The influence of isomorphism and practice variation on
the public management control system: The
case of the WMO in Dutch municipalities
Word count: 13.036
June 25th, 2018
Subject: Master Thesis
Master: Master Controlling (EBM870B20) Supervisor: Dr. E.G. van de Mortel
Co-assessor: Dr. C.P.A. Heijes Name: Robert Anno Bültena Student number: S3272982
Telephone: 06-28762653
2
ABSTRACT
This research studies the influence of isomorphic pressure and practice variation on decentralizing three new tasks to municipalities under the Social Support Act (WMO). This research uses transaction cost theory (TCT) with particular emphasis placed on the intended financial changes to efficiency and the management control system (MCS) of municipalities. While MCS are widely studied, the focus has primarily been on the private sector. Not enough effort has been put into explaining how the MCS of municipalities work in the public sector, and how the opposing pressures of isomorphism and practice variation similarly influence the MCS.
Through the use of semi-structured interviews with financial controllers and internal documents, this research finds evidence that both mimetic and normative isomorphism affect the MCS of municipalities. The isomorphic pressures demonstrate homogeneity in the WMO. Practice variations were also found. Particularly, different political preferences, steering on municipal features and social cohesion demonstrate heterogeneity in the WMO. Although both opposing forces were present, each influenced the MCS in an efficient way. This study contributes to both the public interest and institutional theory concerning the combination of isomorphism and practice variation.
Key words: WMO, public sector, decentralization, efficiency, municipalities, management control system (MCS), transaction cost theory (TCT), isomorphism, practice variation.
3
Index
1. Introduction ... 4
2. Sector description ... 6
2.1 Decentralization in the public sector ... 6
2.2 The WMO ... 6
3. Literature review ... 8
3.1 What is a management control system? ... 8
3.2 Management control systems in the public sector ... 8
3.3 Transaction cost theory (TCT) as a framework ... 9
3.4 Isomorphism ... 11
3.5 Practice variation ... 12
3.6 Towards the conceptual model ... 13
4. Methodology ... 15
4.1 Method of data collection ... 15
4.2 Interview method and development of interview questions ... 15
4.3 Selection of respondents ... 16
4.4 Data analysis... 17
5. Results ... 18
5.1 Transaction cost theory (TCT) ... 18
5.1.1 Asset specificity ... 18 5.1.2 Programmability of activities ... 19 5.1.3 Ex-post information ... 20 5.1.4 Conclusion ... 21 5.2 Isomorphism ... 24 5.2.1 Mimetic isomorphism ... 24 5.2.2 Normative Isomorphism ... 25
5.2.3 Homogeneity in the WMO due to isomorphism? ... 26
5.3 Practice variation ... 29
5.3.1 Heterogeneity in the WMO due to practice variation? ... 31
6 Discussion and conclusion ... 33
6.1 Answering the research question ... 33
6.2 Comparing of findings with academic literature ... 33
6.3 Limitations and future research ... 35
4
1.
Introduction
Research conducted by Sociaal Cultureel Planbureau (SCP, 2015) demonstrates that municipalities face difficulties in executing the new tasks under the Social Support Act (WMO). In particular, budget cuts mean that municipalities struggle to deliver improved custom-made care. This struggle arises because these tasks are decentralized to municipalities in order to create a more inclusive society for people with disabilities or chronic diseases. It has also been assumed that municipalities are better able to deliver the needed custom-made care with smaller budgets (Bekkers, 2016). As a result, municipalities have become responsible for fulfilling this task with a greater emphasis on efficiency.
The decentralization of new tasks to municipalities is the result of a change in legislation. The Algemene Wet Bijzondere Ziektekosten (AWBZ), which was the prior law that arranged care for the elderly, disabled and chronically ill patients, has been abolished. The decentralization implies that municipalities are broadening their WMO care services with three new tasks: assisted living, day activities, and short-term stays (Movisie, 2015). Municipalities have to execute these tasks more efficiently, because the budgets for the WMO care have shrunk. However, uncertainty is inherent to change. Municipalities face uncertainty due to the complexity of custom-made care and the lack of extensive knowledge about the demand in the healthcare field (VNG, 2015). Bekkers (2017) found that some municipalities had a deficit for the WMO care budget, whereas others had a surplus. The reason for regional differences is that municipalities receive significant amounts of freedom in executing their tasks, because there are few guidelines on policy and budget making. This is amplified by the fact that municipalities only have to justify their decisions to their own local council (Bekkers, 2016). In order to counter uncertainty, municipalities develop an effective management control system (MCS) that helps organizations to achieve organizational and strategic goals (Romney & Steinbart, 2006). Therefore, the MCS can help municipalities to design and implement the new tasks. However, municipalities face legitimacy pressures in executing the new tasks (DiMaggio & Powell, 1983). Legitimacy explores social norms and the impact of rules on organizational practices (Brandau et al. 2013). Organizations are forced to integrate commonly accepted norms and concepts to gain social legitimacy, which inevitably leads to isomorphism. This isomorphism can be further classified into three isomorphic mechanisms (DiMaggio & Powell (1983): coercive, mimetic, and normative. Isomorphism is important, because it contributes to a better conceptualization of how institutional forces are exerted on individual organizations and how imposed practices are adopted (Tsamenyi et al. 2006). Besides isomorphic pressures, practice variations can also be present. Practice variations are seen as deviations from norms and explain how and why individual organizations introduce variations in their systems and practices (Cruz et al. 2009). The reasons for adopting practice variations are based on different insights and priorities, which in turn can lead to different outcomes. Moreover, these practice variations may be needed to satisfy the different but complementary needs at the local level (Cruz et al. 2009).
5 Overall, both isomorphic pressures and practice variations have to be considered when designing and implementing the MCS. It has been proposed that isomorphism leads to similar practices while practice variation leads to different practices (DiMaggio & Powell, 1983). Since both of the above variables can influence organizational results, this research considers them both and studies how they influence the MCS of municipalities. Hence, the research question is as follows:
How do isomorphic pressures and practice variations influence the management control systems (MCS) of municipalities? The case of the WMO in Dutch municipalities.
To answer this question, several sub questions are drafted:
1. What are the responsibilities for WMO care? 2. What does a MCS entail for a Dutch municipality?
3. What does isomorphism look like for Dutch municipalities? 4. What does practice variation look like for Dutch municipalities?
