New public service in the Dutch social housing sector : an exploration of the value of new public service in the governance of social housing in the Netherlands

(1)

1

New Public Service in the Dutch social housing sector

An exploration of the value of New Public Service in the governance

of social housing in the Netherlands

Master’s thesis Political Science

Author: Kees van der Zee

Mentor: Dr. A. Reynaers

Second reader: Dr. I. Verhoeven

Date of completion: 24-06-2016

University of Amsterdam

(2)

(3)

(4)

(5)

5

Preface

When I enrolled in the course ‘Managing Public and Private Relationships, What’s the Difference?’, I was looking for a challenging problem in the Dutch policy sector to explore in detail. When thinking about a subject, I thought about an episode of the Dutch talkshow ‘Zondag met Lubach’. It discussed the behavior of the management of Rochdale, a housing corporation in Amsterdam. The message of the show was that housing corporations are focused on self-enrichment and that they neglect their public task. The public moral in the sector was gone (Lubach, 25-10-2015).

When delving in to housing corporation policy, I found out that much effort has already been put in reforming the sector and changing the behavior that has led to the incidents that were criticized in the TV show. The interesting thing was that the information raised more questions than answers. It was not very clear how the sector is organized, on what ideas reforms were based or how the central problems in the sector could be explained. Therefore, I chose social housing policy as my thesis subject, to spend the next five months finding out how the sector is organized and what problems have caused incidents in the housing sector.

I chose to study political science in the first place to contribute to the improvement of governance, so the work I would write on the housing sector also had to contribute to the improvement of housing policy. That is why this research roughly consists of two central parts: what are the problems in the social housing sector and how can they be solved?

In the beginning of the research, most of my time was focused on finding out how the policy works. This gave me insight in the rich historical tradition of the corporation and how important they are to the Dutch society. After I had gained a basic understanding of the sector, I started interviewing people about their view on problems and potential solutions. This gave me an even richer insight in the current governance processes. It showed the viewpoint from the another side, of the people trying to deal with the problems they face.

I want to thank all the people who were willing to make time for me and share their insights, it was very inspiring and it made doing research more interesting. I also want to thank my family and fellow students, who gave me very helpful feedback to improve the structure and content of my texts. And lastly I want to thank my mentor, who really helped out in sharpening my research focus and the coherency of the thesis.

(6)

(7)

7

5.2.2. Amount of steering ... 55 5.2.3. Effect of (de)centralization ... 56 5.2.4. Level of competition ... 57 5.2.5. Definition of goals ... 58 5.2.6. Result orientation ... 59 5.2.7. Tenant approach ... 60 5.2.8. Enterprising behavior ... 60 5.2.9. Anticipatory policy ... 61 5.2.10. Market functioning ... 62 5.2.11. Conclusions ... 62 Chapter 6: Conclusions ... 67

6.1. Answer to the main question ... 67

6.2. Discussion ... 70

6.2.1. Implications for the social housing sector ... 70

6.2.2. Implications for NPM and NPS ... 71

6.3. Suggestions for further research ... 73

(9)

9

News articles and television shows ... 79

Websites ... 80

Appendix 1: Interview structure ... 81

Appendix 2: Coding categories ... 82

(10)

10

List of abbreviations

AW* Authority Housing Corporations

AFWC* Amsterdam Federation of Housing corporations

BBSH* Decision on Management of Social Housing

CCM Constant Comparison Method CFV* Central Fund for Social Housing

DAEB* Services of General Economic Importance (non-commercial)

NPM New Public Management

NPS New Public Service

PEW* Parliamentary Inquiry Commission

PVT Public Value Theory

SER* Social and Economic Council

WSW* Guarantee Fund for Social Housing * Abbreviation of Dutch concepts.

(11)

11

Executive summary

Social housing governance in the Netherlands is influenced by New Public Management (NPM) since the 1990s. This policy model provided an answer to the problems of the bureaucratic organization that governed corporations before (Van der Schaar, 2009: 42). NPM itself has been the cause of several problems in the sector in later years. Bad governance by corporation managers, lack of supervision and poorly functioning market mechanisms have led to costly incidents (Gerrichhauzen, 2014: 17). This thesis investigates what problems are caused by NPM and how the New Public Service model can contribute to a better form of governance in the sector.

The shortcomings of NPM are investigated with an analysis of policy documents and scholarly work on the sector. An overview of the influence of different aspects of NPM is established. This analysis shows that the earlier reforms were mainly focused around decentralization and a retreat of the state. In a later stage, the state enlarged its influence on the sector again. It steered the market more actively. The analysis shows that the NPM ideas of decentralization and influencing the market contradicted each other in social housing policy.

The second part of the thesis focuses on a case study of De Key, a housing corporation located in Amsterdam. In the case study, the potential of NPS to improve governance in social housing policy is investigated. This is done by interviewing people working for or with De Key. Respondents stated that dialogue with stakeholders, particularly tenants, is a potent method to improve management quality. Introducing a form of network governance could also lead to an improvement of governance. The case shows that corporations are not equipped for, and unwilling to act as result-oriented, competing enterprises.

Some NPM doctrines are suitable to be (partially) maintained in the sector. Goal setting and result orientation have had a positive effect on corporations and need no alternative governance ideas. Other doctrines should be replaced with NPS values. The social housing sector is for example not suitable for competition between corporations, which led to a steering role of the state. The steering role of the state could be replaced with network governance, in which other stakeholders are able to provide input in decision making.

(12)

(13)

13

Chapter 1: Introduction

1.1. A sector in transition

In the 1990s, the social housing sector of the Netherlands was made independent from the state. This operation was called the bruteringsoperatie1 (hereafter brutering). State loans of corporations were remitted, while subsidies were abolished. This put an end to the expensive, bureaucratic social housing system that pressured state budgets (De Jong, 2013: 14-15). Corporations were enabled to operate independently so they had more room to compete with each other to have better performing businesses.

The way the sector became arranged is to a large extent based on the New Public Management (NPM) model. NPM is a set of doctrines that aims to deal with problems of traditional bureaucracies by imposing market-oriented reforms. The monopolistic bureaucracy had to be replaced by more independent and flexible organizations that would compete with each other. These forms were shaped more like private companies, with a clear set of goals and a focus on output (O’Flynn, 2007: 354). Corporations were expected to adapt to this paradigm. The result was a large reduction in public expenditure, which reduced the pressure on the national budget (Van der Schaar, 2009: 43).

