Adaptive building and its institutions

Master thesis T.D. Ravestijn

all know that many office buildings have become redundant during the last decade. Nonetheless new buildings are constructed on the triple A locations of today. During my travels to the university of Amsterdam, the many vacant buildings at the Amsterdam Sloterdijk area struck my eye. A few kilometres further, in the south-eastern part of the city, yet another office area has to deal with an increasing number of vacant buildings. While these two areas deal with a decreasing demand for office space, at the Amsterdam Zuidas (literally: south-axis) new offices are being constructed. As emphasized during the urban planning programme at the university of Amsterdam, the Zuidas is the new business area of Amsterdam; Easily accessible via the main highways around the city and perfectly suitable for international business due to the proximity of the airport. I do not tend to bring in doubt the importance of a business location, but the older office areas may have been constructed with a comparable conviction of choosing the right location, as in the case of the Zuidas. As I began reading about office conversion it became clear that many buildings are not suitable for reuse; in such cases leaving the building vacant or demolition are two options for the proprietor. Both vacancy and demolition are not desired for the direct surrounding, it is often heard that buildings are part of a larger real estate portfolio and function merely as an investment object, which makes vacancy less troubling. Obviously, this is not always the case and these rumours alone, are worth a research. This interest in commercial real estate made me wonder what the real mechanisms are behind the construction, conversion and reuse of buildings. Vacancy and demolition do not happen solely in Amsterdam. All over the world new high-rise buildings are constructed for a particular use. Many cities have large areas with office buildings as landmarks. Think about the skyline of New York, Hong Kong, or Tokyo and imagine the impact on the city if those buildings become redundant because of unforeseen events. Potential for flexible use can increase the resilience of cities. This research combines the theory of increased flexibility with the experiences in Amsterdam. Who are at stake and what key drivers are there for

have become redundant. Many of those vacant buildings were transformed into other forms of usage. However, some are not suitable for reuse initiatives because of, amongst others, technical and locational reasons. Hence, a significant number of the structurally vacant office buildings have been demolished since they were inflexible and not able to fulfil a different form of exploitation. One way to overcome the need for demolition is adaptive building. This entails that the parties, involved during the development of a new project, take measures to increase the potential for adaptation to other forms of use. Langston et al (2008) call this the adaptive reuse potential (ARP). This research focuses on adaptive building in the Netherlands. To what extent is it standard practice and how do institutions influence the measures to increase ARP during the early planning phase of development projects? Interviews with actors involved in three different cases in Amsterdam and policy analysis provide insight in the main incentives for and hurdles to adaptive building. The study shows that the cases developed from different institutional settings and all have their own ARP.

Inhoudsopgave

Preface v Abstract vii Inhoudsopgave 9 Figures and Tables 11 1 Introduction 13 1.1 Research introduction 13 1.2 Problem statement: adaptive reuse potential and the institutional setting 14 1.3 Research question 15 1.4 Relevance 16 1.5 Structure 17 2 Context of Office Transformation and Adaptive Building 19 2.1 Motives for building transformation 19 2.2 Opponents of transformation 20 2.3 Transformation barriers 21 2.4 Adaptive building 22 3 Theoretical framework 27 3.1 Concepts and policy 27 3.2 Institutional Setting 29 3.3 Rational choice institutionalism 30 3.4 The Game-Theory 31 3.4.1 Actors 31 3.4.2 Actor constellations 32 3.4.3 Modes of interaction 34 3.5 Law and Regulation 35 3.6 Hypothesis 36 4 Methodology 39 4.1 Research Design 39 4.2 Case selection 39 4.3 Assessing Adaptive Building 40 4.4 Assessing the institutional setting 44 4.5 Linking the institutional setting and adaptive building 46 4.6 Data collection 47 4.7 Data Analysis 50 5 The Dutch context 53 5.1 Policies on transformation of vacant buildings 53 5.1.1 National government 54 5.1.2 Province of North-Holland 55 5.1.3 Municipality of Amsterdam 56 5.1.4 NEPROM 57 5.2 Policy summary 59 5.3 Law and Regulation 62 5.3.1 The zoning plan 62 5.3.2 The building Act (Bouwbesluit) 64

5.4 interview results 65 5.4.1 Legislation 65 5.4.2 Functioning of real estate markets 66 5.4.3 Public sector practice 67 6 Adaptive building projects: Three cases 69 6.1 Steigereiland 69 6.2 Solids: The concept 70 6.3 Solid 1,2 73 6.4 Solid11 73 6.5 Transformation meter 74 7 Institutional setting of the cases 77 7.1 Actor analysis 77 7.1.1 Actors in project development 77 7.2 Actors involved in the cases 79 7.2.1 Case: Apartment building Steigereiland 80 7.2.2 The Solids: general institutional setting 82 7.2.3 Solid 1,2 83 7.2.4 Solid 11 85 8 Analysing results 89 8.1 Institutional setting 89 8.2 The Adaptive reuse potential 91 8.2.1 Private actor, public goals 91 8.2.2 Individual approach vs. solidarity approach 92 8.2.3 Path dependency 92 9 Conclusion & Discussion 93 9.1 Empirical findings 93 9.2 theoretical implications 94 9.3 Future research 95 References 99 Appendix 103 Topic guide interviews 103

Figures and Tables Figure 1: Conceptual Model 14 Figure 2: Conventional development scheme 24 Figure 3: Adaptive building scheme 25 Figure 4: Institutional setting by Scharpf 31 Figure 5: Concepts to be operationalized 41 Figure 6: Transformation barriers emphasized in this research 46 Figure 7: relations between organizations of which polices are analysed 53 Figure 8: Dutch policy regarding reduction of structural vacancy 61 Table 1: Sub-questions to support research question 16 Table 2: Reasons pro and against building reuse 21 Table 3: Transformation barriers 22 Table 4: Types of ideas 28 Table 5: Institutional setting according to Scharpf (1997) and Ostrom (1986) 36 Table 5: Shortened version of the transformation meter 43 Table 6: Operationalization of institutional setting 45 Table 7: Institutions and ARP per case 47 Table 8: List of interviewees 49 Table 9: Analyzed policy documents 50 Table 10: Coding scheme 51 Table 11: Summarized policies 60 Table 12: Three cases and their transformation meter scores 75 Table 13: Institutional setting of the Steigereiland case 82 Table 14: Institutional setting of solid 1,2 85 Table 15: institutional setting of Solid 11 87 Table 16: Cases, ARP and institutional setting 89 Photos: All photos are taken by the researcher, unless stated otherwise.