The theoretical contribution of this study is fourfold. First, this study contributes by providing insights to the understanding of institutional pressure in the public sector. It aims to explain and understand the dynamics of institutional change and the implications for financial information systems. This study uses isomorphism to describe homogenization and follows the following quote by (DiMaggio and Powell, 1983 p.149): “The process of homogenization
is best described by the concept of isomorphism which is a constraining process that forces one unit in a population to resemble other units that face the same set of environmental conditions.” Second, by studying practice variations, this study explores differences among
local practices. By incorporating practice variation, this study demonstrates how differences occur and how this affects local policies and procedures. Third, there has not been much research to date on the influence of isomorphism and practice variations as opposing forces in the public sector. Consequently, this study provides useful guidance how both forces simultaneously affect the MCS in the public sector. Fourth, this study increases the generalizability of existing theories on isomorphism and practice variation.
The remainder of this research is organized as follows. Chapter Two describes contextual information involving Dutch municipalities and the WMO. Chapter Three contains the literature review. This chapter uses transaction cost theory (TCT) for describing the efficiency aspect of the new WMO tasks as well as theories about institutional pressures and practice variations. All this literature is used to map the conceptual model that is presented in the end of the chapter. Chapter Four describes the methodology. Chapter Five describes the results of this research. The last chapter, Chapter Six, provides the discussion, conclusion, limitations, and room for further research.
6
2.
Sector description
This section will answer the first subquestion: What are the responsibilities for WMO care?
2.1 Decentralization in the public sector
The Dutch governmental system is organized in three different layers: central, provincial, and municipal. The Netherlands can therefore be characterized as a decentralized unitary state (Speklé & Verbeeten, 2014). Decentralized parties have their own set of tasks and responsibilities but might in some instances be supervised by the central government (Schuszler & Deth, 1990). However, decentralization also results in considerable freedom in the design and use of performance measurement systems (Speklé & Verbeeten, 2014).
Decentralization, which is the transfer of authorities and responsibilities from central government to provinces and municipalities, has gained increasing attention as a means of improving public services (Kim, 2008). Decentralization has the advantage of being more efficient and resulting in the governing authorities possessing more knowledge of the demand in the field. This implies that municipalities’ knowledge of local care is better suited to the local needs and wants of citizens (Gallagher et al. 1995). Whereas central approaches lead to standard products and services, local solutions are supposed to offer greater flexibility and specialization (Putters et al. 2010). The result is that each municipality can make its own policy based on the demand of its inhabitants.
2.2 The WMO
Municipalities have been responsible for WMO tasks for years. For instance, municipalities have been responsible for addiction treatment care since 1987. Municipalities also have been responsible for the wet voorzieningen gehandicapten (WVG) since 1994, which is a law that is responsible for delivering adequate wheelchairs and other residential facilities. However, both laws were replaced and merged into the WMO in 2007. Domestic help also became covered under the WMO. The 2007 WMO arranged for citizens of the municipality to receive adequate support. With the WMO of 2015, the tasks have been broadened for municipalities with assisted living, day activities, and short term stays.
The WMO facilities are related to four domains. These are the ability to run a household, the ability to move within one’s own residence, the ability to move with an appropriate means of transport, and the ability to make contacts within society (Movisie, 2015). The rationale behind the WMO 2015 is that the self-reliance of citizens will be enhanced. This is the main goal as described by the secretary of public health, welfare, and sport:
“With the WMO 2015, municipalities are able to deliver adequate, suitable support that is in line with the personal needs of citizens. The social environment, the residency situation of the citizen, and the needed support differ from people to people, making sure that custom-made care is needed” (Kamerbrief voortgang en ambitie nieuwe WMO, volwaardig meedoen, May
7 However, since 2015, the categorization has been extended from four to five domains: domestic help, delivery of appropriate tools and services for social care, assisted living, day activities, and short-term stays (Movisie, 2015). Each of these domains will be discussed in more detail below.
Domestic support consists of care such as doing the laundry, vacuuming, and cleaning the house. This task was already covered in the WMO of 2007. Therefore, municipalities have experience in areas such as task execution, relevant needs, the quantity of citizens that ask for domestic support, and costs. Domestic support has the goal of delivering adequate support to citizens so that they can live longer in their residences.
Delivery of appropriate tools and services can involve assigning wheelchairs and stair lifts. This type of support was also part of the WMO of 2007. Therefore, the municipality also has experience with task execution, relevant needs, the quantity of citizens that ask for domestic support, and costs. The main goal of this kind of support is for citizens to be able to move around the household.
Assisted living is the first new task for municipalities since 2015. The goal of this kind of support is to increase the self-reliance of citizens. Assisted living also provides help for citizens that need structure in their lives. Examples of this kind of support include helping by doing the groceries and residential guidance.
Day activities are the second new form of support. This form of support is helpful for citizens that have psychological or physical limitations and therefore cannot fully participate in society. This kind of support increases the self-reliance of citizens, relieving the burden on family members. Day activities help people to participate in society, such as by meeting people.
Short-term stays are the third new task for municipalities since 2015. This type of care has the goal of relieving family members and care givers. Municipalities can provide help by selecting the appropriate care organization.
People have two options for receiving WMO care. The first is care in kind (zorg in natura) and the second is an individual budget (PGB). Care in kind means that the municipality arranges the WMO care for the citizen. The municipality searches for the relevant care organization and arranges the appropriate care for the citizen. Hence, municipalities have to sign contracts with care institutions. Citizens can also choose to receive the needed care via the PGB, which involves the citizens self-arranging the needed care.
The rest of this research will focus on the new tasks since 2015. The reason for focusing on the new tasks is that at the moment of writing, municipalities are struggling to design and implement the MCS. The first year, 2015, was a transition year that focused mainly on continuity of care; therefore, municipalities did not organize or implement MCS. However, since the transition year has been completed, municipalities are freer to design their own MCS. This creates an ideal research ground to investigate how municipalities develop and implement their MCS.
8
3.
Literature review
This chapter describes the literature and ends with the conceptual model. Sections 3.1 – 3.3 answers the second subquestion: What does a MCS entail for a Dutch municipality? from a theoretical point of view. Section 3.4 consists of literature on isomorphism that is relevant to subquestion three: What does isomorphism look like for Dutch municipalities? The last section 3.5 focuses on the fourth and last subquestion: What does practice variation look like
for Dutch municipalities?