The transition from bureaucratic dependent organizations to more enterprising companies has been challenging for corporations. Supervisory institutions of the independent corporations failed to recognize problems and mismanagement in time. Mismanagement has in some situations led to costly incidents. A famous example is the purchase of a ship, the SS Rotterdam, by housing corporation Woonbron. The ship was purchased for €1,75 million, the plan was to be rebuild it into a congress-center, hotel, restaurant and several other things. Woonbron estimated a total rebuild cost of €24 million (Gerrichhauzen et. al, 2014: 6). The corporation simply started the process without a thorough plan or contractual guarantees. This led to a total cost of €257 million in the end.

This example shows the broader problems that exist within the sector and can be related to NPM. The first problem is poor management. Corporations CEO’s engaged in costly projects which were not carefully planned out or based on unfounded commercial ideas (Ibid, 6-7). Decision making authority was centralized in the corporation’s management which could hardly be impeded by supervising institutions. The second

1

The bruteringsoperatie or grossing act (freely translated) was a large financial transaction in which the ties between the state and housing corporations were cut.

(14)

14

problem was poor supervision. Both internal and external supervisory institutions lacked the capability to put a halt to escalating problems. The third problem was an uncritical and optimistic culture within the sector and related political institutions, which was inattentive to potential threats (PEW, 2014: 54).

The call for more government intervention in the sector became louder in 2009. Some small reforms in the supervision were made, but this was not enough (De Jong, 2013: 2). Vestia, the largest Dutch corporation, announced in 2012 that it had financial problems due to controversial derivatives trade (PEW, 2014: 46-47). Dutch Parliament lost its patience and a parliamentary inquiry was announced. The Parliamentary Inquiry Committee2 (PEW) proposed a state-led sector management, which resulted in stricter regulation and supervision of the sector (Ibid: 30).

1.2. The problem

The problem with the proposed solutions of the PEW is that although they do address several problems directly, they do not question the underlying NPM paradigm. This thesis investigates if the different doctrines of NPM could be the cause of problems in the sector. Subsequently a potential alternative paradigm is investigated: New Public Service (NPS). This management model has been proposed as an alternative of NPM. NPS aims to serve citizens on a broader range of values than the narrow output performance hat is focused on by NPM. It states that competition is unsuitable for public agencies, it should be replaced by cooperation in networks with stakeholders that often work with corporations (Denhardt and Denhardt, 2000: 556). When corporations operate in networks they are accountable to more stakeholders which makes them less vulnerable to irregularities. Due to more frequent interaction in networks, corporations can be more responsive to needs and interests of tenants, municipalities and other stakeholders.

The central research question that is investigated consists of two parts. The first part of the question asks if NPM is the underlying cause of problems in the social housing sector. The second part of the question, asks whether NPS might be a viable alternative for the NPM doctrines. Combined, they form the central question of this thesis: what effect does

(15)

15

New Public Management have on the governance of the Dutch housing sector and to what extent can New Public Service be an alternative to improve governance in the sector?

1.3. Relevance

Earlier reports and scholarly work on the sector have looked at the problems in a more practical manner. They made a list of the larger incidents and analyzed what trends could be recognized in these incidents (Bieleman and Snippe, 2010; De Jong, 2013; Gerrichhauzen et. al, 2014; PEW, 2014). What was lacking in this work is the proposal of an alternative foundation on which the sector could be organized. This thesis analyses the underlying ideas when the housing policy was reformed. In this implementation the rhetoric of NPM is clearly present from the 1990s onwards (De Jong, 2013: 18). It would be good to put this rhetoric to a critical review and see if NPS is a more suitable alternative to organize the sector.

The thesis is both relevant for the governance in the sector and for the scientific discussion on the value of NPM and NPS. The findings can be used to critically review how the sector is structured. They can also recognize potential weaknesses based on the underlying NPM-paradigm. On the scientific part, potential weaknesses or contradictions in NPM can be exposed by looking at how they are realized in practice.

There is little scholarly work on how NPS can be used in practice. This thesis can be valuable in exploring potential values and challenges NPS could encounter. NPS seems to be a promising alternative to the problems caused by NPM. It is only useful if there is knowledge on how to implement it though.

1.4. Outline

In the next chapter, NPM and NPS are explained and discussed. NPM is defined in nine central doctrines, subsequently its criticism is described. The central ideas of NPS are also defined. The question why NPS is a better alternative than NPM is discussed at the end of the chapter.

In chapter three, the methods that are used to gather- and analyze data are discussed. An explanation is provided on how both parts of the central question will be approached in a different way. The impact NPM has had on the sector, is investigated with a document study. The contribution of NPS to improvement, is assessed with a case study of one housing corporation: Woonstichting de Key (hereafter De Key).

(16)

16

As the social housing sector is a very complicated sector to understand, a brief history is provided in chapter four. The process towards NPM reforms is explained at the begin. Then the consequences and reactions on the reforms are discussed.

Since the central problem consists of two parts, the results chapter will be divided into two parts as well. In the first part, the results of the document study on the influence of NPM on social housing policy are presented. The second part is based on the results of the analysis of interviews. This analysis discusses what respondents of De Key and related organizations think of the current governance and what value they see in NPS as an alternative. Both sections in the results chapter are structured around the nine doctrines of NPM, which are described in chapter two.

In the conclusion, the findings of both parts of the results chapter are summarized forming the answer to the main question. Subsequently, the potential meaning of these results for NPM and NPS in general is discussed. Lastly, suggestions for further research are made.

(17)

17

Chapter 2: Theoretical framework

The social housing sector reforms that started in the 1990s have largely been based on NPM ideas (Heerma, 1990: 2; Van der Schaar, 2009: 93). In this section, the theoretical foundations of NPM and NPS are explained. Each of the theories is firstly defined in a few central characteristics. Subsequently, criticisms and responses are discussed.

NPM has been an influential model of governance since the 1990s and it is still relevant in current policy debates. NPM has mainly been a response to old-fashioned bureaucracy and its inefficiencies. Increased focus on efficiency and results are proposed as solutions to the problems of bureaucracy (Osborne, 1993: 353-354). NPM is criticized for seeing the public sector as a private firm, for seeing citizens as clients and for measuring immeasurable values (Alford and Hughes, 2008: 6).