1 Introduction

1.1 Research introduction Many cities across the world deal with structurally vacant office buildings. Internet and communication developments changed society and the way business is attended. Furthermore the global economic crisis of 2008 lead to budgetary cuts in businesses and many employees and their offices became superfluous. As a result of these trends many office buildings became structurally vacant. According to a research conducted by OfficeRank (2014) more office space will become available for conversion in the coming years. If the office supply stays the same, in 2025 40 per cent of all office space will be redundant. These empty buildings offer potential solutions for two problems cities are facing: (1) For sustainability reasons, cities need to grow within their boundaries and (2) they are simultaneously confronted with a housing deficit. Fortunately, in many cases, structurally vacant office buildings are being converted into apartments. However, while more and more buildings are being reused, another considerable part needs to be demolished (Financieel Dagblad, february 2016,). Barriers to conversion make them not suitable for other means of use. In the 1960s, 1970s and 1980s, when most of the now vacant office buildings were built (Gann and Barlow, 1996), the need for reuse was hardly considered. The impact of societal changes and a financial crisis on the building stock were not taken into account. According to scholars, governments and developers could consider increasing the potential for reuse when a building becomes redundant in its original function (Wilkinson et al, 2016; Langston et al, 2008); Because we will never know which societal trends will appear and what effect they will have on the demand for specific buildings (Langston et al, 2008). First of all, vacancy has negative effects for the immediate surroundings. Second, the required demolition of all those ‘inflexible’ office buildings is not sustainable at all. Langston et al (2008) encourage increasing the adaptive reuse potential (ARP) of future buildings. This potential is calculated on the basis of the barriers encountered, while trying to convert a building. Heath (2001) and Remøy & Van der Voordt (2007) have defined, amongst others, the technical and locational barriers as the biggest

hurdles to overcome. Taking these barriers into account during the construction makes buildings more adaptive and therefore more feasible to different forms of use, which can prevent structural vacancy in the future. This in turn reduces the negative externalities of vacancy and makes cities more resilient for changes in society. 1.2 Problem statement: adaptive reuse potential and the institutional setting In many cities across the globe, new office buildings are constructed while others; a few kilometers away remain vacant. The location clearly is an important aspect for companies when deciding where to build their offices. Yet, it is the municipality who decides, which is reflected in their zoning plan, where different spatial functions can be located. As the downsides of redundant and inflexible buildings become visible, different government layers are actively promoting the transformation of vacant buildings. Private initiators are encouraged to give buildings a second life. Encouraging transformation processes alone is not enough though: It focuses merely on reducing the amount of structural vacant buildings after they have been abandoned in the first place. It is only one way to solve the vacancy issue. Adaptive building can be a means to prevent the structural vacancy in the first place because it increases the building’s potential to get reused. However, embedding the idea of adaptive building (Langston et al, 2008) in the building and spatial development industry appears too cumbersome as only a few projects show its features. The main assumption in this research is that it is foremost the institutional setting that influence the institutionalizing of new concepts like adaptive building. The following conceptual model (figure 1) forms Adap%ve Reuse Poten%al (ARP) Transforma%on barriers Actors Rules/ Laws Ins%tu%onal Se?ng

the basis of this research. Actors, involved in the development of new projects, can influence the institutions/ the “ways of doing” in this sector, as they are part of this institutional setting themselves. Therefore, the government and developers and other actors can influence the transformation barriers and thereby the adaptive reuse potential. Simultaneously, prevailing rules are important and can lead to certain policy outcomes. The actors and the rules together form the institutional setting. Hence, the ARP is the dependent variable and the involved actors and rules, as they influence the ARP, the independent variables. There are external forces, which influence the institutional setting in the field of spatial development too: Different policy sectors are intertwined and changing rules and norms in one sector affects others. For practical reasons though, this research focuses on the spatial development and these external influences are left out of the scope of this research. The timeframe of this research limits the possibilities to bring in multiple policy sectors. Moreover, it improves the in-depth understanding of the institutional setting in the spatial development sector. The initial focus lies on all barriers identified in by Heath (2001) and Remøy & Van der Voordt (2007) (see next chapter) and what the actors can do to lift these barriers. An analysis of the institutions (Scharpf, 1997; Ostrom, 1986) can help understand why certain institutions prevail and why ARP is more successfully applied in some cases. 1.3 Research question The notion that adaptive building is hardly implemented in practice and the presumed influence of the institutions on embedding it in practice, led to the following research question: “To what extent are measures to increase the Adaptive Reuse Potential applied during the development of new buildings and how do institutions affect this?” This is a twofold question: The first part of this question focuses on the current development practice. The second part of the research question focuses on the influence of the institutional setting on embedding adaptive building. To answer the second part of the question, three cases, which are adaptable to functional

change, are examined. Especially the second “How” part of the research question is suitable to get tested in real-life situations (Blaikie, 2009). A few sub-questions can aid to get a theoretical understanding of the concept and provide insight of adaptive building practice. Table 1 shows the different questions. SUB-QUESTIONS 1) What is adaptive building? 2) What are the main reasons to implement adaptive building; and what are the downsides? 3) What are institutions? 4) How is adaptive building embedded in spatial development practice in the Netherlands? 5) What is the adaptive reuse potential of the different cases? 6) Which institutional setting was prevalent in the different cases? Table 1: Sub-questions to support research question Sub-questions 1, 2 and 3 are sub questions, which can be answered by thorough analysis of the existing body of literature about adaptive building and institutions. These questions serve the purpose of getting adequate knowledge about the main concepts of this research. Question 4 provides insight in the context of adaptive building in the Netherlands. Polices of different government layers and developers are described so their standpoint on adaptive building becomes clear. Then, to answer sub-questions 5 and 6 empirical data is needed. The different cases are analysed and the concepts of this research are measured. Afterwards, these different results are compared and certain mechanisms can be discovered: Are there conditions in the institutional setting that lead to a higher ARP? 1.4 Relevance This research is socially relevant because it helps to get a better insight in the aspects of office vacancy. Society benefits for a number of reasons: First of all, sustainability is an important issue in almost every government sector (Municipality of Amsterdam, 2016; AgentschapNL, 2013). An increase of

