control of van der Meer-Kooistra en Vosselman to add to the knowledge of control in new organizational forms
Shows how control pattern may change over time (due to three contingent factors: characteristics of environment, parties, and transaction)
Development of trust may be compatible with tighter accounting controls and contracts
Risk was not specifically investigated
Interdependencies between development of control in the outsourcing relation and the internal system could have been investigated in more depth
A model derived from TCE was used so one could say that it is too fixed on transactions rather than relational aspects. But the model goes further than TCE since it incorporates contingent factors
Evidence is based on a single case study
The study focused on the IT&T function
Chapter 10: Major, M., Hopper, T. Managers divided: Implementing ABC in a Portuguese telecommunications company
A case study found that different employees responded differently to the implementation of ABC:
Production managers: skeptical about accuracy and usefulness, tolerated resistance workers
Workers: resistance by inputting inaccurate data late
Production personnel: difficulty understanding, fear work intensification and redundancy
Commercial and senior managers: satisfied; more accuracy, meets requirements, eases consolidation of accounts. They use it for decision making.
Proponents of ABC say that it gives more accurate product costs and helps managers understand cost causation. But definitions and precise use are controversial. ABC may also be a source of conflict.
Cooper and Kaplan laid the basis for ABC. ABC allocates resources to activities, and then activities to cost objects, through imputed causal relations based on (non)volume drivers. This does not ensure correctness. Managers often deny ABC due to it being costly and disruptive. Selection and definition can be very difficult. Two approaches are used to study the problems arising during ABC implementation. The first focuses on implementation success and the second on which factors have influence during various implementation stages. Issues of power, resistance, and conflict are often rejected.
Factor and process issues are used. It is found that top management and union support, ties to rewards, and adequate resources influenced managers’ evaluations of ABC. Local management support was associated with top management support and the quality of existing systems and union support was linked to local management support. The factors that influence the perceived accuracy of ABC were stable among recent and mature adopters but factors that influence ABC usage vary at each stage.
Marconi, the company where the case study was done
Marconi was founded in 1925 and is Portuguese. It had a monopoly on the long distance telecommunications between the mainland and the Portuguese colonies until that sector was liberated fully in 2000. It expanded internationally and to other business areas. Marconi was reintegrated into Portugal Telecom (PT) in 1995 and privatized. The full competition arrangements made Marconi more market-oriented. That affected the development of its MAS, and two other things too:
Private investors and Stock Exchange listing conditions required better financial information.
The Portuguese telecom regulator made a regulatory framework that asked for a lot of information.
In 1997, Marconi started to replace its MAS system with an ABC system.
Research methods and theory
Case study on MA practices to explore how and why questions. Interviews, plant visits and documentation checks were used. The two main issues:
Why did Marconi adopt ABC? Importance of consultants, regulators, and financial markets in diffusing ABC through European telecommunications firms.
Why did only parts of Marconi use ABC? Partly due to technical and implementation issues, the rest explained by labour process theory.
Orthodox labor process theory: class relations and firms’ need to get maximum effort from workers while maximizing value results in struggles over control of work and distribution of rewards. But there is not always conflict, often employees accept the control. So to use it, conflicts and resistance associated with implementation should be studied, thereby keeping in mind the possibility of consent. The subjective beliefs of employees operating at the various levels plays a role in the control system.
The MAS of Marconi early 1990s
The main concern was the annual report, they only had some kind of system for the external world. A new MAS was implemented when the telecommunications market was liberized. Activities for each main functions were identified and resource costs were allocated to these on the basis of labor hours. But this system did not provide all the information that was needed and was replaced by ABC in 1997.
Marconi’s board and Portugal Telecom (PT) both supported this. The Board prioritized commercial managers’ cost accounting needs. Production managers and engineers did not like that because they used to have the greatest power. External consultants helped Marconi to implement ABC. There were six implementation steps:
Selection of teams
Definition of activities
Definition of conceptual model
Collection of data
Definition of ABC software
Two committees were created to facilitate the implementation. The goal was to share ownership and involve all departments. But the engineers in the production department were never committed to ABC. The consultants asked the employees how much time they spent on what activity and developed schemes. Then they identified cost objects. They divided Marconi in five business segments. PMOs (time sheets) were used to track time spent on activities. Direct costs as a percentage of total costs became a little bit higher after implementing ABC. The consultants were around a lot and emphasized good communication and involvement. But there was resistance from the beginning.
Internal resistance and operational problems
Engineers described the system as too detailed so that they spent too much time feeding it and got too less in return. The engineers did not like the commercial departments. Production engineers said ABC data was too generic, found common costs too high and said the data was unreliable and inaccurate. They continued using non-financial data and refused using the ABC. Workers also resisted, especially in the production department. They did not disclose information because they feared for their autonomy and their job. Also they thought that the activities did not describe their job correctly. This may be increased due to the lack of training and education for employees.
Perceptions of ABC in commercial and accounting departments
They were satisfied because ABC was more accurate than previous systems. They saw that it was not perfect but did not perceive that as an impediment.
Accountants had several opinions about it. They believed it was an improvement because it signaled good accounting practice. However, they were concerned about its technical accuracy, slowness, and decision relevance because of unreliable data input and high common costs. They also did not like the inter professional rivalry it caused.
ABC could become operational and was accepted by managers due to support from the top management during the implementation. Marconi had clear goals for ABC:
Satisfying external demands
Better information for pricing and investment decisions by commercial departments
Goal consensus was not complete; the production department opposed. That affected the path of the ABC system so that it was not accepted nor used in the production department. Senior production management were part of the problem as they were hostile and tolerated late and inaccurate PMOs from the production workers. The ABC in Marconi was not directly linked to rewards or performance evaluation, which was a critical factor. The consultants and the implementation team espoused ownership of the project by non-accountants. There were implementation committees and Marconi’s accountants ensured communication and involvement at all levels. However they assumed that the members of the implementation committee would persuade colleagues of the benefits of ABC. Later they sought to rectify that by pointing facilitators but that was unsuccessful because production managers filled in that role the way they wanted to. The lesson that Marconi should learn is that education and training is important to implement ABC at all levels (although that does not necessarily generate consent).