3.1 What is a management control system?
There are many conceptualizations of MCS (Strauss & Zecher, 2013). For example, Anthony and Govindarajan (2007, p.17) defined a MCS as “system(s) used by management to control
the activities of an organization”, whereas Simons (1995, p.5) defined it as “the formal, information-based routines and procedures managers use to maintain or alter patterns in organizational activities”. In addition, Merchant and van der Stede (2003, p4) defined a MCS
as “all the devices or systems managers use to ensure that behaviors and decisions of their
employees are consistent with the organization’s objectives and strategies.” These three
definitions are the most commonly used in the literature (Strauss & Zecher, 2013) and are updated continuously because of changing control needs in the current environment (Nixon & Burns, 2005).
Within the literature, MCS can split into two categories: MCS as either cybernetic control systems or systems that have a broader scope (Merchant & Van der Stede, 2003). Cybernetic control systems are usually financial systems and are used by applying planning and control cycles (P&C). Cybernetic control systems have established procedures, rules, and guidelines that are linked to the strategy. Examples of cybernetic control systems are budgets, financial, and non-financial performance measures (Malmi & Brown, 2008). In contrast, MCS with a broader scope are more focused on a broader set of components within the organization, such as the cultural and administrative controls (Malmi & Brown, 2008). However, this research is particularly interested in efficiency and therefore left out the administrative and cultural controls. The rest of this research accordingly focuses on cybernetic control systems.
3.2 Management control systems in the public sector
The MCS definitions described in section 3.1 are mainly profit-based (Merchant & Van der Stede, 2003). Central to this kind of MCS is that the financial result is a key measure to determining the efficiency of the organization. However, this does not apply for non-profit organizations like municipalities. Municipalities do not have the goal to make profit; instead, they focus on maximizing the well-being of their inhabitants (VNG, 2015).
Municipalities have become more interested in profit-based insights called new public management (NPM) (Speklé & Verbeeten, 2014). New public management is used in government and public service organizations as part of an effort to make the public sector more “businesslike” (Hood, 1991). However, NPM is not a new phenomenon, since these approaches were already being developed during the 1980s. The reason for introducing NPM was to improve efficiency by using private sector management tools. Consequently, the public
9 sector becomes more result-oriented, which is new because the focus of municipalities was previously on internal procedures (Putters, 2010).
Despite much attention on NPM, there is still limited insight on the usage of MCS within the public sector (Speklé & Verbeeten, 2014). Although NPM can be valuable for public sector organizations, the “best fit” control for the public sector organization is dependent on several factors. For instance, Merchant and Van der Stede (2003) found that the type of activities influences the “best fit”, and Sandino (2007) found that the choice of the MCS is also dependent on the strategy and structure of the organization.
3.3 Transaction cost theory (TCT) as a framework
The introduction of the three new WMO tasks intends both policy and financial changes. The Dutch central government assumed that municipalities would have more insight on the personal situation of their inhabitants and would therefore be able execute this task with smaller budgets. Emphasis on efficiency is in line with transaction cost theory (TCT) (Williamson, 2007). TCT states that the price for a good or service is higher than the actual market price. This is because the price for a good or service is not only based on the goods or services themselves, but also on the transaction cost that is inherent to the good or service. While Williamson’s research (2007) is mainly based on profit organizations, he also states that TCT is suitable for non-profit organizations.
Another author that incorporated TCT in his research was Ouchi (1979). Ouchi (1979) used TCT to look at different organizational structures and determine how to manage people in an organization. He distinguished three control mechanisms: market, clan, and bureaucratic controls. He divided these control mechanisms into “knowledge of the transformation process” and “ability to measure outputs.” However, it is not clear if the usage of one or more control mechanisms is efficient for a specific organization. Moreover, the extent to which this model fits the public sector is questionable, since it is a model that focuses primarily on the private sector.
Speklé (2001, 2004) is also an author that uses TCT, but in a more specific sense. In this framework, the nature of the organizational activities is defined along three dimensions: the asset specificity, the extent of programmability, and the intensity of ex post information. These attributes differ in their problem-solving abilities. Moreover, the framework explains the alignment of a control archetype by focusing on efficiency. He developed five archetypes in a comprehensive way that makes it clear which mechanism should be used in different circumstances. The archetypes will be discussed below.
Arm’s length control is the first archetype that has low levels of uncertainty and specificity. Within this type of control, rules and procedures are included, but the market determines how it will be executed. Therefore, this type can be considered as “market control,” using the term mentioned by Ouchi (1979). In addition, Speklé (2001) found that this type of control has a steering mechanism that is linked to performance management. This is possible due to a high programmability of activities and a low specificity of resources.
10 The second form of control, machine control, has a high level of specificity and a low level of uncertainty. Machine control refers to pre-established rules and procedures that are mainly based on standards and are used in order to influence behavior. As a result, this ends up with comprehensive planning and control cycles where tasks and responsibilities are clear and unambiguous. Speklé (2001) found that this archetype can be divided into result and action controls, which were mentioned by Merchant and Van der Stede (2003) above.
Exploratory control is the third archetype which can be seen as a control type that has a high level of uncertainty and a low level of ex post information. Within this type of control, it is more difficult to control prescribed norms and values due to the low programmability of activities (Merchant & Van der Stede, 2003).
The last form of control, boundary control, has a high level of uncertainty and a high level of ex-post information. Speklé (2001) argues that the activities are hard to control in this situation. Boundary control can therefore be seen as a type of control that set limits in order to prevent incorrect behavior.
The specificity of resources determines the extent to which the activities are specific and to which extent the organization is able to steer itself in the desired way. When the programmability of activities is high, it is clear what results will be reached by the activities. Ex post information involves the extent to which it is possible to determine the delivered performance. These three features are, according to Speklé (2001), crucial for an efficient MCS. Figure 1 below provides an overview.
11
3.4 Isomorphism
Institutional theory offers unique insights into the ways that organizations react to institutional processes (Oliver, 1991). Institutional pressure forces organizations to integrate commonly accepted norms and concepts to gain social legitimacy, which inevitably leads to isomorphism (Brandau et al. 2015). The isomorphic pressures can be split into three types: coercive, mimetic, and normative isomorphism (DiMaggio & Powell, 1983). Each of them will be discussed below.