sustainability in the spatial development sector is beneficial for all. Conversion of office space is a sustainable solution to deal with the housing shortage, as agricultural land does not need to be used for housing, while demolition of vacant buildings can be prevented (Heath, 2001). Secondly, vacant offices at centre locations are easily accessible and do not necessarily foster mobility by car (Remøy and Van der Voordt, 2007; Bertolini, 2006). Thirdly, apart from the sustainability theme, negative externalities from vacancy can be reduced, which enhances the liveability of an area (Koppels, Remøy and Messlaki, 2011). Although there is a body of literature about adaptive building, it only emphasizes the adaptability of the buildings to changing environmental conditions. For sustainability reasons this is an interesting perspective. The current body of literature does not cover the topic of adaptive construction of buildings; whether it is already happening; or how it could be stimulated remains unanswered. From a scientific perspective this research ameliorates the knowledge about adaptive building. The barriers and incentives for office conversion are known (Heath, 2001), but little has been written about the actual barriers encountered by private and public parties when and if they are willing to build in a flexible way. Furthermore it is interesting to see what role the public and private sector play in making sure the cities stay liveable on the long term. 1.5 Structure In this thesis both theoretical and practical aspects of adaptive building and the influence of institutions are elaborated. Chapter 2 starts with a thorough explanation of the theories of office transformation and adaptive building; what do we know and what are the standpoints of the scholars? Chapter 3 describes theories about institutional analysis. What are institutions and how do they influence ideas from scholars towards a policy outcome? Chapter 4 gives an overview of the research methods of this research. How are the different concepts of this thesis measured and how is data obtained? The next chapters describe the obtained empirical data of this research Chapter 5 starts with the Dutch context of adaptive building. In this chapter different policies are analysed and the first part of the research question is answered. Chapter 6 and 7

show the results from the research: Which ARPs and institutional settings are measured in three cases in Amsterdam. Chapter 8 analyses these results and comparisons between the cases are made; which different institutional settings are measured and how did this influence ARP (the second part of the research question. Finally, Chapter 9 presents the conclusions of this research followed by a discussion about this, and future, research.

2 Context of Office Transformation and Adaptive Building

This chapter describes the body of literature about the transformation of vacant buildings: what arguments are there to reuse buildings or for what reasons should this be avoided? Then, the idea of adaptive building comes to the fore. What is the difference with conventional transformation processes and how can it become common practice? 2.1 Motives for building transformation The conversion of empty office buildings to housing units started in the late 1980s (Barlow and Gann, 1993). Since then, due to the changing society, more and more office space has become vacant and over time some of properties have been converted into other functions (e.g. housing or neighbourhood facilities). But why is vacancy a bad thing? Wilkinson and Remøy (2015) identify a couple of reasons why private and public parties foster the reuse of vacant buildings. For sustainability reasons the reuse of empty buildings can prevent new land to be taken in use. Secondly, the reuse of empty buildings decreases the environmental impact. Demolition and the use of new natural resources clearly have a bigger impact on the environment than the usage of an existing construction. Besides, demolition increases the risk of uncovering asbestos or contaminated soil, which in turn are expensive to “clean” and extend the duration of a project dramatically. For urban policy reasons the reuse of empty buildings can be solution to get a better mix of spatial functions in an area. Many cities traditionally have central business area, which is featured by its many offices. Experience from the past have exposed the downsides of mono-functional areas (Jane Jacobs, 1992 ) and are rarely seen in contemporary planning practices. These often centrally situated central business districts offer new inner city housing opportunities. Empty offices in downtown New York were converted into small apartments for firsts-time renters (Beauregard, 2005); an area, which under ‘normal’ market conditions, often is unaffordable for first time renters; so a social mix of residents can be fostered. Then there are economic reasons; the tight housing market leads to higher rental prices for houses. Hence it is financially interesting to create houses in the, little demanded, office buildings. Moreover, vacancy has a negative impact

on the immediate surrounding; within a 500-meter radius, vacancy affects lease prices of other properties (Koppels et al, 2011). Finally, some buildings have a social value for a neighbourhood and demolition plans can lead to resistance of the public. In general the above described reasons can be summarized as political, economic, social and environmental drivers for office conversion The private and public sector both use different reasons why they would like empty offices to be used again (Remøy and Van der Voordt, 2007, Heath, 2001). For the real estate owners, for example, economic reasons are of greater importance than social reasons; The lack of income from rents of their assets needs to be prevented. For municipalities economic reasons prevail next to social reasons. The tax revenue from real estate is worth mentioning but especially liveability is of public interest. It is better to have lively neighbourhoods instead of abandoned business areas. Moreover, derelict business areas with many vacant buildings can lead to criminal activities (Remøy and Van der Voordt, 2007). 2.2 Opponents of transformation Opponents to office transformations argue that the vacant buildings are empty for a reason. These buildings belong to the mediocre and bottom-shelf stock of buildings: they are out-dated and therefore do not offer the contemporary comfort or requirements which new buildings do (Gann and Barlow, 1996). For the opponents of office transformation it makes no sense to invest in those buildings. Another argument is that in the end all buildings are conversable if you change the building regulation. In practice this leads to a disturbance of the market, because new suppliers can intervene with less strict rules and therefore have a benefit over the conventional housing suppliers. Table 2 summarizes the different reasons pro and against reuse.

REASONS FOR REUSE OF VACANT BUILDINGS Pro Con Sustainability/ Environmental: if a building can be reused no new land has to be cultivated and building materials do not end as waste. Out-dated buildings are vacant for a reason. Investments should not be made in these obsolete constructions. Urban policy: transformation can affect the population in an area. A more mixed population is often desired. Market disturbance: unfair competition between transformation initiators and conventional developers. For economic reasons transformations are desired. Rent for the owners and tax revenue for the municipality. For social reasons: some buildings have a social or emotional value to residents and should not be demolished. Table 2: reasons pro and against building reuse 2.3 Transformation barriers While trying to convert empty offices, developers face some particular barriers. A first set of barriers concerns the building itself; not all buildings are suited for A conversion due to technical reasons. Requirements for housing differ from those for office space (Remøy and Van der Voordt, 2007). Second, it is all about the location; people are not willing to live in peripheral areas, which are hardly accessible and lacking amenities. Thirdly, the current market situation determines whether it is a good moment to start a transformation process (Koppels et al, 2011). A fourth set of barriers is based on the economic situation of the owner/ developer. The initial costs are an important aspect whether or not conversion is economically feasible. Finally, laws and regulation affect the possibilities to change spatial functions. Table 3 gives a brief overview of the afore mentioned barriers.