Criteria for evaluating ABC vary by function and hierarchical level:
Commercial managers: accuracy relative to previous systems, satisfying external constituents, usefulness for pricing and investment decisions.
Production personnel: absolute accuracy, secure cost reductions within operations (they evaluated ABC negatively on both but they didn’t even try to work with it)
Some argue that ABC is indeed no improvement compared with traditional costing if used for production decisions (unreliable allocations). Employees resisted ABC because of fear for their job and their autonomy. The ABC system symbolized the dominance of marketing and financial controls over productions so there was rivalry between the departments.
Management innovations like the ABC system is actually a change in philosophy of governance. In Marconi there was no program to achieve that.
If factor and process studies want to find out why ABC might fail, sociological work that identifies factors relating to conflict and resistance must be addressed. The findings of this study show how such initiatives can reinforce perceptions of injustice, division, and inconsistencies.
ABC might have failed in production due to not articulating and persuading production personnel of an idealistic vision or failures in the implementation process. But if the reasons for resistance are structural, then user involvement would not have helped.
A single case does not give enough evidence but it adds to ABC implementation literature in three areas:
Management of change processes
Technical issues concerning ABC’s relevance and accuracy (problems with joint and common costs, no valid data made available by employees)
Resistance and consent
In the conclusions the previously described findings are discussed, read these again to recapture them.
Chapter 11: Marginson, D., Ogden, S., Coping with ambiguity through the budget: the positive effects of budgetary targets on managers' budgeting behaviors
This study examines the extent to which budgets have a more positive role in individuals’ work experiences. The outcome is that budgets may be as useful as problematic.
Budgets are often viewed negatively. It is said that they stifle innovation and learning and that they create behavioral side effects that are dysfunctional to the firm. Others find more positive results. However, research seems to focus on the negative effects and there is a lack of a counterbalancing positive perspective. That is weird, because budgets may be very functional. Path-goal theory provides an example in that it suggests that where managers do not have obvious paths and clear-cut goals; they welcome accounting based controls such as budgets to structure things. Currently, we know little about how or why managers may use budgets to cope with uncertainty and ambiguity and what that means for their budgeting behavior. We also do not know what the implications of that may be for individual and organizational performance. Some empirical evidence has been found, for instance that senior managers like budgets because they provide clarity and one can oversee where he will be evaluated on. So there are two positive elements:
Goal specificity and goal clarity that the budgets provide
Performance evaluation is based on how successfully the budget has been achieved
Role ambiguity is a central element in this research. The authors examine the extent to which managers may commit to meeting budgetary targets because doing so offers a sense of clarity and security (and not because of the threat of accountability or the promise of reward).
The authors did a case study in the UK. They performed interviews with senior and line managers, had meetings with personnel from the accounting and human resource functions and used documentary evidence. Infotain, the company where they did the case study, had largely abandoned budgeting in its traditional form. Some targets were set by the accounting function but there was little formal accountability or reward attached to budgetary performance. Tolerance limits were attached to overall budget allocations. Budgets could also be renegotiated, so everything was very flexible so that managers could adapt to the rapidly changing and highly uncertain environment. There was an ethos of empowerment.
Some interviewees appeared uneasy with the high levels of ambiguity and uncertainty. Managers wanted to pursue budgets and that was at least partly to create some structure and certainty in the face of the ambiguities and uncertainties that resulted from the empowerment they got.
Role ambiguity and budgetary commitment
To test the strength of the relationship between role ambiguity and budgetary commitment some arguments are developed.
Managers are empowered to detect new ideas and to mobilize resources around these. Actually, they get a strategic forcing role and are encourage to apply individual initiative, undertake boundary-spanning activities and to work as team players. But these activities may result in the blurring of role responsibilities and may lead to poor communication. So empowerment may create or increase role ambiguity. Role ambiguity occurs when an individual is unsure about others’ expectations of himself. Or officially; “discrepancy between the amount of information a person has and the amount he requires to perform his role adequately”. Unclear to the manager are:
Duties and responsibilities
Consequences of action
People have an inherent need for role clarity.
Coping strategies are attempts to regain clear, orderly and meaningful cognitive experiences. These may for instance be the use of defense mechanisms, in this case to reduce role ambiguity. One is to identify with the budget by emphasizing the achievement of short0term budgetary targets. Several reasons are brought forward why budgets are a useful coping strategy:
Budgetary control is organizationally good citizenship; those who control their costs benefit the firm, even more because accounting information is visible.
The manager autonomously knows what needs to be done to acquire a positive identity, no communication is needed because of the mechanistic credentials of budgetary systems. So it is clear what is required (achieve the budget), who is responsible (the manager) and what must be done to meet the budget (instruct the subordinate to do so).
So managers who experience role ambiguity will probably not adapt a flexible approach because that increases perceptions of uncertainty and ambiguity. The coping strategy is committing to the budget, where the reward is a more gratifying role experience.