Coercive isomorphism emerges through social pressure and norms that are enforced by governmental and other powerful organizations (DiMaggio & Powell, 1983, Brandau et al. 2015). This type of isomorphism may force organizations to adopt enforced strategies, introduce particular quality standards, or implement information systems (Cheng & Yu, 2008). The main driver of coercive isomorphism is legislation (Brandau et al. 2015). However, the WMO works as a border. Within this border, municipalities receive significant freedom in executing the task and only have to justify their decisions to their local council (VNG, 2014). Therefore, the border legislation for the WMO does not seem to play a role and is left out for this research.
Mimetic isomorphism occurs when organizations imitate practices, processes, and structures that are employed by a successful organization to counteract uncertainty (Brandau et al. 2015). Uncertainty ensures that organizations look to their peers, particularly the leading peers, because it is assumed that those peers have already developed appropriate solutions (Benders et al. 2006). For example, a municipality that faces significant uncertainty could look to innovative neighbouring or regional municipalities (Cheng & Yu, 2008). Mimetic isomorphism therefore occurs by imitating best practices and has a “follow-the-leader” mentality (DiMaggio & Powell, 1983). Elnatan et al. (1996) conclude that organizations imitate the legitimate and successful practices of competitors and therefore benchmark themselves frequently. As a result, benchmarking is an important driver of mimetic isomorphism.
Normative isomorphism results from the creation of social network standards and business professionalization (Galaskiewicz, 1985). DiMaggio and Powell (1983, p.152) define professionalization as “the collective struggle of members of an occupation to define the
conditions and methods of their work.” Professional training institutions such as consultants
are important drivers of normative isomorphism. To strengthen the formation of social expectations regarding appropriate practices, professional training institutions develop and promote common ideas (DiMaggio & Powell, 1983).
However, there is a tendency in sociology for organizations to become more similar without necessarily becoming more efficient (DiMaggio & Powell, 1983). This trend is supported by Meyer et al. (1977) who posit that organizations are more evaluated on the basis of legitimacy rather than efficiency. Furthermore, DiMaggio & Powell (1991) found that organizations strive to be efficient, but that the drive for efficiency comes from internal and external pressures from the environment. For example, competitors exert enormous amounts of pressure on organizations in order to force them to conform to particular legitimized models.
12 This is reinforced by Meyer and Rowan (1977, p.343), who state that “organizations are
driven to incorporate the practices and procedures defined by prevailing rationalized concepts of organizational work institutionalized in society.” In addition, organizations
maintain a strong emphasis on similarity that is neither a result of competition nor an objective requirement of efficiency; instead, it is caused by attempts to attain legitimacy (Mizruchi & Fein, 1999). Conformity to institutional rules can impact organizational systems. For example, myths of best practices are supposed to be rationally effective and hence induce a legal mandate status. Although mandated based on rational assumptions of efficiency, these best practices are not always efficient for every organization. This may result in the development of internal organizational strategies aimed at buffering against the prevailing best practices. This leads to the following propositions:
Proposition 1a: Mimetic isomorphism leads to less efficient MCS in the execution of WMO
care services for municipalities.
Proposition 1b: Normative isomorphism leads to less efficient MCS in the execution of WMO
care services for municipalities
3.5 Practice variation
The study of practice variation becomes interesting when studying organizational heterogeneity (Lounsbury, 2008). By broadening the scope beyond isomorphism and thus incorporating practice variation, the relationship between institutional forces and micro-processes can be established. Whereas isomorphism examined how organizational fields and individual organizations react to isomorphic pressure, practice variation focuses at convergent change processes (Scott, 2001).
Cruz et al. (2009) found that organizational fields are comprised of multiple logics and that the existence of these diverse logics enables actors to segregate and distinguish themselves. Lounsbury (2008) emphasizes that in order to demonstrate the power of logics, it is necessary to demonstrate how a multiplicity of broader cultural beliefs and rules can structure cognition and guide decision-making for actors in a field.
Hopper and Major (2007) found that organizational actors can transform imposed accounting practices at any time during the enactment of those practices. Managers can experience contradictions and conflicts resulting from tensions between institutional and technical pressures. These variations in the imposed practices are typically conceptualized as stemming from actors resisting institutionalization, or in other words, following divergent change processes. Additionally, Ahrens and Chapman (2007) performed extensive fieldwork that highlights how management controls are a resource for action. This analytical strategy promises the generation of new knowledge and insight that will explore the relationship between institutional forces and micro-processes.
Practice variations are also influential for organizations. Benito (2010) investigates practice variations between municipalities based on two measures: size and political sign. First, he found that size could be one explanation for difference, because the greater the population, the greater the dedication to more responsibilities. Accordingly, there will be less efficiency in
13 the provision of the service. This is reinforced by Giménez and Prior (2007), who argue that larger size is linked to greater inefficiency. However, size is not used in this research, since this research mainly focuses on small municipalities.
A second explanation is based on the political sign of municipalities. Cusack (1997) found that ideology matters in the public sector. These findings supports the so-called “convergence”, which argues that industrialized societies of the twentieth century have become increasingly similar in that they face the same kind of problems and thus use the same kind of solutions. Consequently, political differences do not matter much in explaining variations in efficiency. However, Borger and Kerstens (1996) conclude that socialist party participation in municipality coalitions means a higher degree of efficiency than liberal party participation.
Additionally, Brand (2005) found that practice variations can also occur due to differences in steering municipal features. Municipalities receive the total budget based on several factors like population quantity and total surface of the municipality. However, municipalities can have “luck” or “bad luck” depending on how the distribution key for the budget is established. Brand (2005, p.3) notes that “the economic theory applicable to decentralized systems of
government suggests a balanced approach to the distribution of financial resources and obtains the most efficient and equitable solution.”
Another interesting aspect is the level of social cohesion within a municipality. Tolsma et al. (2009) found that the mean income of the neighborhood is the most important measurement. The authors found that residents have more contact with their neighbors in more prosperous neighborhoods. As a result, there is a higher level of trust, and people are more willing to help each other. Practice variations can therefore also occur due to differences in social cohesion. The literature assumes that practice variations are needed at the local level. The practice variations that are introduced should therefore also deal with the efficiency aspect. Hence the following proposition:
Proposition 2: Practice variations lead to more efficient MCS in the execution of WMO care
services for municipalities.