NPS is a reaction to NPM and aims to look beyond efficiency and results. It criticizes NPM for showing a one-sided picture of public policy (Denhardt and Denhardt, 2000: 554). NPS is a suitable alternative for NPM because some elements of it are already present in the current housing sector. This makes a more gradual transition to an alternative management of the sector possible. NPS can be explained as a combination of network governance and Public Value Theory (PVT). Network governance argues that governance is executed more effectively and democratically through networks (Sørensen and Torfing, 2009: 248). PVT argues that creating and executing policy should be done from the viewpoint of creating public value (Stoker, 2006: 46-47).

2.1. New Public Management

NPM is a set of doctrines that aims to deal with problems of traditional bureaucracies by imposing market-oriented reforms. These reforms include changing the large bureaucratic organization into smaller, more flexible and competitive organizations. This management method is largely based on how private companies work. An increase in output performance and customer orientation are examples of the approach (O’Flynn, 2007: 354).

A popular description of NPM is provided by David Osborne (1993: 356). The goal of Osborne’s theory is to make governments more entrepreneurial by pursuing ten principles, or several of these (Ibid: 351). The ten principles summarized below form the basis on which

(18)

18

social housing policy can be analyzed. In the results chapter, nine central principles are used as an analysis framework.

Others, for example Hood (1991: 4-5) have also provided descriptions of NPM as a set of doctrines. However, the definition by Osborne has been most influential in the scientific debate and more widely used as a basis for governance programs. Moreover, it provides a clear distinction between doctrines that makes his definition useful as a framework to analyze governance in practice.

2.1.1. The ten principles of NPM

(1) The first principle states that governments should be ‘catalytic’, meaning problems should be solved by steering third parties. In a bureaucratic system, problems were solved by setting up bureaucracies. This means governments are producing solutions themselves, which is expensive. The government needs to ‘steer rather than row’ to become more efficient and flexible (Osborne, 1993: 352). (2) By introducing competition, government services create incentives to achieve better results. Monopolies were the norm in old bureaucracies, this organization form lacked incentives to work hard. If the state services have to compete with other organizations, they have to provide better services in order to keep citizens satisfied (Ibid: 352). (3) Governments should focus on output goals, rather than budgets and rules. In traditional bureaucracies rules and budgets determine policy behavior. Since officials would only focus on spending their entire budget to avoid cuts, this system is not cost-effective (Ibid: 353). (4) The focus on goals can be pursued more easily when results are measurable. Officials should have a clear ideas of how to improve policy and what costs are linked to this improvement. Therefore, goals should be described in measurable terms, and indications of costs for improvement should be given (Ibid: 353-354). (5) The best way to tie results to spending is to let citizens (consumers) choose how to spend funds. By letting citizens choose, governmental organizations are forced to become more responsive and to improve their service quality (Ibid: 354). (6) Governments should consider earning money as a more serious option. By widening the scope from only budget spending, more efficient and cheaper solutions to public problems can be achieved (Ibid: 356)3. (7) Smaller-sized organizations are easier to manage. Therefore, a decentralization

3

Since this doctrine is not very different from doctrine four or ten, its content is discussed under these doctrines in this thesis. Only nine doctrines are discussed.

(19)

19

strategy should be pursued. (8) A community-owned government should be better equipped to contribute to democratization and increased social control. (9) Anticipatory governments focus on prevention rather than cure. (10) A market-driven government changes the scope from finding solutions in public programs to solutions in influencing supply and demand within markets (Ibid: 356).

NPM as it is described by Osborne emphasizes the need to change the attitude of public servants. This is especially a reaction to the shortcomings of the old bureaucracy, such as inflexibility and an unresponsive attitude of public servants. This inefficiency was mainly derived from a focus on process, rather than outcome. NPM reforms advocated a change of attitude, the services could remain under control of the same organizations (Alford and Hughes, 2008: 5).

In later NPM theory, there was more attention for cooperation between public and private organizations. This cooperation is argued to bring forth the best of both: the market conditions and effectiveness of private actors and the legitimacy and accountability of public actors (Ibid: 6-7). Since the 1990s, an increasing number of agencies and organizations has been given more autonomy (Kickert, 2001: 137). The focus has changed from internal reforms to reforming government services by outsourcing them (Alford and Hughes, 2008: 5-6).

Outsourcing public tasks to private firms or hybrid organizations (private organizations with a public task), can increase transparency and clarity of goals (Kickert, 2001: 147-148). Although the combination of public and private values, practices and goals is a challenging combination (Ibid: 147).

2.1.2. Criticism on NPM

The work of Christopher Hood (1991: 8) forms a good starting point for discussing different critiques on NPM. They can be divided into four groups. The first criticism states that NPM is nothing more than a new hype which does not change anything for real (Ibid: 8-9). Dunleavy and Hood describe this as a ‘fatalist critique’ on NPM. This critique states that NPM cannot change the basic dilemmas of public policy. Only the jargon and control frameworks change. The human errors, system failures, misdirected programs and corruption will stay (Dunleavy and Hood, 1994: 10-11).

(20)

20

The second critique argues that NPM does not cut costs but just enlarges the management by inducing measurements everywhere. The NPM reforms might reduce costs on one place, but these costs are negated by the NPM reforms themselves (Hood, 1991: 9). Moreover, NPM could absorb the attention of experienced policy-makers, who originally watched the consequences of new policy strategies. NPM pursues a more individual approach, with less control from other public actors. Controlling and monitoring also bring substantive costs with them, making the system less efficient (Dunleavy and Hood, 1994: 11).

A third criticism asserts that NPM creates a management elite that does not address public values aside from efficiency and profitability (Hood, 1991: 9). Market forces and a decentralized structure form opportunities for the abuse of public funds. Another risk is overly enthusiastic budget-minimizing by public servants. Policy makers might hurt the policy execution by their personal interest of cost-reduction, for which they are rewarded (Dunleavy and Hood, 1994: 12). The entrepreneurial attitude that is promoted could lead to a self-serving group of managers. Where economic values are based on gaining personal profits, there exists a risk that this goal is maintained in the public agencies (Kaboolian, 1998: 190). Moreover, the fragmentation of public service might pressure on accountability of public servants. Public servants will only focus on their specifically defined goals, impeding the possibility to deal with problems that cross agency boundaries (Dunleavy and Hood, 1994: 13).