However, those who feel little or no role ambiguity may adopt a more flexible attitude towards the issue of budgetary control because they have little coping needs. However, they only do so to the extent that variables such as leadership style, the supervisor’s behavior towards the budget or previous work experiences do not themselves result in budgetary commitment. Therefore it cannot be suggested categorically that budget flexibility is associated with an absence of role ambiguity. The authors only argue that the ones experiencing role ambiguity are likely to commit to meeting the budget in such circumstance. So the argument that experience of role ambiguity leads to the manager committing to meeting the budget irrespective of:
His natural tendency to behave otherwise
What the superior may expect
The effects of occupational socialization
A questionnaire survey was used. There was a lack of participation in setting budgetary targets which is normally expected to make managers less committed to achieving the budgets. Tight budgetary control may be exhibited in two ways: imposed by the budgetee on himself or imposed by the budgetee on subordinates. An expectations approach is used to measure the degeree of tightness as applied to the subordinate. Further, the authors assessed experienced empowerment, role ambiguity, the superior’s budgetary expectations, occupational socialization/business unit affiliation and leadership style. The latter was measured along a high task/low consideration – low task/high consideration continuum. Managerial performance was measured with a self-rating measure (absolute and relative measures).
Significant positive relationship between perceived empowerment and role ambiguity.
Support for H2; those who experience high levels of role ambiguity are more likely to commit to meeting the budget than those whose experience of role ambiguity are minimal.
Each variable (role ambiguity, leadership style, the superior’s expectations of budgetary performance and occupational socialization) is significantly related to budgetary commitment.
Partial support for H3; several regressions were done and:
First regression: role ambiguity and business unit affiliation significant results.
Second: business unit affiliation, leadership style, hierarchical contagion and role ambiguity significant results.
Full regression: role ambiguity and occupational socialization are significantly related to levels of budgetary commitment.
Support for H4; the higher the level of commitment to the budget, the higher the level of self-reported performance.
Support for H5; performance is inversely related to role ambiguity. So a more gratifying role experience is seemingly regained through the budget.
Overall the authors say that the experience of role ambiguity may be sufficiently strong to suppress some explanatory variables, but not all. Further studies should examine that.
Limitations and conclusions
Respondents from a single organization (limits generalizability)
Certain research instruments used for the first time
Self-reporting measures used
Causal statements are made
No other potential coping mechanisms available to the manager were investigated
But the result is valuable. Budgets may have a more positive role to play in people’s work experiences, people may commit to meeting budgetary targets because budgets can offer structure and certainty in situations of high ambiguity and uncertainty. Also, the degree of commitment may be such that it overrides the potential for known explanatory variables to influence managers’ budgeting behaviors. So the role of ‘image management’ as a motivator of budgeting behaviors may need to be reconsidered, as well as the motivational impact of budgets themselves.
Expectancy theory suggests that budgetary targets need to be accepted as legitimate, as difficult but achievable for managers to commit to attaining them (whether this is for intrinsic or extrinsic reasons). Some assurance is also needed that the outcome is controllable. This research suggests that issues as accountability, achievability and extrinsic rewards are less important than previously thought. The mere visibility of budgetary performance may be sufficient to ensure commitment and formal inducements may not always be necessary.
Further research should validate the arguments presented.
Chapter 12: Norreklit H. (2000), The balance on the balanced scorecard – a critical analysis of some of its assumptions
The balanced scorecard is distinct from other strategic measurement systems in that it is more than an ad hoc collection of financial and non-financial measures. Also, it is not only a strategic measurement system but also a strategic control system which can align departmental and personal goals to overall strategy. It suggests four areas of measurement:
Internal business process
Learning and growth perspectives
The paper examines to which extent there is a cause-and-effect relationship between these and whether the balanced scorecard can link strategy to operational metrics which managers can understand and influence.
The financial measures of accounting systems have been criticized for failing to focus on the future. So the accounting figures do not emphasize the elements which will lead to good or poor financial results. One problem is that if actions are not completed, their financial consequences are not taken into account (for instance intangible assets). So companies can feel forced to pursue short-term results and be reluctant to make investments.
Gaps may exist between the strategy expressed in the activities planned and the strategy expressed in the actual activities. The aggregate financial measures of the accounting system are not sufficient to close that gap.
Many financial and non-financial tools have been developed to address these two problems. One is the balanced scorecard. It is different from the others in that it contains outcome measures and the performance drivers of outcome, linked in a causal relationship. So the performance measurement system is a feed-forward control system. Also, it is said to align departmental and personal goals to overall strategy. This paper evaluates its underlying assumptions to see whether the balanced scorecard is a valid tool.
Research questions and methodology
Porter’s concept of strategy is used for the balanced scorecard. Competitive forces are to be examined, then internal business processes should be identified to deliver value propositions to the customers in the market segments targeted. So strategy is based on the environment and not on core competencies. Four areas of measurement are included in the balanced scorecard:
Financial (how company wishes to be viewed by shareholders)
Customer (how company wishes to be viewed by customers)
Internal business process (business processes the company has to be good at to satisfy customers and shareholders)
Learning and growth perspectives (changes and improvements to be made to make the vision come true)
The difficulty lies in linking these. Kaplan and Norton emphasize that the intention is not to just randomly collect financial and non-financial data, but to purposeful collect data that makes sense. They assume a link from measures in the fourth area (learning and growth) to financial measures. So the measures for one area are the drivers for the measures in the following area. That cause-and-effect assumption is crucial because it claims that financial measures say something about past performance and non-financial drive future performance. The validity depends on the causal assumption, which this article will investigate. It is also important because the balanced scorecard drives the technique to be used to decide on actions that are drivers of future financial performance.
The second research question is whether the balanced scorecard is a valid strategic management control tool. It is claimed to be a strategic control system and not just a strategic measurement system. An analytical tool is used to make the model useful and to develop it, so provide more clarity and precision. See table 2
Analysis cause-and-effect chain
The criteria for a cause-and-effect chain are the following:
X precedes Y in time
Observing X implies observing Y next
X and Y can be observed close to each other in time and space
X and Y are logically independent events and cannot be determined empirically. Accounting methods are logical and cannot be proved or rejected empirically. The relationship between X and Y cannot be both logical and causal. The relationship among events should be known because that determines whether the effect of an action will occur or whether the effects have to be financially calculated. The balanced scorecard seems problematic, but the underlying assumptions will be discussed next.