BARRIERS Physical/Design Size/height/depth Layout Structure Locational Quality of the environment Safety/security Conflicting use ‘next door’ Views Parking space Demand Level of demand Interest rates Facilities Target market sector Price structure Financial/Economic Attitudes of owners/investors/developers Available funding Taxes Legislation Planning policies Building regulations Environmental policies Table 3: Barriers identified by Heath (2001) As mentioned above some crucial barriers are faced which determine whether or not a transformation is possible. When these buildings were constructed, up to fifty years ago, the need to change the function of the building was hardly considered: Buildings, which were no longer considered useful, were demolished and new constructions arose (Bullen and Love, 2010). To meet the requirements for housing, adaptations on for example ceiling heights and noise regulation are needed. In some cases though, municipalities are more flexible than others and ease the rules regarding building requirements for housing. In practice this leads to residential units, which are not completely in line with the housing standard and therefore can possibly lead to conflicts (RVO, 2014). 2.4 Adaptive building To what extent though is it possible to reduce these barriers and create a more resilient development/ building system? Resilience is often defined as the ability of a system to maintain its current form under external pressure (Buckingham and

Turner, 2008; Goudie, 2006). In the case of structural vacancy of office buildings, the external pressure is created by changes in society. In a relatively short time span, many offices have become redundant due to unforeseen circumstances. As future developments remain unclear, no one knows what will happen with the actual demand for office space. Resilience can be ameliorated in the future by building in a more flexible manner: Adaptive building (Wilkinson et al, 2016; Langston et al, 2008). Even though the Amsterdam canal belt probably wasn’t built according to adaptive building concept of Langston et al (2008), this area has shown to be resilient over time. The former warehouses, have served as shops, offices and residential units when the demand for a particular usage rose. The barriers encountered during the transformation of vacant buildings can be prevented or reduced by embedding adaptive building into the spatial development practices. Encountering fewer barriers implies that adaptive building gives real estate a larger potential to be reused in another function. It differs from conventional transformation projects in the sense that adaptive building does not require changes in the construction before a building can be reused (Wilkinson et al, 2016). This enhances sustainability because, instead of only partial reuse, the whole construction is suitable for reuse. Basically it eases the future transformation of real estate but on top of that, it also prevents buildings from not being convertible at all. It is this second feature that makes this an innovative concept. Like conventional transformation processes are a result of the abundance of structural vacant buildings and the pressure on the available city space; adaptive building is an answer to all those buildings that are still unconvertible. It requires awareness of the negative externalities and institutional change before it becomes common. The way buildings should be constructed, according to adaptive building principles, is well elaborated by Langston et al (2008). Basically it entails that potential transformation barriers are dealt with in the first construction and planning phases of the development. By taking the different norms for different spatial usages into account, no adjustments to the construction are needed when functional change is desired. The downturn of this concept is that considering all

the potential future barriers in the contemporary building practices requires extra investments (Wilkinson et al, 2016). Figure 2 and 3 shows two models of cash flows during real estate development and exploitation: one of conventional building practice and one of adaptive building. Prior to any development, agreements to buy or lease the land and investments for the construction are needed. Then every year the developer generates revenue from the lease of the property (Middelkoop, 2010). After X years a large-scale renovation is needed. Then, after another X years, the building becomes obsolete and the renters move to another building (Jaffe and Sirmans, 1995). Subsequently, the owner is left with a vacant building. Or, as the recent economic downturn has showed, companies go bankrupt or reduce their workforce and need less spacious buildings (Officerank, 2015). If this happens, an owner can leave the building vacant, try to start a reuse project or demolish the construction. Anyway, the land value remains. Land value Construc0on investment Yearly rental income T1 T2 T3 …. Renova0on costs Demoli0on costs Land value Building becomes obsolete Figure 2: Conventional development scheme (Middelkoop, 2010)

When adaptive building is applied, Additional investments are needed to comply with requirements from the building code for multiple uses. Then, similar to conventional development tactics a yearly income in generated and renovations remain inevitable. Yet, the stage where the building gets obsolete is postponed and the annual income continues. Renter may extend their rental contracts or the building is used for other purposes and therefore becomes attractive for a new pool of potential renters. In the current situation some buildings are not suited for transformation due to locational, physical/technical, financial, market and legislative barriers (Heath, 2001; Remøy & Van der Voordt, 2007; Langston et al, 2008). The potential for reuse increases if the aforementioned barriers can be lifted. The institutional setting influences the way things are done in the building sector. So it is the same institutional setting, which can take action to lift the transformation barriers. For some barriers it may be a matter of time before they are overcome; market situations can change in favour of demand for office space; and a changed financial situation of real estate owners can give them enough funds to initiate a transformation process. Other barriers on the contrary can be taken away by effort of the involved actors. Influential actors can change legislation, technical features Figure 3: Adaptive building scheme (Verweij et al, 2014) Land value Construction investment Annual rental income T1 T2 T3…. Renovation costs Additional required investment for adaptive building Land value No demolition costs Extra rental income due to adaptive building

and locational aspects of the current practice. This implies that some barriers are more easily overcome than others. Three examples are given: (1) If the technical design is a hurdle to convert a building to other usage, costs are often the reason to leave the building in its current state (Bullen & Love, 2010). These transformation costs can be lowered if; important aspects as the layout, access to the building, structure and cladding of the building, are adaptively built. (2) If there are legal restrictions to change the spatial function of a building because the zoning plan prohibits other usage; then, changing the zoning plan is a possible solution to make other usage allowed. (3) If the locational aspects of a vacant building form a barrier to transformation, the lack of nearby public facilities and poor accessibility, are often the main causes (Geraedts and De Vrij, 2004). These locational aspects can be improved by changing spatial planning practices: creating mixed-use areas and providing multiple modes of transportation (Jacobs, 1992). The three examples above elucidate that some transformation barriers can be taken away by the institutional setting: Laws can be changed and willingness and effort of the actors lead to locational and building characteristics that are in line with adaptive building principles. Adaptive building though, as presented by Langston et al (2008) and Wilkinson et al (2016), is an innovative concept. The implementation process of new ideas can be very time-consuming (Campbell, 1998). As the problem and solutions are set, it remains the question who has to start taking action? Is it the public or private sector that has to solve the set issues and bring the concept into the political agenda? And is there enough willingness and power to act? The challenge is to get this new concept institutionally embedded and replace the current conventional way of real estate development. Especially the interaction between the involved actors is paramount for the institutionalizing of adaptive building. The next chapter elaborates on this institutional framework.

3 Theoretical framework

This chapter elaborates on the theory about institutions. What are institutions? And why is the institutional setting so important? In contemporary politics, public policies are a way to solve issues that concern the whole society. Especially those issues that cannot be solved by individual action in a normal functioning market, but with collective actions, need active political involvement, in the form of policy. As an influential academic in the field of institutional analysis, Scharpfs work about institutional setting is thoroughly described. The institutional setting plays an important role in the process of embedding adaptive building into development practices. New concepts however, first need to get on the political agenda. Therefore, this chapter starts with an account of how new concepts are reflected into public policy. 3.1 Concepts and policy When new ideas emerge from science or practitioners develop new insight in their routines, an important next step is to get it institutionalized so it is reflected into policy. So also adaptive building needs to find its way to the political agenda and eventually get included in public policy. Campbell (1998) tried to get a better understanding of how different forms of ideas influence policy making. Therefore he conceptualized how ideas influence policy outcomes. To understand the “game” of policy making it is important to grasp theories of action arenas and rational behaviour (next paragraph). Within these action arenas four different types of ideas can be recognized and they all become part of the institutions in a different way. These different types of ideas all have their own features and influence policymaking in their own way. The types Campbell describes are: programs, paradigms, public sentiments, and frames (Table 4). 1) Programmatic ideas help to get a clear and concrete solution to problems. A clearly defined course of action needs to be taken to reach a desired outcome. These are often technical solutions. 2) Ideas as paradigms are, in contrast, to programmatic ideas not as straightforward. Multiple solutions are feasible and paradigms often reside in the background of policy making.