The fourth criticism argues that NPM is not a universal model that can address all challenges in governance at the same time. NPM especially focuses on values such as efficiency (Hood, 1993: 13-14). By pursuing this strategy, it often assumes a culture of honesty among public servants as a given. Measures to ensure these values are upheld are often neglected or weakened by NPM reforms (Ibid: 16). The argument that NPM measures immeasurable values and treats citizens as consumers can also be linked to this fourth critique. The focus on one set of core values ignores the existence of other values that are important in the public sector (Ibid: 11). In multiple countries, including the Netherlands, studies have reported that public values such as accountability have come under pressure due to NPM reforms (Brereton and Temple, 1999: 460). However, the critique that NPM removes traditional public values completely is overdone. In most cases, a balanced mix between the old bureaucratic public values and the new private values is sought (Ibid: 464).

(21)

21

2.2. New Public Service

Where NPM is a reaction to the shortages of old bureaucratic policy making, NPS can be seen as a reaction on the weaknesses of NPM. In NPM, the focus is put on reducing costs and improving efficiency. NPS focuses on serving citizen’s needs, which often means focusing on different values (Warner and Hefetz, 2008: 163). NPS is a model which introduces several new policy techniques, but also a model that carries out a set of public values (Denhardt and Denhardt, 2000: 551). Denhardt and Denhardt have given a clear definition of NPS (2000: 554-556), which is provided below. This model is based on a different set of values and goals than NPM (Ibid: 551).

2.2.1. A definition of the New Public Service

NPS argues that instead of focusing on steering and managing society, public servants should shift their focus to serving citizens. Government should not be the distant leader, but a participant in a network that interacts with citizens and other stakeholders. By cooperating with citizens, solutions for public problems are found and turned into policy (Denhardt and Denhardt, 2000: 553). This is in contrast with NPM, which advocates a steering government at more distance from the policy process (Osborne, 1993: 352).

By deliberating, a shared notion of public interest is created among stakeholders. It is important that every stakeholder is heard. The fact that every actor can express his interests is more important than formulating a clear goal that leads to partial exclusion (Denhardt and Denhardt, 2000: 554). This prevalence of inclusion over clear goal setting by NPS contradicts the call for clear goal formulation of NPM (Osborne, 1993: 352). Inclusion and trust building are more fruitful strategies for long-term goals than a short-term focus on cost reduction and efficiency (Denhardt and Denhardt, 2000: 557). Public servants must always be flexible in changing governance according to the wishes and suggestions of actors (Ibid: 555). NPS broadens the scope of public values (Hood, 1991: 15-16). Citizens should not merely be seen as consumers, but also as individuals with more complex interests. The narrow focus of NPM on entrepreneurialism needs to be widened. Public servants should take community values, political and professional norms and standards, citizen interests and responsiveness into account (Denhardt and Denhardt, 2000: 555-556).

The lessons discussed above give insight in how NPS differs from NPM. However, very little is written on how NPS could be implemented in practice. There is little indication

(22)

22

to how public servants could best serve their citizens, only that a focus on customers is too narrow. Therefore it is worth experimenting with alternatives that do address problems in a broader way. It is not necessarily related to the values of NPS, but it is potentially a suitable organization form to market competition. Network governance focuses on cooperation of stakeholders instead of competition (Sørensen and Torfing, 2009: 242).

After network governance is discussed, there is attention for PVT. PVT can give direction to which values NPS strives for, or should strive for. It focuses on what can be defined as an important service for citizens. It also pays attention to which values are important in NPM and NPS, giving more insight in what focus each governance model has (O’Flynn, 2007:).

2.2.2. Network governance

In this research, network governance is defined as a governance strategy that promotes cooperation between the government and a variety of other actors, aimed at the cooperative provision of public goods and services (Rhodes, 2007: 1247). Actors negotiate and argue with each other in an institutionalized framework of rules, norms and shared knowledge (Sørensen and Torfing, 2009: 236). NPS does not provide a clear indication on how its ideas should be put to practice. Networks are mentioned as a useful means to enlarge inclusiveness. The focus of NPM on more effective decision making (catalytic government) has reduced influence of other stakeholders (Sørensen and Torfing, 2009: 254). Therefore, network steering should be useful as a carriage on which NPS can be realized in practice.

Networks are the key to more flexible and proactive governance. On the downside they are often unstable, diffuse and opaque (Ibid: 236). Policy makers can deal with this by focusing on management skills and strategically influencing networks (Skelcher, 2000: 16). The success of networks also depends on the policy area. If goals are vague, or knowledge of the field is lacking in a policy area, more input from stakeholders can be beneficial (Agranoff, 2006: 60).

There is a large pool of measures and strategies that policymakers can use to influence networks. Four groups of measures can be distinguished: designing, framing, managing and participating in networks. Designing and framing network governance means the government influences the policy making process passively on the background.

(23)

23

Managing and participating in a network requires a more active attitude from the government (Sørensen and Torfing, 2009: 246-247).

The government can influence the effectiveness and inclusiveness of a network with a number of means and strategies. Effectiveness means improving efficient and fast decision making. The amount of inclusiveness shows how many stakeholders are involved and how democratically a network operates (Ibid: 249). By supporting and lowering interaction costs, governments can make the network function more effectively. Empowering actors can speed up the decision-making process as well. Defining clear goals and deadlines enhances effectiveness as well. Framing network coordination is also a method to make decision making more efficient: framing a sense of urgency can for example help reaching an agreement faster.

By including underrepresented and weaker stakeholders, one can make the design more democratic. Creating public awareness also makes a network more approachable. Another way to enhance inclusiveness is endorsement by legitimating actors (for example elected politicians) it is framed as more democratic (Ibid: 249-250).

Enhancing effectiveness and inclusiveness can enhance the bonds of trust in the network. Moreover, participation shows which strategies work and are suitable to institutionalize (Ibid: 248). Insisting on dialogue and open-mindedness makes sure that a policymaker can enhance participation. Loosely defined goals and arrangements gives stakeholders space to exchange ideas and negotiate on sensitive issues (Agranoff, 2006: 60). Governance networks can be undemocratic (Sørensen and Torfing, 2009: 243). This depends on whether a network consists of a small exclusive group with high boundaries to join, or consists of a broad range of actors with low boundaries to participate (Fawcet and Daugbjerg, 2012: 199). NPS is a model that focuses on more responsiveness and inclusiveness of citizens (Denhardt and Denhardt, 2000: 554). One can thus expect their networks to become more society-centered and inclusive. A consequence of this approach is that networks receive a lot of input. This large amount of different interests makes it harder to find consensus and make decisions (Fawcet and Daugbjerg, 2012: 204).