X precedes Y in time
A time lag between cause and effect is required, but time dimension is not part of the scorecard; cause and effect is measured at the same time. One could say that the time between cause and effect is very short and thus a time dimension not relevant, but then the accounting figures would already fully show consequences of actions. Numerous factors influence the result so it is difficult to say when the financial effect of an action will occur. Measuring effects of actions is especially difficult for new activities. The coordination of operations is of statistical nature and the coordination of development of dynamic nature.
Relationship between X and Y
Kaplan and Norton are ambiguous on this. They claim causality in that a financial result occurs if a given cause exist. But they also say that actions have to be assessed on the basis of financial reasoning to make sure that they are financially profitable. The focus here will be on the last link (customer satisfaction good financial results). The relation between loyal customers and satisfied customers is logical. Another research found a link between customer satisfaction and good financial results, but it was not really clearly described. That research describes loyal customers as the ones with low costs and high prices, so that inherently concerns profitability and the relationship is logical. Assuming causality is thus misleading. The assumed relationship between customer satisfaction and good financial results can also be criticized using neo-classical economic analysis. A lower than optimal price would yield higher customer satisfaction but is not beneficial to companies. Economics see the relationship as logical too since profits are conditioned by customer satisfaction because the transaction only takes place if the value to the customer is higher than the price he pays. However, you may say that in the short term, a company may use a non-optimal price to create higher satisfaction while building image and market share in the long term. But then profitability is not a likely outcome.
Interdependence of the four areas
There is no causal relationship between the four measures; they are interdependent and circular reasoning is used:
Capital needed for investments is limited by unsatisfactory financial results but to produce financial results, investments are necessary.
Company image is part of the customer value proposition (which creates satisfaction and loyalty) but an image is generated by customer satisfaction and loyalty.
Production processes may become inefficient if sellers make customers too satisfied by promising them things so customer value proposition and production need to be balanced.
Conclusion on cause and effect
The causality assumed is not valid. Claiming that some factors are necessarily profitable is only possible if it follows logically, which is not the case. Kaplan and Norton themselves are unclear, arguing for both a logical and a causal relationship. It may be that they actually infer finality relationships. That occurs when human actions, wishes and views are related, so when: someone believes a given action to be the best means to an end; and the end and this view actually cause the action. There is not one best means and every means has several outcomes so the means and ends are not generalizable. If finality is indeed assumed then the relationships become more ambiguous and less simple, also it makes the balanced scorecard no different from other approaches. Another criticism is that it is based on empiricism; a gap between the empirical world and their theory exists.
Analysis of the balanced scorecard as a strategic control model
Whether it is usable as a strategic control model depends on whether the methods are realistic. That required relevant information as a base and a bridge between planned and existing strategic patterns of actions.
Relationship with external stakeholders and environment
Mission and vision are redefined as measures in four areas to balance the activities with the shareholders. But not all stakeholders are included (suppliers, public authorities, institutional stakeholders, business networks). Competition and technological development are also not included. During the planning stage benchmarking may occurred but that is only static and not dynamic (does not evolve during model). Next to asking what has to be done, one also has to consider what might prevent the realization of the vision. A solution may be to ask employees to continuously gather information outside the usual routines and channels to be able to uncover external shocks and opportunities. Kaplan and Norton recommend interactive control and double-loop learning. However, the control method of the balanced scorecard is mechanical and top-down. The underlying assumption is that top management’s plan is the right plan.
Relationship with management and employees
Management needs to contribute resources and the project of the balanced scorecard needs to be rooted in the organization if it is to be successfully implemented. The control and implementation procedure of the balanced scorecard does not ensure this rooting. It should be interwoven in management’s behavior but it may be very different from how management thinks so that rooting may be difficult. To root it in employees Kaplan and Norton suggest a top-down method, disregarding implementation and support problems. Not just the external commitment should be sought, also the internal commitment to ensure active rather than reactive behavior. If the focus on external commitment is too high, employees will be motivated to focus on what is measured and less on what is not measured but important too. A solution raised by Kaplan and Norton is to add diagnostic measures to the balanced scorecard but these would also be oriented towards external commitment. So a method is needed to root the balanced scorecard in the internal commitment and the language of the employees. The balanced scorecard has to ensure that strategy is communicated throughout the organization. Performance measures should reflect the strategy and be correctly interpreted by employees. The advantage of the performance measures is that they are concrete, the disadvantage is that they are reduced. But at least the message is deliberately spread. An interactive manner should be used.
Analysis of the strategic control model; conclusion
Kaplan and Norton’s control model is top-down and not easily rooted in the environment or internally. The control process should be more interactive during strategy formulation, during the building of the scorecard and during its implementation.
The balanced scorecard is a tool which systematically expands the measurement areas traditionally involved in accounting. It is a measurement and control system. It may sharpen communication due to the inclusion of non-financial measures.
Problems with key assumptions and relationships:
Not causal but rather logical relation between areas. So invalid assumptions are made which may lead to wrong performance indicators.
The balanced scorecard is not a valid strategic measurement tool because it does not ensure organizational rooting and it has problems ensuring environmental rooting. So a gap between strategy and actions may be expected.
Suggestions for development:
Financial calculus is required for financial consequences of factors such as increased customer satisfaction or quality improvement.
Activity-based costing may identify products and customer types most profitable to the firm and the costs, cost drivers and customer value which result from different policies.
Coherence (match/complement) should be established between measurements (rather than a cause-effect relationship). An action is coherent when the actions used and the means are appropriate with respect to the intended end. Coherence analysis on two levels:
Strategy formulation: overall coherence among areas
Activities: coherence between resources and performance measures in individual groups of activities.