3) Ideas as public sentiments are based on the views of the public and offer a range of solutions to a certain phenomenon. These solutions however are not always politically acceptable 4) Ideas as frames provide the actors with symbols and concepts, which can be used to frame solutions to a problem on a more abstract level. Actors actively use these symbols to influence public opinion so their solutions become more acceptable. TYPES OF IDEAS

Cognitive idea Programs Paradigms

Normative idea Frames Public sentiments

Table 4: Types of ideas (Campbell, 1998) These four types differ in their normative and cognitive characteristics. Normative ideas fit into the existing norms and values of the political elite. A problem arises when plausible solutions to a certain problem cannot be implemented because they are against the prevalent normative structure. Especially a historical institutionalist administration governs according to these principles. Cognitive ideas and the organizational institutionalists on the contrary, can break through the wall of normative routines and are only limited by cognitive capabilities. Based on these fundamental differences Campbell (1998) states that different types of ideas have different effects on policymaking since they are not always politically and socially acceptable. Some ideas are perfectly acceptable to some actors and unimaginable to others (Campbell, 1998). Often, well defined policy proposals, are not being implemented because multiple actors are involved during the policy making process. They all perceive the issue and the possible solutions in a different way (Campbell, 1998). The concept of Adaptive building, encouraged by Langston et al (2008) and Wilkinson et al (2016) can be regarded as an idea to the issue of the amount of buildings, which cannot be converted. Adaptive building is a typical idea as a

program, a cognitive idea. The above-mentioned scholars, in the field of construction and urban planning, introduced the adaptive building concept. These scholars offer clear-cut solutions and indicate where changes in the current practice should occur. Some actors can see this cognitive idea as radical and therefore it can take a while before it can become common practice. Existing institutions can react reserved to radical changes (Campbell, 1998). The political elites might not want to conform themselves to new norms in the building industry. Scharpf (1997) states that actors are responding differently to external ideas, laws and regulation. These new ideas offer threats, constraints and opportunities to the actors and influence their perceptions and preferences and ultimately their behavior. This results in a continual interchange between the institutional setting and the emergence of new ideas. In practice it can take years before a, relative easy, solution can become institutionalized (Campbell, 1998). A shift in certain actor positions, a contingent event or reframing a solution can function as a window of opportunity for ideas to become institutionalized (Scharpf, 1997). The next paragraphs elaborate on the institutional setting; the actors involved, their motives and how they (co) operate to seek after their goals. 3.2 Institutional Setting This paragraph clarifies how different schools of institutionalists developed during the 1960s and 1970s and how they consider political decision-making. The main idea of institutionalists was to get an insight on how social and political outcomes can be influenced. During this rising interest in institutions three different groups of thought appeared: Historical institutionalism, Rational choice institutionalism and Sociological institutionalism Hall & Taylor (1996). All streams of thought have a somewhat different view on two issues regarding the institutional analysis. First of all, they wish to explain the relation between behaviour and institutions. Secondly, they try to explain the origin of an institutional setting and its change over time. These schools of thought developed independently and are not created as a reaction to one another. Hence there are, apart from fundamental differences, similarities between them too. The aim of this research is not to elaborate on the ideas of all types of institutional analysis. However, the features of rational choice institutionalism and the ‘game-theory’ (Geddes, Marks, Scharpf, 1997) are of

interest for this research and share a great amount of ideas; both will be explained below. 3.3 Rational choice institutionalism In this paragraph rational choice institutionalism and its main features are explained. The first and most distinctive aspect is the idea that all involved actors in policymaking act solely based on their fixed preferences (Hall & Taylor, 1996). This leads to strategic behaviour in which fulfilment of these preferences is the main objective. In response to the preferences of other actors and their expected behaviour, each step in a policymaking setting requires thoughtful action; often referred to as the calculus approach. A second feature entails the thought about politics. Rational choice institutionalists observe politics a series of collective action dilemmas (Hall & Taylor, 1996). As all actors pursue the attainment of their own preferences a potential policy outcome indirectly bring sub-optimal results for other actors. In practice, this means that any policy automatically has winners and losers. Because there is no institutional arrangement for collective action, actors do not tend to collaborate to search for a superior outcome. The third feature regards the possible policy outcomes. Policy outcomes are limited, as actors are expected to behave rational (Hall & Taylor, 1996). As the calculus approach makes clear that every move an actor makes is based on its own preferences and expectations of other actors, advantages can be gained if knowledge and resources between actors are exchanged. This shared information makes actors to cooperate and new policy outcomes, which at forehand seemed impossible, become plausible. The last characteristic of rational choice institutionalism is the idea that institutions are created by the involved actors themselves on a voluntary basis. Hence an institutional setting exists because it provides more benefits than another institutional form. As rational choice institutionalism differs from the two other thoughts about institutional analysis, some scholars criticized this type of analysis. Primarily the assumed rational behaviour is a problem for the sociological institutionalists. They argue that, apart from rationality, culture is a key determinant of behaviour (Bates, 1988). Secondly, the idea that institutions are created by the actors themselves and

will continue to persist as they offer the greatest benefits is a false assumption according to historical institutionalists. Decisions in the past are of importance for future policy pathways (Pierson, 1993). Contemporary practices are path dependent because they are formed by institutions of the past. Lastly, cooperation between actors is not always voluntary (Moe, 1990). Different actors hold more power and can influence the actions of other actors, which makes relations asymmetric. 3.4 The Game-Theory As a reaction to these critiques, Scharpf (1997) embedded the importance of culture and norms in the rational choice institutionalism. His game theory emphasises the need of different actors to participate in the ‘negotiation game’. Figure 4 shows the institutional setting according to Scharpf Figure 4: Institutional setting by Scharpf (1997) The institutional setting consists of the involved actors, the actor constellations and the modes of interaction. Ostrom (1986) created a similar model and called it the action arena. To understand how certain actors react to a problem in the current practices, it is important to get a clear understanding of the whole institutional setting. In the upcoming paragraphs the different parts of the institutional setting are worked out. 3.4.1 Actors The different actors involved in policy making all have their own capabilities and orientations. The capabilities of an actor are of different effectiveness under different circumstances. The specifics of a situation determine what the actor’s