Network governance can be seen as an alternative way of organizing the provision of a public good. The market forms the basis of public good provision for NPM. Not every sector functions well by the laws of the market though. Absence of market forces or a lack of incentive to compete can be important reasons why market functioning does not work.

(24)

24

Embedding (semi) public institutions in a network reinforces cooperation and interaction (Warner and Hefetz, 2008: 162-163).

2.2.3. Public Value theory

PVT is described as a theoretical foundation for network governance (Stoker, 2006: 44; O’Flynn, 2007: 359). The overarching goal of PVT is to create public value. Public value can be created when the government solves problems that citizens are concerned about. Government services are judged by their value for citizens from a PVT perspective (Stoker, 2006: 47).

PVT should be carried out through deliberation in networks in which stakeholders can make clear what public value means to them (O’Flynn, 2007: 60). If public value is the measure by which the quality of services is measured, then input from citizens must be enhanced (Stoker, 2006: 48).

Not only should public servants learn from- and cooperate with citizens and stakeholders for solving public problems, they should also work together. Large and complex problems often cross the boundaries of a single agency, therefore cooperation and interaction between agencies is necessary. This is especially a weaker point in NPM, since this system has led to a more fragmented service system (O’Flynn, 2007: 362).

Within networks, a pragmatic approach should be kept. Pragmatically choosing the strategy that works best makes policymaking most effective (Stoker, 2006: 44). NPM, and to a limited extent also network governance, are theoretical models that argue for a ‘one size fits all’ solution. Public value can be realized by pragmatically choosing from a list of options, including market functioning, network governance and bureaucracy. One can thus argue that it makes no difference which means are used, as long as the results are achieved (Alford and Hughes, 2008: 1). However we have to keep in mind that the legitimacy of the process must remain intact, or the public value will easily be lost (O’Flynn, 2007: 359-360).

Public servants have to be committed to a public service ethos that should be open to input from citizens to improve services (Stoker, 2006: 48-49). This ethos does not have to be organized according to bureaucracy. With training programs, staff support, leadership and organizational learning, a motivated public workforce can be created (Aldridge and Stoker, 2002: 26).

(25)

25

Many scholars provide lists and oversights of public values (Jørgenson, 1999; Jørgenson and Bozeman, 2002; Jørgenson and Bozeman, 2007; Van der Wal, 2008). However, these lists and their goals differ in their purposes and goals. Some work focuses on empirical aspects of PVT (what values are used?), while other articles discuss the normative dimension (what values should be used?) (Reynaers, 2013: 43).

In the NPS model as it is described by Denhardt and Denhardt, lawfulness, altruism and a professional attitude of public servants are almost literally mentioned as values to be strived for (Denhardt and Denhardt, 2000: 554). Other important values that keep returning in the model are responsiveness, accountability and transparency. The call for more attention for citizens interest and a dialogue in which these interests are the subject of the policy debate are the central calls of NPS (Denhardt and Denhardt, 2000; Warner and Hefetz, 2008).

Public servants should pay attention to more than just market values (Denhardt and Denhardt, 2000: 555). The essence of the value of responsiveness is that public servants see citizens as individuals, not as consumers. Citizens should be taken seriously in the public policymaking debate (Jørgenson and Bozeman, 2002).

Public servants should be accountable, not only to politicians but also more directly to citizens (Denhardt and Denhardt, 2000: 555-556). In the NPM reforms, accountability has come under pressure because services have been outsourced to private firms (Reynaers, 2014: 43). In network governance, there are more options to keep multiple service providers accountable through the interaction in the networks (Sørensen and Torfing, 2009: 244-245).

Public servants are not only responsive and more accountable to citizens, they also give citizens more insight in to their decision making procedures and considerations that influence these procedures (Denhardt and Denhardt, 2000: 556). By giving more insight in these procedures the value of transparency is realized (Jørgenson and Bozeman, 2002: 68). In privatization processes, some scholars argue that transparency can come under pressure (Reynaers, 2013: 44; Jørgenson and Bozeman, 2002: 68), this is important to keep in mind when analyzing the Dutch housing sector.

Lawfulness means that public servants abide laws and procedures (Jørgenson and Bozeman, 2007: 365). Public servants should pursue an altruist attitude which entails engagement in helping citizens, not just motivation by salary or job benefits (Ibid: 68). Lastly,

(26)

26

public servants should be able to act according to a professional code, independently of commercial interest. This is hard to realize when cost reduction and outsourcing to private firms is done (Ibid: 68).

2.3. NPS as an alternative

Cooperation between governments, municipalities, corporations, tenants and other stakeholders in networks could have a positive effect on governance. Since the sector has been fragmented, more emphasis on coordination without forcing certain actors into the submission of others is a possibility (Denhardt and Denhardt, 2000: 556). Moreover, NPM with its doctrines of output focus, customer orientation, goal formulation and market conditions do not always seem suitable as a theoretical foundation for the social housing sector. First, the elements that can already be recognized in the sector are mentioned. Then the potential added value on other aspects of governance is discussed.

2.3.1. A starting point: existing elements of NPS

A new policy can never be introduced out of thin air. Social housing policy has been reformed a lot in the past two decades (De Jong, 2013: 11). Since NPS is a theory that provides an alternative for NPM, it has several connections with NPM that ensure the possibility for a more gradual shift to the alternative steering model.

Before the housing act was introduced in 1901, corporations were associations of tenants. In 1901, the government started funding corporations, this gave them a dominant position. Since then, corporations were primarily accountable to the state, and secondly towards their members (Bieleman et. al, 2010: 20-21). Tenants were represented in corporations until the 1990s. After the reforms in the 1990s, citizens were treated as consumers rather than as members of a corporation. The idea was that citizens could exert influence by sharing preferences and by their possibility to go to a competitor. In practice this relationship is so unbalanced in the favor of the corporation that customers have little influence (De Jong, 2013: 42).

After corporations became financially independent and their set of tasks expanded in the 2000s, more opportunities for input from tenants became possible. Although the input did not reach the management of the corporation, tenants participated in urban renewal and other projects by corporations (Ibid: 41). In the reform of the housing act initiated by

(27)

27

the PEW, tenants received more serious influence in corporation policy. They can now block the nomination of a new CEO and appoint one-third of the supervisory committee (Aedes and VTW, 2015a: 4).