The paper suggests strategy formulation procedure and performance measurement formulation procedure, approximately coherent. Strategic dialogue is important here because that reduces differences in perception and understanding so that goal congruence increases. Management should continuously gather information and communicate.
Coherent strategy: one where the properties of the different strategic areas are integrated and harmonized, so that the ends planned to be achieved through the working together of the properties of the different areas. Coherence control concerns the present state and the future state. Dissonance and imbalance can be seen as prerequisite for performance measurements. So then any imbalance needs to be controlled by being balanced over time. So the implementation of a coherent set of performance measurements should be based on a coherent strategy.
Analysis of strategy coherence can be quantitative, logico-quantitative or financial. Concerns raised during the strategic dialogue may also be included. Coherence evaluation can be done using several theories (Porter’s three strategic archetypes; Mintzberg, or Miles and Snow). Financial coherence can be measured through long-term profit and cash-flow budgets.
Coherence control at activity level: linking of the goals and resources of each process to the overall goals of the company. They have to be coherent from different perspectives:
Vertically, hierarchically determined (top management point of view)
Vertically, organizationally determined (point of view personnel)
Horizontally, organizationally determined (interdependent activities mutually support each other)
De Haas and Kleingeld say that an organization is composed of multiple constituencies. The vertical top-down perspective is a relation of interdependence where the performance of the agent constituency contributes to the performance of the principal constituency. The horizontal perspective is a relation of interdependence where the performance of the supplier constituency contributes to the performance of the customer constituency. The design process is initiated from the top but the various constituencies interact in a strategic dialogue. So the agents and suppliers have the opportunity to contribute to the design of performance measurements. Also, they distinguish between result-oriented and process-oriented performance indicators. The former are used for feedback control and the latter to monitor whether throughput processes are executed in a way that will contribute to the achievement of the targets for the related indicators (feed-forward control). The result indicator of the agent then is the process indicator of the principal. Observing process indicators at lower levels then makes it possible to anticipate distortion of future results. If then conditions of the input factors of the constituencies are added and the behavior of competitors, knowledge of future results can be yielded for a big part.
The conclusion is that predicting performance is a highly complex issue involving many elements. The causal relationship along the lines suggested by Kaplan and Norton is certainly not the case. Models are needed to understand how business performance is created. Evaluation will remain partly subjective because next to the past, impact of future opportunities have to be part of the performance picture and since the future is uncertain, subjective assumptions will remain to be made.
Chapter 13: Ouchi, W. G. (1979), A conceptual framework for the design of organizational control mechanisms
Problem of organization: obtaining cooperation among a collection of individuals or units with only partially congruent objectives. Rewards have to be equitable, otherwise individuals will adjust their efforts. There are three mechanisms to cope with this problem of evaluation and control:
Markets: measure and reward individual contributions
Bureaucracies: mixture of close evaluation with a socialized acceptance of common objectives
Clans: socialization process which eliminates goal incongruence
Questions about control addressed in this paper:
What are the mechanisms through which an organization can be managed so that it moves towards its objectives?
How can the design of these mechanisms be improved, and what are the limits of each basic design?
Mechanisms overlap in organizations, but we can treat them as conceptually different from each other.
Market: prices convey all of the information necessary for efficient decision-making. Prices also provide a mechanism for solving goal-incongruity: reward each employee in proportion to his contribution. But that would be in a frictionless market. Also, the employees working in such a system are subject to bureaucratic mechanisms because they are controlled in such a manner.
Bureaucracy: rules rather than prices, rules differ from prices in that they are partial bundles of information because to establish a price a comparison between alternative buyers or sellers has taken place but a rule is arbitrary and not yet compared. Rules are used because it is difficult to determine the price of a task and determine when they are completed.
Clan: if managers know that employees are pursuing the ‘right’ objectives, market and bureaucratic control is less necessary. Socialization processes that characterize groups in different organizations but with similar values are professions. Socialization processes that characterize all the citizens are a culture. In one organization it is a clan.
There are two dimensions to each mode of control:
Informational requirements necessary to operate the control type
Market: prices (rarely perfect, often completed with set of rules)
>> An information system can be explicit (build and maintained) or implicit (grows up naturally as a result of social interaction)
Social underpinnings necessary to operate the control type (set of agreements between people)
Market: norm of reciprocity (so that cheating is not possible)
Bureaucracy: norm of reciprocity, legitimate authority
Clan: norm of reciprocity, legitimate authority, shared values/beliefs
Effective people control can be achieved in two ways:
Find and select people who fit your needs exactly (results in high commitment)
Find people who do not exactly fit your needs and create a managerial system to instruct, monitor, and evaluate them.
People treatment can be summarized as follows:
If the people treatment is totally unselective (anyone is taken, no further treatment), the form of commitment will be internalization and the corresponding control type the market.
If the people treatment is highly selective and extensive screening is performed, the form of commitment will be identification and the corresponding control type the clan.
If selection is not perfect and skill or value training is performed, the form of commitment will be identification and the corresponding control type the bureaucracy.
If selection is not perfect and behavior or output is monitored, the form of commitment will be compliance and the corresponding control type the bureaucracy.
A new view on organizational rationality: loose coupling. It implies that bureaucratic forms of control are unsuitable for many contemporary organizations. Any bureaucratic or market form of control (both rational) assumes that it is possible to measure performance. There are two dimensions to that; the ability to measure outputs and the knowledge of the transformation process:
Ability high, knowledge perfect: behavior/output measurement
Ability high, knowledge imperfect: output measurement
Ability low, knowledge perfect: behavior measurement
Ability low, knowledge imperfect: ritual and ceremony, “clan” control
It is argued that few organizations possess the underlying rationality assumed in market and bureaucratic control. Also, some say that organizations do not have a single goal or objective and that subunits are thus only loosely joined. Moreover, the environment has influence too.