3.6 Towards the conceptual model
Literature has demonstrated that varying socially constructed environments may lead to different imperatives for care services. On one hand, pressures from isomorphism can influence municipalities to adopt similar practices. On the other hand, practice variations can be used to explain why organizations adopt different practices. Since both pressures appear to be present within the WMO of municipalities, it is interesting to investigate how both pressures influence the MCS of municipalities. This is mapped in the conceptual model below.
4.
Methodology
This chapter elaborates on the methodology of the research by describing the research method, method of data collection, and data analysis. In addition, attention has been paid to the controllability, reliability, and validity of this research.
4.1 Method of data collection
Central to this research is the following research question: How do isomorphic pressures and
practice variation influence the management control system (MCS) of municipalities?
This research limited its scope to the public sector and only focuses on the three new WMO tasks. In order to provide direction, the conceptual model presumes that legitimacy factors do influence the phenomenon. However, how those factors influence the phenomenon is yet unclear. This how will be investigated by performing a qualitative study. A qualitative study is useful when practical insights are obtained (Edmonson & McManus, 2007). In addition, Parker (2003) suggests that qualitative research can offer a deeper understanding of management control issues that are associated with change.
This research also assures replicability and controllability by providing a clear description of the research execution so that different researchers in different time periods can generate similar findings (Smith, 2017). Controllable research can be replicated by other academics by representing results as precisely as possible (Smith, 2017).
The data that is used in this study was collected via semi-structured in-depth interviews and by analysing documents to control for instrument bias (secondary data) (Smith, 2017). The internal documents that were used are listed in references.
4.2 Interview method and development of interview questions
This qualitative research made use of semi-structured interviews to collect data. The advantage of semi-structured interviews is that they have an open structure, which means that events are easier to recognize (Smith, 2017). The author further states that semi-structured interviews consist of questions that cover all the information without setting boundaries. Because the research questions are not given in fixed order, there is room to ask additional questions as the interviewer sees fit. In so doing, the interviewer can examine associated issues that arise in the interview (Smith, 2017). Moreover, open questions avoid the provision of suggestions of a certain answer (Ahrens & Chapman, 2007).
The development of interview questions is based on the conceptual model. First, questions were asked based on the distinction of Speklé (2001): the degree of asset specificity, extent of programmability, and the intensity of ex post information. Second, questions were asked regarding both mimetic and normative isomorphic pressure. Third, questions were asked regarding practice variations. Questions were also asked on political preference, steering on municipal features, and social cohesion. The interview questions are listed in Appendix 1, and Table 1 below provides the topic list.
16 Efficiency factors
- Asset specificity
- Extent of programmability - Intensity of ex-post information Isomorphism
- Mimetic isomorphism - Normative isomorphism Practice variation - Political sign
- Steering on municipal features - Social cohesion
Table 1 : Topic list
4.3 Selection of respondents
This research conducted six interviews although it related to ten municipalities. The reason for this is that the Bedum, De Marne, Winsum, and Eemsmond municipalities will merge into a single “Het Hogeland” municipality in 2019. There is, however, just one financial controller that currently controls the WMO for all four municipalities. Also, the municipalities of Veendam and Pekela have one financial controller. This research will use the abbreviations BMWE and P&V to refer to a quote from these financial controllers.
This research analyzes the results based on the number of municipalities instead of the number of interviews. Each interview lasted between 50 and 70 minutes; the transcripts are single spaced and are approximately seven pages long. This research focuses on the efficiency aspect of MCS, and therefore the interviewees were all financial controllers. In addition, there was one interview conducted with a policy officer. This research follows Eisenhardt (1989), who argues that it is best to use cases that are the same in most areas, but differ significantly in specific and important areas. This makes sure that conclusions are meaningful, since all cases are related, but that possible changes can be attributed to the important differences among the cases.
This research mainly focuses on municipalities in the north of the Netherlands, where most of the municipalities are small. Korsten (2015) found that small municipalities receive smaller budgets and are slower at changing processes and structures compared to larger municipalities. It is therefore interesting to investigate how small municipalities are dealing with the new WMO tasks. For this research, 21 municipalities were approached that were distributed over Groningen, Drenthe, and Friesland. Ultimately, 10 municipalities participated. An overview of this information is provided in Table 2.
To assure external validity, results must be generalizable. This means reducing results to a general form that can be applicable to the whole research field. Generalizability is assured by conducting interviews with municipalities that are located across the Netherlands broadly. This research does not include care organizations for two reasons. First, this research does not focus on the relationship between care organizations and municipalities. Second, this research focuses on the MCS of municipalities, not the MCS of care organizations.
17
General information Function of interviewee
Municipality Inhabitants Abbreviation Financial/Controller Policy officer
Bedum 10.477 BMWE x De Marne 10.083 Eemsmond 15.864 Winsum 13.610 Pekela 12.708 P&V x Veendam 27.497 Aa & Hunze 25.337 - x Delfzijl 24.961 - x x Groningen 202.567 - x Oldambt 38.108 - x
Table 2 : Participating Municipalities
4.4 Data analysis
A large collection of information was derived from the interviews. To assure controllability, all interviews were conducted in Dutch, audio recorded, and transcribed. The transcripts were used to interpret interview data per section and per question, so as to make connections between municipalities regarding similarities and differences. Statements that were frequently mentioned were presented in the results. In addition, the interview data gathered from interviews were checked on supportive grounds; internal documents were used to strengthen the analysis that was used to answer the propositions. Using multiple instruments cures biases and shortcomings. The multiple instruments complement and revise each other (Smith, 2017). Furthermore, by comparing interview data with internal documents, internal validity increases. Eisenhardt (1989) states that using multiple types of data collection leads to stronger propositions and constructs.
The interviews found that the differences between the three tasks were negligible. The three tasks were all monitored and controlled in the same way. Therefore, the three tasks were treated as one and the same.
18
5.
Results
This chapter presents the interview results. Section 5.1 presents the results pertaining to transaction cost theory (TCT), Section 5.2 presents the results on isomorphism, and Section 5.3 presents the results on practice variation. The results establish an image of a municipality’s MCS and whether it has different or similar practices. The data was analyzed in the same order as the interview questions, and section headers were used to provide a structured overview. Findings are illustrated by quotes of interviewees.