In 2009, the Centraal Fonds Volkshuisvesting4 (CFV) started monitoring the integrity of corporations. This monitoring process paid attention to a more diverse set of values than just NPM-values. Values like integrity, transparency and accountability received increasing attention (CFV, 2011: 4). Corporations policies on these values has since improved (Ibid: 5). The new Authoriteit Woningcorporaties5 (AW) has put more priority on supervising integrity and lawfulness (Blok, 2015: 3).

Governance codes are an important element of the increased attention for public values and a more open attitude towards citizens. The Aedes-code6, which is embraced by most corporations, states in its fourth principle that the management and Supervisory Board should realize societal goals through dialogue with tenants or their representatives. The corporation should also be open to input from other stakeholders, such as municipalities (Aedes and Attrivé, 2015: 5). The management should pursue a policy to create the institutional foundation for active tenant committees that can regularly interact with the corporation (Ibid: 25-26).

The first wave of NPM-reforms brought forth a certain level of network steering. Sector-association Aedes wanted to create a system of corporations embedded in a network of stakeholders, including the state, municipality and tenants. Corporation management had to be organized in a more horizontal way. This plan did not succeed, since corporation management was too vertically organized. Moreover, corporations turned out to be unresponsive towards other ‘stakeholders’, which made network cooperation harder. With the gradual centralization in the second wave of NPM reforms, the network steering ideal faded (De Jong, 2013: 43-44).

Corporations are held accountable on an increasing number of public values. Besides financial stability, corporations are since 2009 increasingly monitored by the CFV on

4

The Centraal Fonds Volkshuisvesting (CFV) is the institution that was responsible for the financial supervision of corporations until 2015, when it merged into the larger Autoriteit Woningcorporaties (AW) (website AW, 2016).

5

Housing Corporations Authority

6

Aedes is the sector association of Dutch social housing associations, most corporations are member and thus follow the governance code (Aedes, 20-04-2016).

(28)

28

integrity and lawfulness (CFV, 2011: 8). These values have received even more priority in the new AW (Blok, 2015: 3).

There are several elements in the sector that can be a starting point for NPS to build upon. Tenant participation in the governance of corporations is taking place in a limited form. Tenants can exert influence through tenant committees (Volkskrant, 11-12-2014). Corporations have slightly broadened the set of values they take into account when managing the organization. Supervision institutions and governance codes stress that corporations work on values such as integrity, transparency, lawfulness and accountability (CFV, 2011: 4; Blok, 2015: 3).

2.3.2. Elements of NPS that could improve social housing management

One of the problems NPM faces in the sector is caused by the market balancing strategy. The government is forced to intervene in the market of social housing because otherwise it becomes imbalanced. This is due to the fact that corporations do not respond to market incentives (Bieleman et. al., 2010: 104; De Jong, 2013: 28). If competition and market functioning would be replaced by interaction with the target population, there would be a better insight in the demands. Market incentives do not have to play a dominant role in this system anymore (Denhardt and Denhardt, 2000: 555-556).

On the contrary, a certain amount of strategic cooperation and interaction should be beneficial to oversee the sector. Another benefit that interaction brings is that clear goals do not have to be formulated. By deliberating on problems and possible solutions on a regular basis, goals can more easily be set and adjusted to remain adequate. Corporations should be embedded in a network of stakeholders to whom they are accountable. In this network, interaction is important and different actors should have equal voices to express their interests7 (Sørensen and Torfing, 2012: 249).

A way to improve interaction with the target population is to start treating them as tenants, not as customers of goods. When tenants are treated as customers, many other values they may cherish are not addressed. This means that corporation management should become less authoritative and more open to input and deliberation with tenants (Denhardt and Denhardt, 2000: 555).

7

To create a network of ‘equal’ stakeholders, one overarching power is required to force all actors to cooperate constructively. This is often the central state (Sorensen and Torfing, 2012: 255).

(29)

29

One critique by the PEW on corporations is the lack of clear goals within their organizations (2014: 14). Although a certain direction would be beneficial for the management strategy of a corporation, a narrowly defined goal is not necessary. By leaving policy goals open for input, the corporation remains flexible for wishes from stakeholders and changing circumstances (Sørensen and Torfing, 2012: 249). A flexible goal formulation can help realizing the NPM doctrine of anticipatory policy. Maintaining flexible and open for changing environments is a better attitude for anticipating problems in the future than being bound to specific and detailed goals (Denhardt and Denhardt, 2000: 555).

There is room for more emphasis on the NPS model to improve governance of the sector. Artificially created competition between corporations could be replaced by a form of network steering, in which all stakeholders receive a more equal voice. Detailed goal formulation should be abandoned, and replaced by more flexible goals with space for adjustments (Denhardt and Denhardt, 2000: 555; Sørensen and Torfing, 2012: 249).

2.4. Conclusions

The current governance in the social housing sector is largely based on the governance model of NPM. This model stresses that output performance and a private company management style are important. The governance strategy can be caught in a framework of nine doctrines. NPM is criticized for its emphasis on measuring output, which costs a lot of energy that could be invested in the policy itself and neglects the process towards the output. A second critique is that NPM does not take public values into account sufficiently. In NPM only profitability and efficiency matter. A fourth critique is that NPM can only partially deal with policy problems. A focus on output requires honest and motivated personnel, which is taken for granted by NPM, but is not self-evident.

These weaknesses of NPM will likely have repercussions for the quality of social housing governance. NPS tries to provide an answer to the shortcomings of NPM by widening the scope from a narrow set of output goals to a broader question of how service can best be provided to citizens (Denhardt and Denhardt, 2000: 556-557). It is expected that interaction and exchange of information can enrich and improve decision-making in the sector. These ideas can be realized through network governance in which all actors involved in the housing sector can interact and exchange information (Rhodes, 2007: 1.247). PVT can contribute by defining which public values are important to pursue (O’Flynn, 2007: 359).

(30)

30

Table 2.1. shows how the doctrines of NPM can be compared to the ideas of NPS. It forms the framework for the results in chapter five. In the first part of the results chapter, the table is filled in with descriptions of how the doctrines have been carried out in the sector. In the second part, it is used to identify the way in which De Key carries out the doctrines and whether these decisions are based on NPM or NPS thoughts.