Organizations can be loosely or tightly coupled. Many organizations (especially those in relatively stable manufacturing industries) fit the requirements for behavior control or output control. The market or bureaucracy fits that. Organizations in the public sector, in service industries, and in fast-growing technologies may better have cultural or clan forms of control.
Control mechanisms must focus on achieving cooperation among individuals with partially divergent objectives. There are three devices to do so:
Market: evaluate each person’s contribution and permit each to pursue non-organizational goals, but at a personal loss of reward
Clan: selecting and socializing individuals such that their individual objectives substantially overlap with the organization’s objectives
Bureaucracy: little of each → evaluate performance as closely as possible, engenders feelings of commitment to the idea of legitimate authority in hierarchies
To determine which form of control is more efficient:
Clarity with which performance can be assessed
Degree of goal incongruence
→ Related because high goal incongruence can be tolerated so long as performance can be precisely evaluated; and if there is high goal congruence then high ambiguity concerning performance can be tolerated. So there must be trust or monitoring must be possible.
Development of goal congruence is undermined by growth, turnover and specialization so bureaucracies and markets are dominant.
Also, technological interdependence among organizations increase over time and that hinders clear performance assessment. That is another reason why bureaucracies and markets are dominant.
So a balance of socialization and measurement should be found, while society becomes more pluralistic and thus more goal-incongruent. A bureaucracy does not fit the increasing interdependence and ambiguity and should be avoided.
Chapter 14: Speckbacher,G., Wentges, P., The impact of family control on the use of performance measures in strategic target setting and incentive compensation: A research note
This paper researches the influence of a firm’s governance structure on its management control system (MCS). The authors find that founding family involvement in the top management team (TMT) is associated with the TMT making less use of performance measures in its strategic target setting and incentive practices, and that this impact is moderated by firm size.
Firm owners are viewed as the principle stakeholders who determine the primary goals, and various mechanisms and institutions are brought forward to make sure that the TMT acts in the best interests of the owners and implements the goals that the owners have set. Management control is concerned with the relationship between the TMT and the other managers or employees who perform the operations. It is defined as the set of mechanisms designed and implemented by the TMT to influence and control the behavior of managers and employees to better attain organizational goals.
Family ownership occurs when the founding family is a firm’s main owner. Then the motivation for monitoring managers is greater than when ownership is dispersed. Also, if some or all members of the TMT are also members of the founding family, a close transfer of owner interests into management decisions seems possible so that formal governance mechanisms are not necessary anymore. Some researchers argue that family involvement reduces agency costs. But there is a ‘dark side’ to family involvement too. They may use their controlling position to obtain personal benefits at the expense of minority shareholders. Also, the entrenchment effects of family ownership may lead to inferior performance. Lastly, when the TMT has family and non-family members, this may provoke subgroup conflicts that harm its effectiveness as a whole.
The authors argue that founding family involvement in the TMT results in a more personal and informal way of communicating strategic targets and controlling the behavior of subordinates. Also, they say that this informal and personal way of implementing strategies is only effective in small firms.
A sample of 304 small and medium-sized firms is used. The two main results:
Founding family involvement in the TMT is associated with the latter making less use of performance measures in its strategic target setting and incentive practices
This impact of family involvement is moderated by firm size, so that the impact of family involvement on the use of performance measures is weaker in larger firms. Family ownership alone does not seem to make a difference
Theory and hypotheses
Family involvement may be a strategic resource. They provide actual resources (human capital, financial capital) but more importantly, their personal relationships with managers, employees and other stakeholders (social capital) form a valuable and unique resources specific to family firms. It can be a source of competitive advantage and even superior performance. The latter may be due to the reduction of agency costs. The translation of the goals into activities is based on family relationships, shared views, and implicit and informal ways of communication. Therefore, the TMT may use informal ways of transferring that tacit knowledge of goals and strategy to lower level managers and employees too, rather than using explicit performance measures and formalized incentive contracts.
Family relationships and social networks are valuable because they are able to:
Replace expensive formalized procedures to transfer knowledge or information
Facilitate transferring tacit knowledge and non-codifiable information
So these relationships and social networks are valuable, rare, and difficult to replicate and imitate.
If the family owners are executives themselves, they may be reluctant to bear the costs of making their entrepreneurial ideas known to outsiders. They may see it as an advantage to transfer such insights directly and personally into the TMT instead of writing everything down. That way, complexity is reduced. Explicit performance measures may even be counterproductive if tacit knowledge of goals and strategies is important. Mintzberg and Waters say that in if the owner is also manager, new strategies emerge as a vision in the owners’ minds so that they can be implemented informally through the owners’ omnipresence in the company. A synthesis of planning and implementation occurs, two management functions that usually are separated in other firms.
H1. Involvement of founding family members in the TMT is associated with firms making less use of multi-perspective performance measures in strategic target setting practices.
Family firms have employment relationships which are based on goal congruence, shared views and valued, mutual trust and reciprocal altruism. These help to reduce cognitive distance, “honest disagreements”, misunderstandings and conflicts of interest and opportunistic behavior. Therefore, these values substitute for incentive contracts.
H2. Involvement of founding family members in the TMT is associated with firms making less use of incentive contracts for managers.
The effects in H1 and H2 are not expected in firms where families are owners but do not act as executives.
A BSC-type (Balanced ScoreCard) of performance measure is one which fulfills the following points:
It groups performance measures into one financial and several nonfinancial perspectives
It formulates strategic targets for both financial and nonfinancial performance measures
It links strategy targets with incentives
H3. Involvement of founding family members in the TMT is associated with firms making less use of BSC-type performance measurement systems.
Research says that the bigger the firm, the more it uses formalized and developed management control systems because of their greater complexity, which makes them need more coordination and communication.
Also, larger firms can use economies of scale and have resources for formalized and developed systems. Conflicting results are found concerning the effects of size on performance measurement systems, which may be due to neglecting a certain variable.