capabilities are. According to Scharpf (1997) there is no general framework for the capabilities of actors since they change from situation to situation. Nevertheless, the resources possessed by the actors determine their capabilities for a large part. An actor with veto power as a capability can, for instance, limit the possibilities of other actors. Scharpf added some features of sociological and historical institutionalism to the theory of Hall and Taylor (1996). Behavior is not fully determined by calculated moves, but also by cultural values and socially embedded norms: the cultural approach. Rationality is bounded and socially constructed. Thus social and physical resources are determining the capabilities of an actor too. Their orientations (perception and preferences) are mainly determined by rational choices. Whether they want to stay at the status quo and keep the current policy or they pursue change, is determined by achieving set goals (Ostrom, 1986b; Scharpf , 1997). Because one actor is not capable of making policy on its own in a democratic system, cooperation/interaction between multiple actors is needed. Scharpf makes a distinction between individual actors and composite actors. The last group can act as a collective or a spokesperson can represent a corporation. When actors have multiple functions a role conflict can occur. To get insight in the actor’s orientations research is needed, as it is not always simply observable. Especially private sector parties do not always publish their business strategies and objectives. Public sector actors on the contrary, publish their opinion and strategy in public documents and therefore are easier to deduct. 3.4.2 Actor constellations Solutions to a problem, invented by one actor, are always affecting the other actors involved in an institutional setting. Often, resources possessed by others are required for these solutions. This interdependence of one another, makes actors seek for a mutual interest. In this process they can exchange resources, which makes them dependent on one another. This process is recognizable in the different actor constellations.

One should notice that these models of actor constellations assume that actors are only interested in their own pay-offs. In reality however, actors are certainly interested in the benefits or losses of others; they even may act in such a way that can increase the losses of other actors, even if that move may not yield the highest revenue for that actor itself. Scharpf makes clear that not all problems need to be solved through public policy. Sometimes, (market) actors can solve problems together without interference from the public sector. But not all problems can be solved in this fashion, because the market itself creates many societal problems. Then a public policy is required, in which often multiple actors are involved. The actor constellation describes the degree to which actors may cooperate based on their different orientations and capabilities. It gives an indication whether their strategic behaviour is compatible or not and if conflicts are likely to occur when they cooperate. From a game theory assumption it is expected that actors with comparable orientations will have a comparable strategy during policy interactions and therefor will form clusters with a shared desired policy outcome. Scharpf distinguished three basic actor constellations Pure conflict: In a pure conflict situation all gains for one actor imply direct losses to another actor. The counterpart is pure coordination. This situation leads to a well organized cooperation between the involved actors. Working together, with coordinated strategies towards a best outcome, all actors are likely to achieve a best pay-off. As the two situations above are quite rare in practice, a mixed motive game is more likely to occur. This situation leads to partially conflicting and partially compatible strategies of all actors. Actors constantly have to communicate with one another and choose to commit to a cooperative relationship instead of a defective one. Five sub-sets of this mixed-motive game are identified: 1) Individualism: Only profit and losses are considered. An actor aims at maximizing its profits. This situation reflects the assumption of neo-classical economics where self-interest is pursuit in all actions. 2) Solidarity: In a solidarity constellation cooperation is sought. Gains to all negotiation partners are regarded as positive; even when there is a clear asymmetrical division between profits. Desired outcomes imply gains for all

partners and shared risks. If an actor in a solidarity setting inevitably loses, often a move towards a competitive situation follows. 3) Competition: this situation is best described as a sports or war setting. The gap between one’s profit and the loss of another actor has to be as large as possible. Loss can still be regarded as positive if the competitor has a bigger loss; and gains can be perceived as a loss, when the competitor has larger gains. 4) Altruism: this situation implies that an actor is unconditionally helping another actor. Own gains or losses are irrelevant as long as the other party benefits from a situation. 5) Hostility: in this situation an actor’s own situation is irrelevant as long as the other actor is harmed or loses in a certain situation. This constellation describes sadism and hatred. Altruism and hostility are, according to Scharpf (1997), hardly seen in the modern democratic politics. Individualism, solidarity and competition are therefore most recognizable in modern political settings. 3.4.3 Modes of interaction Finally these different groups of actors interact with one another in what is called a “game of interaction”. Different modes of interaction are possible: From very hierarchical directions cooperation on the one side to loose arrangements between groups of actors on the other. Four types of interaction models are elaborated. These interactions are, except for unilateral action, shaped by the institutions itself. 1) Unilateral action occurs in the absence of any institutional structure in which actors on their own can change an existing situation. Simultaneously, other actors can do the same. These actors then, can decide to work together but agreements are easily infringed. This way of acting is only effective if no one is affected by changes that an actor has made, so no countermoves are expected. 2) Negotiated agreements occur when multiple actors are united in a network. They can come to various forms of agreements. If actors negotiate rationally, they will come to voluntary agreements of cooperation. In such cooperation, transaction costs are relatively low and the benefits from a product, created with shared resources leads to synergy. Reaching an agreement however can take time and trust is an important aspect in this way of interaction.

3) Majority Votes occur when many actors are involved in policy making. The group of actors is too large to make sure all interests are covered by one policy decision. All actors of relevance vote for a policy direction and even though some actors “lose”. In this majority vote compliance to the new rules/policy is expected. Majority vote is a very legitimate way to create policy as many supporters of the policy are known in advance. 4) In contemporary society however, citizens rarely make decisions. Referenda are not used for topics, which require in-depth knowledge and besides the outcome is most of the time only advisory. By voting, the population chooses representatives in a parliament, based on the programs of their political parties. Thereby, politicians do not need the consent of the citizens to impose rules changes or new laws. The government has the authority to govern and to give hierarchical directions but transparency and participation are more and more valued in policy-making processes. In the end the politicians are accountable for the decisions made, but less resistance is expected when active participants are consulted. 3.5 Law and Regulation Apart from the institutional setting, described above, Ostrom (1986) emphasizes the importance of the prevailing law and regulation. Existing laws are a product of the institutions and are based on decisions made in the past: They are path-dependent. These institutional rules influence the behavior of the actors and therefore the amount of possible outcomes in a process towards a policy. Rules determine what kind of behavior is required, prohibited and permitted. Ostrom (1986) makes a distinction between the formal and informal rules of the game. Formal rules (laws) are easy to understand and all involved actors should know them. Formal rules are requiring, prohibiting and permitting particular behavior. Parties, which are not complying with these rules, are easily left out of a negotiation game and sanctions are often known in advance. Informal rules are harder to grasp and are more socially embedded in habits of cooperation between actors (Ostrom, 1986). The actors in Scharpfs (1997) institutional setting are influenced by laws and regulation too. Actors’ behavior in the action arena is limited by their knowledge of the rules. In practice the model is more complex, since problems could be the result of certain rules and new policies might go hand in hand with rule-changes. Table 5 depicts the institutional setting used in this research.