5.1 Transaction cost theory (TCT)
In order to describe the efficiency of the WMO, questions were asked regarding the specificity of the WMO, the programmability of activities, and the extent to which ex post information can be collected.
5.1.1 Asset specificity
To investigate the asset specificity of the new WMO tasks, questions were asked on the specificity of the new WMO tasks.
Estimating needed care
All municipalities determine needed care via a “keukentafelgesprek.” This “keukentafelgesprek” is a conversation between the WMO policy officer and the client that is intended to assess the level of needed care. The WMO policy officer subsequently determines what elements of care can be handled by neighbors, family, or other acquaintances. The residual elements become the responsibility of the municipality. All municipalities also describe the WMO as legally mandating custom-made care.
Replacing a care provider
The interview results suggest that all municipalities are able to replace a contracted care provider. For example, the financial controller of BMWE stated that “If one goes bankrupt,
other care organizations should dive into this gap. (...) This is not an issue, since there are many alternatives available.” This is in line with the statements of the financial controller of
P&V: “Care providers are screaming to receive new clients.”
Comparing prices between care organizations
All municipalities are able to compare prices between care organizations. For example, the financial controller of P&V said that “Prices are partly in control because you set limits via
the public-tender.” Although prices are partly in control, subjective elements make
comparisons sometimes difficult. For example, the financial controller of municipality of P&V said that “There are so many factors not defined, meaning that we struggle with our
contract partners about the fair value every year. (…) For example, how much overhead do you find fair? The care provider finds 50%, and we 0%, but what is the truth? In line with
previous quote, the policy officer municipality of Delfzijl said, “We use standardized prices
for each type of care, which is suitable to a large part of our clients, but where it does not work, we have to make separate appointments.”
19
Comparing quality between care organizations
All municipalities have almost four years of experience with the new WMO tasks and use this experience to measure quality. For example, the financial controller for the municipality of Groningen said, “Differences between care providers are negligible. (…) We have filtered the
bad care providers out of our system.” The financial controller of BMWE said, “We now have the history of which care provider is the best in executing each type of care.”
Comparing delivered care between care organizations
Standardization of care makes all municipalities able to compare delivered care ex post. For example, the financial controller of Oldambt said, “We try to standardize our public tender to
make it possible to compare types of care between care providers.” Although standardizing
make comparisons possible, the WMO still remains custom-made. Custom-made care makes it sometimes difficult to compare delivered care as is illustrated by the statement of the policy officer for Delfzijl: “If a client has more indications, it is questionable the extent to which you
can place this client in a box.” In line with the previous quote, the financial controller of
Oldambt said, “Standardizing custom-made care always leads to discussion.”
5.1.2 Programmability of activities
Questions were asked to investigate the programmability of the new WMO tasks.
Predictability of total care
Interview results were equally divided between municipalities. Five municipalities mentioned the open-ended arrangement characteristic. For example, the financial controller for the municipality of Delfzijl said, “It is an open-ended arrangement and therefore we cannot
predict precisely how many clients will need help in the upcoming year.” The financial
controller for the municipality of P&V said, “It is difficult to predict the amount of new
indications and the amount of indications that will not be extended next year”. As a result,
more emphasis was placed on early signaling and prevention. By so doing, municipalities become better able to predict care and help the client at an early stage.
However, on the contrary, five other municipalities are better able to predict the total costs due to a more stable WMO-database. For example, the financial controller of BMWE said:
“The WMO is predictable, since the WMO does not show much fluctuation in clients, but instead is very constant.” The financial controller for the municipality of Aa & Hunze said: “We have a basic database, which is filtered and therefore results in a customer base that is more stable.”
Estimating costs
All municipalities used the acquired experience to estimate the costs. An example is given by the financial controller for the municipality of Aa and Hunze: “I look back in history to see
where differences occur and indicate if this occurred due to an incidental issue or a structural one, which we translate into our new estimation.” In line with the previous quote, the
financial controller of BMWE said, “We now have the management information on such a
20
Usage of information before 2015 and changing control method
All municipalities gathered information from prior healthcare offices before the start of 2015. However, controlling the WMO with this information remains difficult. For example, the financial controller for the municipality of Groningen said, “We received encyclopedias full
of information, but we do not have a clue where to start. As a result, we were swimming and steering was a total joke.” All municipalities therefore changed working methods and
forecasts several times in that year. An example is given by the financial controller for the municipality of P&V: “It became known that numbers were not realistic, with the result that
our calculations did not match reality. Therefore, our method changed at least three times in that year.”
Integrating efficiency
All municipalities used the public tender to integrate efficiency. For example, the financial controller for the municipality of Aa & Hunze said, “We calculate an hour-price per type of
care based on the budget and knowledge of the number of clients.” Although efficiency is
integrated in a similar way among municipalities, different funding methods were found. Nine municipalities calculated the price via PxQ, but the municipality of Groningen used lump sum.
5.1.3 Ex-post information
To investigate the ex-post information of the new WMO tasks, questions were asked relating to the amount of ex-post information that can be collected for the new WMO tasks.
Measurability of results (efficient)
Results show that all municipalities mainly use the basic financial parameters. For example, the financial controller for the municipality of Aa & Hunze said, “We are recording numbers,
amounts, and euros, but this is more based on recording instead of using that for controlling the task.” In line with the previous quote, the financial controller for the municipality of
Oldambt said, “The care provider has to deliver a production schedule for each type of care
where numbers, length of indication, and money used are reported. (...) Since there an audit report from an accountant is usually included, we trust these numbers for controlling these tasks.”
Monitoring to stay within budget
It appears from the interview results that all municipalities monitor the indications and use the “verzilveringspercentages,” which determines the percentage of the available budget that is already used for an indication. For example, the financial controller for the municipality of Oldambt said, “When the municipality has €100,000 as total amount for indications, and the
verzilveringspercentage is 80%, the financial obligations will be €80,000.” Although
monitoring the budget was deemed successful based on the basic financial parameters, controlling the task was still difficult. For example, the financial controller for the municipality of Delfzijl said, “You only see budget and percentage used, but how much care
21
Controlling on financial and non-financial information
In line with previous question, all municipalities were still developing their non-financial information. Using financial and non-financial information strengthens the analysis. This was illustrated by the financial controller of Delfzijl: “The production schedule has the basic
parameters, like numbers, hours, and money. This can be used as a good starting point, but it should be extended in more detail, both financially and non-financially, to get more control.”