H4a. The impact of family involvement in the TMT on the use of multi-perspective performance measures in strategic target setting practices is weaker for larger firms.
H4b. The impact of family involvement in the TMT on the use of incentive contracts is weaker for larger firms.
H4c. The impact of family involvement in the TMT on the use of BSC-type performance management systems is weaker for larger firms.
Data and variables
A survey was used to test the hypotheses. They were completed by TMT (66%), head of accounting (14%), assistant to managing director (3%) or other (10%). The quality of the responses was good. The industries were heterogeneous. A couple of interviews was used because the questionnaire only used single-item scales (might not capture complex concepts).
Size (four size classes), age (four classes), industry, and importance of nonfinancial goals are used as control variables.
H1, H2, and H3 are strongly supported. H4a and H4c are also confirmed.
Very old firms (before 1900) make less use of multi-perspective performance measures.
Highly significant positive effect of firm size on the use of incentive contracts.
Manufacturing and service firms make less use of performance-related pay mechanisms than sales firms.
Firms that put more emphasis on nonfinancial goals use less performance-related pay mechanisms.
No significant size effect for non-family firms in all three models.
Empirical evidence is found that founding family involvement in the TMT is an important contingency factor for management control system design. Family ownership alone does not influence the use of strategic performance measurement systems, but if they act as an executive in the TMT, there is a difference in the use of formal control mechanisms for explicating and implementing business strategies. These effects are weaker for larger firms.
It has long been assumed that because professional managers are chosen for their skills and abilities and family managers just inherit their positions, family firms are managed less rationally and less professionally. Some therefore also assumed that these family firms need to have the same control instruments and techniques as professional firms to be successful. Family firms may well be non-professional in that they rely less on formal performance measurement and control systems but that this does not result in less effectiveness and inferior performance. Their reliance on informal models of management can be a strength that ensures the use of tacit knowledge and non-explicable entrepreneurial insights and visions. By using firm age as a control variable, no evidence was found that family firms relying on informal modes of control don’t survive in the long run or become convinced by the market to use a more formal mode.
The results do not show why family controlled firms make less use of formalized instruments and how they compensate for it. Another limitation is the execution of the research in a limited geographical area.
Chapter 15: Sundaramurthy, C. and M. Lewis (2003), Control and Collaboration; Paradoxes and Governance
Corporate governance is a debate between control and collaboration. The authors use agency and stewardship theories to emphasize the value of empowerment and of monitoring and to elaborate the tensions between them. These tensions are used to examine reinforcing cycles; fostering strategic persistence and organizational decline. A paradox framework is used.
Academic thinking used to focus on either or thinking, but a more paradoxical approach is now required due to environmental turbulence and ambiguity. That approach supports a simultaneous need for collaboration and control. Lewis (one of the authors) developed a framework to assess paradoxes. She examined the tensions, reinforcing cycles and management of paradoxes. Paradoxical tensions occur because we see interwoven and opposing elements. Most people use formal logic to distinguish between things but that may be counterproductive. In this article, agency and stewardship theories are used to assess similarities and differences to explore tensions. OB, strategy and governance research is used to find dynamics that inhibit change when overemphasizing one approach. Four reinforcing cycles that foster organizational decline and strategic persistence are detailed.
Agency theory concerns a control approach, it seeks to reduce self-serving behavior of managers that negatively impact owners’ wealth. Stewardship theory is a collaborative approach. Managers are empowered to enhance ties between board and management and decision making. Both approaches make assumptions about humans, see figure 1.
Agency theorists increasingly see the costs and limits of control while stewardship theorists see how contextual factors may enable or constrain a collaborative approach.
Paradoxical tensions may result in defenses that inhibit change and support counterproductive reinforcing cycles. Response may be cognitive, behavioral and organizational to protect the ego or to prevent actors or groups from confronting the limits of current understandings and practices. These cycles are called self-fulfilling prophecies, downward spirals or strategic persistence.
If the setting is highly collaborative, directors and managers want to defend their collective decision making. Groupthink results in strategic persistence during good times and organizational decline as performance worsens. If control is dominant, separation of responsibilities is clear so that directors want to defend their monitoring efforts and they prompt managers to justify their chosen strategy and its implementation. Distrusts rises so that these actions are reinforced and learning and interactions are hampered.
Reinforcing cycles of collaboration
Groupthink is a pattern of collective defenses aimed at denying or suppressing tensions. The nature of these defenses and their consequences vary and result in different performances:
High performance cycle
A focus on collaboration and past success leads to decision makers expressing confidence in their team (collective efficacy), in their strategies and in their beliefs concerning cause and effect. Directors and executors with a lot of stock ownership may have a strong identification with the firm so that comments regarding firm strategies and performance are internalized. The high collective efficacy and firm identification raise the potential for counterproductive defenses.
Collaboration and strategic persistence may be intensified due to:
Faulty attributions: causes of success no longer or wrongly identified; environmental change discounted, both due to overconfidence and egos.
Consensus seeking: minority opinions suppressed, critical feedback not sought.
Complacency and entrenchment
The authors expect that firms with high performance focusing on collaboration will experience reinforcing cycles that foster strategic persistence.
Low performance cycle
The collective efficacy declines as performance declines so that potential defenses are altered. Groupthink results in:
Faulty attributions: external attribution of decline and blaming (as opposed to internal attribution of success and discounting environmental changes in the high performance cycle). Some managers say that their strategies are inappropriately implemented rather than inappropriate themselves.
Threat rigidity: low performance leads to stress so that managers only focus on familiar knowledge to reduce communication complexity and anxiety. They may also formalize procedures and further consolidate power.
Escalating commitment to falling course of action: executives intensify their commitment to safe face and directors intensify their commitment to show their dedication to management. Also they make decisions to support a more extreme version of the existing strategy, which seems safe but is very risky. Groups are more likely to do so because supportive interaction reduce risk and uncertainty perception.