INSTITUTIONAL SETTING Laws Laws and regulation form a legislative context in which actors interact and demonstrate particular behavior. Actors Orientations and capabilities determine the actors’ behaviour. Actor Constellation The actor constellation describes the degree to which actors may cooperate based on their different orientations and capabilities. Mode of interaction The mode of interaction indicates how interactions between actors are arranged. Table 5: institutional setting according to Scharpf (1997)and Ostrom (1986) 3.6 Hypothesis The previous chapter, in which adaptive building has been introduced and this chapter about institutions explain the two main concepts of this research. Based upon this theoretical framework some preliminary assumptions can be made: Liveability and sustainability are both positively influenced by a higher reuse potential of buildings in a certain neighbourhood (Remøy, 2008; Wilkinson et al,2016). Yet, liveability and sustainability are mostly and typical public sector policy objectives (Jacobs, 1992). If one considers the institutional setting in which actors strive to reach their own goals and objectives, a conflict can be expected between the orientations of the private and public sector. Private parties often seek profitable ventures whereas the public sector ideally serves the whole community. Therefore it can be expected that: 1) The extra investments, required for a high adaptive reuse potential, restrain private parties to embed adaptive building into their development practice, because it also serves the public interest. 2) The benefits of adaptive building (increased liveability and sustainability) serve the public interest. Therefore, it is expected that an institutional setting, in which a hierarchical direction is the prevailing mode of interaction, results in a higher ARP than cases with other modes of interaction.

3) When private developers’ orientations match public sector orientations, an advantageous situation for the public sector occurs. After all, their

orientations are served without their own effort. Therefore it can be expected that projects where this happens, are encouraged by the public sector.

4 Methodology

This chapter covers the methodology of this research. First the research design is discussed. The next paragraph gives an overview of the different cases. Thereafter the operationalization of the two key concepts of this research, adaptive building and institutions, is explained. The final paragraphs focus on the data collection and analysis of the research. 4.1 Research Design The research design is the link between the collected data and the answer to the main research question. For this research, a case study is used. According to Yin (2009) a case study is (1) useful for research in which the researcher has no control over events and (2) a phenomenon is happening in a real-life context. The cases in this research are examples where the concept/ phenomenon of adaptive buildings is applied in practice. The researcher wants to get an understanding of a particular phenomenon from the cases and therefore by no means wants to influence the cases. This gives the researcher the possibility to obtain information about the cases objectively. Another important reason to use a case study is (3) the focus on contemporary events: an understanding of the institutional setting that contributed to a particular physical outcome; that can not be deducted from old policy documents. It is a contemporary, but still valid, time-frame that needs to be understood. Moreover, (4) a case study is a suitable research design to answer “how”-questions. The second part of the research question focuses on how the institutional setting influences the adaptive building practices. This research makes use of a multiple-case design. The three different cases can be compared and gives the opportunity to compare different cases with one another. Conditions in one case might not be present in the other cases and can therefore be a pre-condition to a certain outcome. The units of analysis are the actors involved in these cases (see data collection). 4.2 Case selection Even though not much has been written about adaptive building in the Netherlands, there are some cases in which characteristics of adaptive building

have been applied. This makes them exemplifying cases within this sub-group of adaptive constructions. In order to answer the main research question, the three of those cases are thoroughly examined. Possibly and likely there are more projects where these features are applied. Outcomes from this research are applicable to those projects as well. By approaching the cases in this way, they can epitomize multiple cases in the Netherlands and they allow the researcher to find key social processes (Yin, 2009) of the institutions in the Netherlands that work as barriers/ incentives for adaptive building. The fact that the cases in this research are all within the same municipality limits the research to the context of Amsterdam. 1) An apartment building on the Steigereiland at IJburg, Amsterdam. The first case entails a building on the IJburglaan. The first parts of the large-scale IJburg project were completed in 2002. A few months later the first residents moved to the Steigereiland. Since then, this part of the city is featured by constant developments. 2) Solid 11. This solid is located at the Eerste Constantijn Huygenstraat in Amsterdam west and is in use since 2011. It is located at a former hospital area near the Vondelpark. Old aspects of the building are reused and a new glass construction is attached. It cost about thirty million euros (Nul20, 2016) and has a floor space of approximately eight thousand square meters. 3) Solid 1,2. The second Solid is located at the IJburglaan at IJburg as well; yet, it is another island (Haveneiland) on which this construction is build. Despite the name “Solid 1,2” this is one building and also in use since 2011. The floor space and costs are around the same prize and size as solid 11 (nul20, 2016). 4.3 Assessing Adaptive Building Looking back at the conceptual model makes clear that two concepts need to get operationalized for this research. These two concepts, within the red rectangles (Figure 5), are the adaptive reuse potential and the institutional setting. The independent variable is the institutional setting as this influences the ARP. This makes ARP the dependent variable. The transformation barriers are already identified in chapter 2. The next paragraphs elaborate further on measuring these concepts.

Figure 5: concepts to be operationalized The extent to which an existing building has the potential to be reused can be assessed with the “transformation meter” (Geraedts and De Vrij, 2004). Because Geraedts and De Vrij (2004) use this tool to determine the transformation potential of constructions, it strengthens the construct validity to use the same tool to determine the measures that increase the ARP. This tool is developed based upon the theory of building conversion barriers and entails a checklist approach to determine the ARP (Adaptive Reuse Potential) of a building. This tool uses a three- step model. In the first step, the quick scan, veto-barriers can be identified. Veto-barriers are barriers, which indicate that transformation is not possible. Typical veto-barriers are the legislation, locational and physical aspects of a building. These veto-barriers though, are not necessarily definitive. A change in rules can create possibilities and physical and locational barriers can (partially) be taken away. If there is enough willingness to end the structural vacancy of a building, influential actors can force a transformation. Even though “veto” seems insurmountable, changing conditions can influence the rules and willingness to change. Buildings with technical barriers for example, can be located on perfect locations for residential use. Yet, these buildings can still remain vacant if the owners are not willing to take a risk. The proximity of public transport or amenities can be a reason to persuade owners of vacant buildings to (if possible) refurbish them and change the functionality. The second and third steps focus on an analysis of the financial feasibility and the risks involved in the transformation program. These steps are of great importance to the owner of the property as they can use it to evaluate their venture. However, these steps are not of particular Adap%ve Reuse Poten%al (ARP) Transforma%on barriers Actors Rules/ Laws Ins%tu%onal Se?ng