New Public Management New Public Service

Catalytic government: steer rather than row Serving government: serve rather than steer Introduce competition to improve

performance of public service

Replace competition by network steering to exchange information and opinions

Set clear goals to give direction in

governance which stimulates performance

Keep goals broad and adjustable to new input by stakeholders

Focus on output performance, increase efficiency and measure progress

Focus on inclusive process, include every stakeholder equally

Be service oriented, treat citizens as customers to improve governance quality

Treat tenants as citizens, not as customers with a narrow scope of values

Divide large government services into independent and flexible smaller agencies

Avoid fragmentation by bringing stakeholders together in networks Institutions with private style and behavior

perform better than public equivalent

Privatized services must be accountable to stakeholders and public values

Focus on prevention rather than cure, deal with potential problems in advance

Anticipatory policy is improved by exchanging information with each other Influencing markets is a good alternative for

solving problems instead of public programs

Pragmatically choosing between different options is the best way to solve problems

Table 2.1. An overview of the nine core doctrines of NPM and the alternatives proposed by NPS. Also the framework for the analysis in chapter five.

(31)

31

Chapter 3: Methodology

This chapter describes the methods used to answer the main question: what effect does New Public Management have on the governance of the Dutch housing sector and to what extent can New Public Service be an alternative to improve governance in the sector? The first part of the analyses, which assesses the implementation of NPM in the social housing sector is discussed. This is done with a literature study, based on reports from different governmental agencies and supervisory institutions, as well as scholarly articles about the developments in the sector. The second half of the main question is analyzed with a case study of housing corporation De Key. The data for the case study is gathered with semi-structured interviews. The interviews are analyzed with a qualitative content analysis, using the Constant Comparison Method (CCM).

In the first paragraph of this chapter, a more general description of the document study and case study are provided. Then the data collection of both methods is discussed. Subsequently the analysis of the documents and coding of the interviews of the case study is explained. The consequences for the reliability and (internal and external) validity of the methods are explained8 lastly.

3.1. Document study and case study

The document study is a method used to see how the different doctrines of NPM are carried out in the housing sector. An oversight of the doctrines can be found within the framework provided in table 2.1. This table forms the framework for the document study. The goal is to gain insight in each of the doctrines separately. This provides knowledge on which doctrines work well in the housing sector and why.

A document study is a form of qualitative content analysis. In a qualitative content analysis, documents are gathered and analyzed for their underlying meanings (Bryman, 2004: 392). Because of the incidents in the sector, there is a large amount of information available which describes the developments of governance in the sector. Since this data is readily available, analyzing if NPM doctrines are mentioned is a good way to gain understanding of the impact of NPM on social housing policy.

8_{Reliability and validity are important indicators of the quality of a research. Reliability is high when results are}

the same after repeating the steps of the research. The internal validity is good when the research steps have been carried out carefully. External validity is determined by whether causal relations to other cases can be made (Golafshani, 2003: 598-599).

(32)

32

The second part of the results chapter examines the case of housing corporation De Key. Since there is little knowledge on new forms of policymaking in the sector, a case study is the most suitable strategy to gain new insights. A detailed study of one corporation can lead to deeper and more clarifying results than a shallow survey of multiple organizations (Gerring, 2007: 20).

Data on De Key is gathered by conducting semi-structured interviews with respondents from in and outside of De Key. From De Key, two respondents from the middle management, two from the management and one from the Supervisory Board have been interviewed. Outside of De Key, two respondents from the tenants committee of De Key9, two respondents from sector association Aedes and one respondent from the Amsterdam Federation of Housing Corporations (AFWC) were interviewed. The transcripts of these interviews are analyzed with a directed content analysis. A directed content analysis makes use of an analytical framework based on existing theories (Hsieh and Shannon, 2005: 1.281). This means the analysis is deductive, it aims to answer whether NPS is useful in policymaking or not.

3.2. Data collection

3.2.1. Document study

The data of the document study consists of scholarly work on the sector, reports by inquiry commissions, reports by supervisory institutions and reports by the sector association10. Data was gathered by applying the snowballing method, which is explained below. Based on former literature, insight was gained in which other literature could be relevant, thereby expanding the amount of literature (Atkinson and Flint, 2001: 1). Documents with many references or reports from an organization that is often mentioned were selected. The PEW report, for example, is mentioned in almost every evaluation report and research article published after 2014 (Aedes and Attrivé, 2015; Aedes and VTW, 2015b; AW, 2015). Therefore this report was adopted in the document study.

9_{Although tenant organization Arcade is connected to De Key it can be positioned as an independent advisory}

organization (website Arcade, 09-06-2016).

10

Examples of sources are: the Ministry of Infrastructure and Environment (former Ministry of Housing, Spatial Planning and Environment), Social and Economic Council (SER), Sector association Aedes and the Scientific Council for Government Policy (abbreviated as WRR in Dutch).

(33)

33 3.2.2. Case study

Data on the case was gathered from two types of sources. As a start, general policy information about De Key was gathered from annual reports and strategy documents. Most of the data was gathered through semi-structured in-depth interviews. In the interviews, the opinion on different doctrines of NPM and NPS was asked. Which doctrines did they find problematic and how could they be changed? Most questions were therefore about the doctrines, the majority of respondents discussed multiple doctrines in one answer. An oversight of the interview questions can be found in appendix 1.

Respondents were selected by conducting a method of snowball sampling. The first respondent was approached through personal relations. At the end of the interview, the respondent advised which respondents could be interviewed next. Some respondents from sector organizations were approached when their organization was mentioned a lot in policy documents.

The snowballing strategy has influence on the representativeness of the research population. This problem was attenuated by intentionally asking to recommend respondents from different positions in the corporation. This ensures that a diverse set of perspectives is included and a more representative sample is gathered. Moreover, the snowballing technique was used because the population is hard to reach without personal relations in the sector. This makes snowball sampling the most viable option (Atkinson and Flint, 2001: 3). It is hard to know who has the most knowledge on this subject. Recommendations by respondents might thus be more reliable than my own. A list of all respondents can be found in appendix 3.

3.3. Analysis

3.3.1. Document study

The analysis of the data is structured around the framework of the doctrines of NPM and NPS, depicted in table 2.1. The analysis is focused on which doctrines are mentioned in the social housing policy. NPS-doctrines are barely represented in the results, since this model is hardly used in current governance. Most of the information was found by closely reading reports and selecting descriptions that match doctrines. This was done with the help of a framework of NPM, which is depicted in table 2.1. When a divergence is found between the

(34)

34

literature and the doctrines described by Osborne, this is discussed and explained in the results chapter.