The authors expect that firms with low performance focusing on collaboration will experience reinforcing cycles that foster organizational decline.
Reinforcing cycles of control
Excessive use of rational controls signals and reinforces distrust.
High performance cycle
Distrust may lead to suppressed stewardship, polarization of board and management, and myopic behavior (focus on short term), which results in greater control and strategic persistence.
Suppressed stewardship: higher level managers are often motivated by higher order needs (achievement, growth, affiliation). Rational controls hinder these by limiting social ties and giving extrinsic rewards. So their aspirations are suppressed and the managers:
Reduce their organizational commitment and engage in self-serving behavior
Become withdrawn and resistant
‘Crowding-out effect’: monitoring and bonding mechanisms undermine motivation under two conditions:
Extrinsic rewards do not fulfill agents’ needs if they have high working morale, internalized performance standards, firm identification, and decision-making authority.
Constraining and coercive controls shift the locus of control from agent to principal so that the agent is less prone to comply.
Polarization: impedes communication between board and management so that distrust increases. In successful times, directors and executives may become over confident so that they do not seek each other’s advice. Information asymmetry may cause managers to seek credit for the success and to hoard critical information to gain more power. On the long term, polarization leads to myopic behavior that inhibits learning and risk taking. Boards dominated by outsiders are more likely to use financial controls that results in myopic and risk-avoiding behavior. Top executives are more likely to invest in long-term projects if there are more insiders on the board.
Myopic behavior: learning becomes tied to situations were extrinsic rewards apply so managers only engage in single-loop learning and frame-braking innovations and change are discouraged.
The polarization and myopic behavior are accompanied by information asymmetry and risk aversion so that distrust of executives is confirmed, so that rational controls are increased. Directors may respond to this cycle by decreasing their control, only using rational control in those areas where opportunistic behavior is most likely, and by delegating decision making.
The authors expect that firms with low performance focusing on control will experience reinforcing cycles that foster strategic persistence.
Low performance cycle
Rising tensions and defensiveness cause a deviation-amplify loop. Greater control indicates directors’ distrust. Executives’ self perceptions and performance are negatively impacted. Impression management and political turf wars may occur. These reaffirm the need for control so that managers cannot innovate and avert failure.
Further suppression managerial stewardship: self-efficacy, aspirations and beliefs decrease so that intrinsic motivation is decreased.
Impression management: to increase self-efficacy and avoid blame. Input from outside board members or institutional investors will not be sought due to the fear of being seen as incompetent. Communications may be distorted and information asymmetry increased, so that effective decision making is less possible and suspicions increase.
Splitting: managers and monitors construct we/they distinctions so that negative emotions can be projected on each other. Firm identification may be reduced and problems blamed on the board. A focus on control enforces this due to the insider-outsider distinction. A dual-title holder and close social ties may prevent that.
The costs of increasing controls rise until it is no longer possible. Rising controls and defensiveness may cause managers to invest resources in defending their past actions rather than focusing on solving the current problems.
The authors expect that firms with low performance focusing on control will experience reinforcing cycles that foster organizational decline.
Managing control and collaboration
Control helps to curb human limitations through discipline and vigilance and collaboration uses cooperation and empowerment to tap individual’s aspirations. Using and balancing them both facilitates adaptation and learning.
Self-correcting cycles: trust and conflict
Trust facilitates collaboration and complements rational controls. Too much trust encourages extreme cohesion. Conflicts stimulated critical feedback but may trigger political battles that fuel distrust and undermine social ties. High performing teams may have trust and distrust. Trust is needed for cohesiveness and comfort needed for thoughtful and open interactions. Distrust may even be functional because it encourages questioning of existing assumptions and past decisions. Conflict may address human limitations as long as it is cognitive (task-related) and not affective (emotional).
So there should be trust in actors’ capabilities so that collaborative board-management interactions exist and there should be distrust of human limitations and cognitive conflict to ensure controls and constructive debates.
Self-correcting cycles: diversity and shared understandings
Governance structures that promote diversity and shared understandings are necessary for self-correction and learning. Building diverse viewpoints and skills within the board enhances monitoring and decision making capabilities. Shared understandings among directors and executives increases mutual trust and cooperative problem solving. A dynamic mix of insiders and outsiders is proposed:
Insiders: strong commitment to and rich knowledge of the firm
Affiliated directors: expanded understandings of internal operations, access to critical resources
External nonaffiliated members: insights into market, technological and legislative changes across industries
However, diversity can be difficult for collaboration and result in fragmentation and using safe formal communication methods. Shared understandings should be sought to increase trust. Building shared understandings can be done through:
Cooperative strategic decision making
Greater informal and formal interactions (subcommittees, visits, meetings)
So promoting diversity and shared understandings is necessary to manage control and collaboration.
These are necessary when a system cannot correct itself. They can help to overcome resistance. (Blocks of) institutional investors or shareholders can exercise pressure. Such interventions may provide a reorientation in strategic direction and counter strategic persistence and escalating commitment. Escalating anxiety and defenses in the TMT can also be stopped. So such external interventions help to stop reinforcing cycles.
The authors have sought to go beyond either/or thinking by exploring paradoxes of governance, specifically the control-collaboration tension, the reinforcing cycles that may occur and how to manage them. Other theories may further explore this balance, for instance:
Contingency perspectives: internal and external control mechanisms, roles as substitutes and or complements. Or manager’s attributes, firm characteristics, firm’s life cycle, external environment may have their influence.
Resource dependency theory: may influence mix insider-outsider directors because outsiders are useful when extra valuable resources are needed and insiders when firm-specific knowledge is more valuable.
It is proposed to execute longitudinal studies for examining these findings. Multiple data sources and secondary data should be used.