Another example is given by the financial controller for the municipality of Oldambt: “The
control part for this task was purely financially-oriented in the first years, but we are still developing more non-financial information to extend our analysis.”
Controlling quality
Eight municipalities doubted the way they measured quality. For example, the financial controller of BMWE said, “If a client gives the quality an 8, does it mean that the quality is
good? I do not think so, because this is what the client perceives, but one client is satisfied more quickly than another.” In line with the previous quote, the financial controller for the
municipality of Groningen said, “We find that client satisfaction surveys do not work, since
all care providers score an 8 on quality.” Besides this difficulty, measuring quality is also
difficult due to privacy concerns. An example is given by the financial controller of P&V:
“Quality is hard to measure based on privacy. (...) In order to measure quality correctly, you should know the content of care, but this stays in the consulting room because of privacy.”
5.1.4 Conclusion
The interview results were used to classify the asset specificity, programmability, and ex post information. The specificity of WMO tasks can be specific and thus qualified as high or more general and thus qualified as low. Furthermore, when the programmability of activities is high, it is clear which results will be reached; when the programmability is low, it is less clear. Additionally, a high level of ex post information indicates that the municipality is able to perfectly determine delivered performance; a low level of ex post information suggests the opposite.
The first section on asset specificity demonstrates that the new WMO tasks are all custom-made. However, municipalities try to standardize care as much as possible via the public tender, making it possible to compare between care providers based on price, quality and delivered care. In addition, municipalities can easily replace a care provider if they have to. However, interview results also suggest that standardizing custom-made care leads to discussion for some municipalities. Therefore, the asset specificity is categorized as low to medium.
The second section on programmability of activities suggests different insights. Half of the municipalities mentioned that predicting care was difficult because of an open-ended arrangement, whereas the other half could easily predict due to a constant WMO database. Also, all municipalities received significant amounts of information from care providers, but could not control it. Finally, municipalities integrated efficiency into price and therefore have control over efficiency. Therefore, the programmability of care and costs is assumed to be medium.
22 The third section on ex-post information demonstrates that municipalities control the new tasks based on basic financial parameters and monitor the amount of indications and verzilveringspercentages. However, municipalities were still developing their non-financial information to extend the analysis. Furthermore, results suggest that more than half of the municipalities faced difficulties in measuring quality. Therefore, the ex post information is assumed to be medium.
Based on the classification of low to medium asset specificity, medium programmability of activities, and ex-post information, municipalities are assumed to use exploratory control to control the WMO tasks. An overview is provided in Table 3.
Asset specificity IQ Answers NM Low Medium High
Estimating needed care 1, 2 Determining care via the “keukentafelgesprek” 10/10 x
Replacing a care provider 3 Many alternatives available in market 10/10 x
Comparing prices 4 Prices are partly in control due to public tender
Subjective parts in calculation
10/10 2/10
x
x
Comparing quality 5 Quality measured on experience with care providers 10/10 x
Comparing delivered care 6 Difficult to place a customer in a box
Standardization of custom-made care leads to discussion
4/10
3/10 x
x
Programmability of activities IQ Answers NM Low Medium High
Predictability of total care 7 Difficult due to an open-ended structure
Constant WMO database
5/10 5/10
x
x
Estimating costs 8 Based on larger portion of history 10/10 x
Usage of information before 2015 and changing control method
9, 10 Receive lots of information, but cannot control on it
Changing method due to uncertainty of what was yet to come
10/10 10/10
x x
Integrating efficiency 11 Integrated in hour price 10/10 x
Ex-post information IQ Answers NM Low Medium High
Measurability of results (efficient) 12, 13 Controlling on the approved basic financial parameters from the accountant
10/10 x
Monitoring to stay within budget 14 Monitoring indications and verzilveringspercentages 10/10 x
Controlling on financial and non-financial information
15 Developing non-financial information to extend analysis 10/10 x
Controlling quality 16 Measuring perceived instead of real quality
Difficult due to privacy concerns
8/10 2/10
x x
5.2 Isomorphism
To investigate the influence of isomorphic pressures, questions were asked relating to the isomorphic practices in the execution of the WMO. Also, the interview results were strengthened by internal documents and presented at the end of the section. The following results occurred.
5.2.1 Mimetic isomorphism
Working together with other municipalities in the execution of new WMO tasks?
All interviewees stated the importance of cooperation and discussion between municipalities. For example, the financial controller for the municipality of Aa en Hunze said, “We have
meetings with the other four municipalities and by talking to each other, we learn from each other.” In line with the previous quote, the financial controller for the municipality of
Groningen said, “Before the start of 2015, we went to Eindhoven, Amsterdam, and even to
Denmark to see if we may learn something.”
What do municipalities together? Where does it differ?
All municipalities copied each other’s practices in case of uncertainty. Municipalities were constantly observing other municipalities and copied those practices that were in line with the way they wanted to control the WMO tasks. Results suggested that nine municipalities work together via procurement practices. Also, two municipalities worked together to improve their information system and in particular the visualization tool.
Besides similarities, differences between municipalities were also established. For example, the policy officer municipality of Delfzijl said, “We are working together with Loppersum
and Appingedam, but our policies differ due to different rules.” The financial controller for
the municipality of Groningen said, “We reduce the choice of freedom and reduce the WMO
from 140 contract partners to 10.” Both differences were related to policy, but a difference in
organization structure was also found by two municipalities. For example, the financial controller of Aa & Hunze said, “We went to our neighbour municipality Borger-Odoorn for
their foundation concept to see if we can implement it as well.” Lastly, a difference regarding
norms and standards was found, as is explained by the financial controller of P&V: “The way
we take responsibility for the calculation part of the WMO is because we are ambitious officers.” Due to a different motivational spirit, differences were found in care execution.
Was the municipality a good practice?
Seven municipalities were not approached by other municipalities and therefore did not see themselves as having best practices. On the contrary, the municipality of P&V said, “We are
often approached by other municipalities on how we organize our policy.” However, a
neutral point of view is illustrated by the financial controller for the municipality of Groningen: “We can be a best practice for some municipalities, but we are also looking to
other municipalities to find the best practice. (...) Therefore, I do not think that there is one best practice.”