3.3.2. Case study

The interviews were analyzed with the CCM, which is a form of directed content analysis. Based on the framework of doctrines in table 2.1, interview transcripts were analyzed and marked (Hsieh and Shannon, 2005: 1.281). The difference between the analysis and the document study is the emphasis that is put on elements of NPS. During the interviews, there was opportunity to discuss NPS. This is not possible when analyzing policy documents.

CCM applies to the analysis of the interviews. The analysis is based on the step by step approach as described by Boeije (2002: 395). First, a single interview is encoded and then the core concepts are analyzed. The concepts were sorted into two groups: NPM doctrines and NPS doctrines. These were then divided based on their doctrine division. Subsequently, the respondents were divided into three value groups: corporation managers, corporation employees and other stakeholders. The results of respondents within the same group were compared. After that, the data samples were compared on a group level. Finally, the samples were compared on their value groups (Ibid: 396). A list of codes supporting each finding is provided in appendix 2.

3.4. Reliability and validity

The data collection and analysis methods have impact on the reliability and the internal and external validity11. The selection of documents was based on the potential of documents to enrich the body of information in the analysis. A randomized or quantified selection method would be counterproductive in this situation (Polkinhorne, 2005: 139). Determining whether a document is useful was done with the snowballing strategy. When a document, institution or author was frequently mentioned in earlier documents, they were viewed. When social housing policy was extensively described, the documents were selected as relevant data.

The advantage of snowball sampling is its ability to reach populations or documents that are relatively hard to approach or hard to find without prior knowledge (Polkinhorne,

11_{Reliability indicates whether a research can be repeated or not. Validity in social research determines}

whether a research is trustworthy. Internal validity is strong when research steps have been carried out carefully and in a logical way. External validity determines whether results can be applied in other cases (Golafshani, 2003: 598-599).

(35)

35

2005: 141). Another advantage is the possibility to search for documents or ask for respondents that seem underrepresented in the study, which makes the internal validity of the case study stronger. Respondents can also help the researcher with advise on participant selection when he is not entirely familiar with the population (Atkinson and Flint, 2001: 3). Intentionally pursuing a diverse set of respondents is also beneficial for internal validity, since it reduces the dependence on (subjective) choices by earlier respondents.

Reliability and external validity are weaker when using snowball sampling. The representativeness of a sample is uncertain, since respondents are not selected randomly. Another sample of respondents could have led to different insights, indicating that the findings of this case study might not provide a convincing explanation for other cases. This limits external validity. A replication of the study would improve the reliability of the selection procedure. The same applies for external validity. Since representativeness of the sample is uncertain, only replications of the study can enlarge its generalizability, or show limitations of the study (Atkinson and Flint, 2001: 3-4). Interviews about sensitive subjects can be arranged easier with a snowballing strategy. In the this research, bad governance of the corporation and in the sector that have taken place were sensitive issues (Atkinson and Flint, 2001: 2).

The reliability of the CCM as a form of directed content analysis of the case study is strong. Due to its deductive approach, the essence of the research is to either prove the usefulness of NPM and NPS or to show that the theories contradict (Hsieh and Shannon, 2005: 1.282). This process can easily be repeated. A clear analysis framework is provided in table 2.1, which can easily be used again.

The internal validity of the results is pressured by two challenges. Firstly, the adopted framework gives the researcher a biased view of reality. Although aware, the approach of data will be specifically oriented at finding specific ideas, which narrows the researchers view and makes him blind to eventual contextual events (Hsieh and Shannon, 2005: 1.282-1.283). Secondly, this only applies to the interviews, respondents are more likely to go along with the theories that are represented in the framework. They want to please the researcher. Questions are based on the theories and steer respondents in these directions. This problem is partially negated by giving respondents the opportunity to express what they think are important matters (Hsieh and Shannon, 2005: 1.282).

(36)

36

3.5. Conclusions

This chapter has discussed the methods that are used in the results chapter, which consists of two parts. The first part consists of a document study of which NPM and NPS doctrines are present in the entire housing sector of the Netherlands. The second part is a case study that investigates what potential NPS has in housing corporation De Key. This case study is carried out by conducting semi-structured interviews.

One of the biggest advantages of the methodology of the results chapter is that it makes use of a clear framework that can easily be used again. Moreover, it is providing consistency to analysis of content that can be of differing natures.

A more problematic side of the methods is that case selection was dependent on personal relations and recommendations of others. Because of this, control over the selection procedure is partially lost, which makes it uncertain whether a similar research population would be reached by another scholar.

(37)

37

Chapter 4: A brief history of social housing in the Netherlands

The history of the housing sector in the Netherlands is complicated, it is full of reform processes and changes of direction. Many institutions and stakeholders have influenced in the policy process. The position of many of these actors has changed through the years, making it sometimes confusing to know who is responsible for what. This chapter aims to describe how the sector is organized and provide an explanation for why it is organized in this way.

The recent history of the sector can be divided into three parts. The first part is marked by the bureaucratic governance system of the state. Corporations were dependent on the state and subject to a complicated and inefficient governance system. The second part is mostly about the reforms that are made in the 1990s and early 2000s. In the 1990s, deregulation and stimulation of competition were central in the policy to remove ties between the corporation sector and the state. The more assertive policy by the state is mostly a reaction to the problems caused by the reforms, starting around 2009. Some corporations performed poorly or caused large and costly incidents. Supervision of the sector was insufficient. To ensure the survival of the sector, the government intervened more actively and took a more assertive position. The three parts of history will be discussed below, followed by a discussion on the two forms of NPM in the provisional conclusions.

4.1. Housing policy until the 1990s

Before the twentieth century, social housing was provided by private organizations, for example clerical and philanthropic organizations (Van der Schaar, 2009: 5). As costs for (construction of) housing for the poor rose, the call for state assistance became louder. This finally resulted in the Woningwet (Housing act) in 1901 (Ibid: 6-7).

The public interest was put central in the new policy. The state provided loans to local governments or private organizations to enlarge the supply of good quality housing (Ibid: 11). This policy gradually increased the production of state-subsidized housing. In 1916, the government intervened on a larger scale. Due to stagnating housing production by private organizations, large interventions were made to deal with high housing shortages (Ibid: 20). This process ended in 1923, when the government abruptly privatized the

New public service in the Dutch social housing sector : an exploration of the value of new public service in the governance of social housing in the Netherlands