25 Internal documents
Mimetic practices in the collaboration of procurement practices for the WMO can also be found in the internal documents. For example, eight municipalities procured together via the “Ommelandersamenwerkingsmodel.” These municipalities worked together with pilot partners that supported the municipalities in the execution of the WMO (p.3). The municipality of Aa en Hunze procured together with five other municipalities, as was found in the “Beleidsplan Staan 2017.” This report suggested that municipalities work together in the execution of the WMO in order to increase effectiveness and efficiency (p.1). The collaboration in procurement is based on three measures: reduce costs, improve quality, and improve knowledge. This suggests that the goal of the WMO is to become more efficient while improving quality.
5.2.2 Normative Isomorphism
Did the municipality use the help from professional groups due to uncertainty?
Results indicate that the Vereniging Nederlandse Gemeenten (VNG) has been investigated as the most used organization that provides professional help. This is illustrated by the statement of the financial controller for the municipality of Aa & Hunze: “Look, as the VNG elaborates
a model to deal with a negative trend, we will implement it, since the VNG represents all municipalities.” In line with the previous quote, the financial controller of P&V said, “The VNG publishes, has an online platform, and keeps municipalities updated about latest updates.” Four municipalities noted the importance of the forum discussion. An example was
given by the financial controller for the municipality of Groningen: “We are on the website of
the VNG where we share information with other municipalities, and we learn from each other, really.”
How did municipalities use the help from professional groups and why?
Four municipalities mentioned the help from the advisory board. This advisory board consists of local citizens. However, this board functions differently. For example, the financial controller of P&V said, “The advisory board of the WMO consists of elderly people (over 65
years of age). The result is that they are not always present during meetings and besides that, are not always capable to understand the complex task.” In contrast, the financial controller
of Aa en Hunze said, “Before a municipal meeting, the advice from the advisory board is
discussed, which happens regularly.” Furthermore, two municipalities used help from local
professional organizations once the care request becomes too complicated. Internal documents
The interview questions mentioned the use of the advisory board. However, each municipality is obliged to have an advisory board that consists of citizens. Hence, although the board is mentioned in the interview questions, it cannot be recognized as a professional group. But, since the board is mentioned in the interview questions, internal documents will be used to strengthen the interview results. In the policy plan for BMWE in 2015-2018, the BMWE municipalities mentioned the importance of the WMO advisory board (p.36). However, the advisory board received too many questions and therefore lost its value (Nieuwe adviesraad Jeugd & Wmo BMWE gemeenten, 2018). In line with this, the advisory board in the municipalities of P&V should also be improved. The alderman (p.1) report in “SeniorenRaad
26 Pekela & Veendam 2017” that the advisory board should be restructured for greater value. On the contrary, the “Adviesraad Social Domein Aa en Hunze” suggested that the advisory board works and several recommendations since October 2015 were discovered.
However, the interview results did not mention the advisory board in other municipalities. In order to demonstrate this aspect, internal documents were used. For example, the municipality of Groningen has Stadadviseert. In the report titled “Stadadviseert 2017,” it is mentioned that this board is split into three parts: (1) an urban advisory group consisting of 50 persons; (2) an expertise pool; and (3) collaboration with knowledge parties. Furthermore, the advisory board in the municipality of Delfzijl was also not mentioned in the interview questions. However, the municipality of Delfzijl does in fact have an advisory board called “DAL adviesraad sociaal domein.” This board consists of 10 citizens from Delfzijl, Appingedam, and Loppersum. Finally, the advisory board of the municipality of Oldambt was also not mentioned in the interview results. However, Oldambt also has an advisory board called WMO-Raad Oldambt 2017, consisting of 14 persons. The internal documents of these three municipalities demonstrate that all advisory boards discussed several meetings and therefore attached value to municipalities in the execution of the WMO. However, it is important to qualify this by noting that support was only found in internal documents.
5.2.3 Homogeneity in the WMO due to isomorphism?
The interview results and internal documents were used to classify isomorphic pressures. High levels of mimetic isomorphism indicated significant uncertainty and attempt to imitate other municipalities in order to counteract that uncertainty. On the contrary, municipalities were less triggered to imitation if they faced less uncertainty. High levels of normative isomorphism indicated strong professional help, whereas low levels indicated a low usage of the professional standards or networks that influence change. An overview of the classification is presented in Table 4 below.
The first section on mimetic isomorphism suggested that municipalities strongly cooperate by visiting other municipalities and through procurement practices. Furthermore, municipalities cooperate by implementing structural changes such as the foundation concept and by improving the visualization tool. However, almost three-quarters of the municipalities mentioned that they were not approached and seen as engaging in best practices. Therefore, mimetic isomorphism was categorized as medium to high.
The second section on normative isomorphism suggests that municipalities strongly used the help from professional groups, particularly the VNG. The VNG represents all municipalities and also offers a national platform. Furthermore, ten municipalities used the advisory board, but improvement to the advisory board should be made in six municipalities to attach more value to it. Therefore, normative isomorphism was categorized as medium to high.
Proposition 1a describes mimetic isomorphism as leading to less efficient MCS in the execution of WMO care services for municipalities. There appears to be a strong form of mimetic isomorphism in the WMO. Municipalities attach value by visiting other municipalities to see if they may learn from each other. In addition, nearly all municipalities procure together, meaning that the method of care organization has strong similarities.
27 Proposition 1b describes normative isomorphism as leading to less efficient MCS in the execution of WMO care services for municipalities. There appears to be a strong form of normative isomorphism in the WMO. In addition, municipalities attach value to the advice of professional groups and in particular the VNG. Furthermore, municipalities also use the help of advisory board, but this does not seem effective in all municipalities.
Both propositions assume that municipalities strive to be efficient and are therefore imitating each other or using help from professional groups. The literature assumes, however, that the drive for efficiency comes from internal and external pressures of the environment that exert enormous pressure on organizations to force them to conform to particular legitimized models.
The interview results and internal documents indicated that all municipalities still face uncertainty and are therefore imitating other municipalities or using help from professional groups to counteract that uncertainty. However, instead of conforming to legitimized models, practices, and procedures, all municipalities are instead searching for the best and most suitable practice to execute the WMO. In addition, the VNG also helps municipalities to execute the WMO with budget cuts. The isomorphic pressures help municipalities to deal with the WMO and therefore become more efficiency-oriented. This is in contrast with the literature. As a result, there does not seem to be support for proposition 1a and proposition 1b.