interest for this research. At the end a final score is produced based on the features of a construction. Some of them are directly observable and others require more information (direct and indirect observables (Babbie, 1998)). This score makes it possible to rank different cases based on their ARP. Table 6 shows a shortened version of the transformation meter. This tool grades a construction based on its (1) locational and (2) building aspects. The locational aspects emphasize the direct surrounding of the building: What kind of facilities are located in the surrounding and at what distance; What is the predominant character of the neighbourhood; is the area easily accessible? One can imagine that resident demand a particular surrounding with features of a residential area. A typical (mono-functional) office location is often built according to different norms. For this reason the adaptability of the locational aspects is important too. The building aspects emphasize the technical features of the building. An office building is built for another purpose than living, and therefore does not necessarily meets the norms from the building code. The possibility to change the physical layout of a building makes a transformation more feasible. The quality of the building and the linked maintenance; the possibility to change the façade; and the effects of the surrounding on the building are determining factors whether or not acceptable living conditions can be created within a vacant building. The higher the number of checkmarks after each aspect, the lower the ARP. If only a few or no potential barriers are observed the building has a high ARP.

SHORTENED VERSION OF THE TRANSFORMATION METER Aspect Criterion Location Urban situation Office on remote industrial zone Office in the middle of office park Office in area defined as priority area for offices Land property Land rent Vacancy Vacant or more than one year Vacancy of surrounding buildings Character of urban situation Location near city edge, ring roads No greenery in the neighbourhood Social depreciation, vandalism Accessibility by public transport Distance to train station > 2 kM Distance to bus, metro, tram stop > 1 kM Accessibility by car, parking Distance to parking place > 250 M < 1 parking place/100 M2 _realisable Building Year of construction Building was built or renovated recently (three years) Character of the building Poor maintenance Extensibility Non extensible horizontally Non extensible vertically Structure Dense structural grid < 3,6 M Façade Façade openings not adaptable Daylight entry < 10% of the living area Environment Noise level at the façade > 50 dB Presence of dangerous materials in construction Table 6: Shortened version of the transformation meter (Geraedts & De Vrij, 2004) Even though the transformation meter is originally developed to determine whether a vacant building can be transformed, this research investigates whether measures are applied that make a potential transformation easier to pass through. Two examples will clarify this. (1) If the inventors of the transformation meter see the lack of amenities in the direct surrounding as a threat to a successful

transformation, then, ensuring the presence of amenities in an early developing stage enhances the reuse potential of a building. (2) When, because of legislation, low ceilings are perceived as a barrier to a transformation, then, ensuring certain ceiling heights enables future users to change the spatial function while still following the rules. This line of thinking is applicable for all aspects of the transformation meter. However, some aspects are not easily measured and others may be more idealistic than realistic. The transformation meter can aid to determine the adaptive reuse potential of the three cases. However, some measures can be taken from a different reasoning than Langston et al (2008) encourage. So apart from grading the different cases on their ARP score an understanding of the motives of the involved actors is of importance too. Hence, the stakeholders are interviewed to get information about their motives. These motives are part of the actors’ orientations and come to the fore in the next paragraph: assessing the institutional setting 4.4 Assessing the institutional setting As the theory of Scharpf (1997) made clear, the institutional setting exists of all actors, the actor constellations and the way they cooperate (modes of interaction). Since three cases help to get insight of adaptive building in practice, the institutional analysis focuses on these cases too. To get a better understanding of the institutional setting an actor analysis needs to be carried out. By doing so, an overview of the actors becomes available. Then, a better understanding of their orientations and capabilities is desired, because that information is an indicator of which actors are likely to pursue the same goals and are for that reason willing to cooperate and exchange their resources. In the theoretical framework (Chapter 3) the aspects of the institutional framework are explained. Actors, actor constellations and their modes of interaction however are not observable. Interviews are held with the involved actors in the different cases. Only those actors that had a stake in the actual negotiations about the developments are included in this thesis. Asking question about their objectives, the way actors cooperated and how negotiations worked out, provides direct information about the institutional setting and is the way this concept is operationalized. Then it is up to the interpretation of the researcher to identify the capabilities and orientations

of the actors: what sort of actor constellation was in use; and what mode of interaction was used? As Scharpf (1997) acknowledges too, some distinctive subgroups of the actor constellations and modes of interaction are rare in contemporary politics and are therefore not represented in this operationalization. Table 7 shows the aspects of the model of the institutional setting and their operationalization within this research. ASPECTS OF INSTITUTIONAL SETTING Variables Operationalization Actors Capabilities Possession of relevant resources Orientation Main objective Actor constellation Individualism Actor aims at maximizing his own profit Solidarity Actors cooperate so both can profit from the development Competition Actors try to profit from a development at the expense of other involved parties Mode of interaction Unilateral action Actors applied adaptive building features without interference of other actors of the institutional structure Negotiated agreement Actors share resources to develop projects and come to voluntary agreements Majority vote A large group of actors vote for a certain development decision Hierarchical direction Developments rules and requirements, additional to the standard rules, are imposed by the government Table 7: Operationalization of institutional setting Apart from the focus on actors, law and regulation, regarding the development of new buildings, are part of the institutional setting too. Are prevailing law and legislation a barrier or an incentive to adaptive building? Focusing on the law covers the barrier of legislation from the list provided by Heath (2001). Digging

deeper into the possibilities and restrictions of the zoning plan, the Building code and environmental law operationalizes this aspect of the institutional setting. Because these laws influence the characteristics of the building and the surrounding, they are the ones explored in this research. Figure 6: Transformation barriers emphasized in this research By assessing the institutional setting together with the theory of adaptive building three initial transformation barriers are emphasized in this research (figure 6). 4.5 Linking the institutional setting and adaptive building After operationalizing the institutional setting and the adaptive reuse potential of the cases, some connexions can be made. For each case a table is presented, similar to table 8. Which conditions in the institutional setting were significantly different and therefore led to a higher ARP in one case and a lower ARP in the other?

Linking the two concepts allows drawing conclusions about the effect of the institutional setting on adaptive building. INSTITUTIONS AND ARP PER CASE Case: X Involved actors Capabilities and orientations Actor constellation Mode of interaction Table 8: Institutions and ARP per case 4.6 Data collection In this research the techniques for collecting data are interviewing, the analysis of documents and observations. The use of semi-structured interviews gives the opportunity to go beyond the information from the answers of the interviewee; follow-up question can be used to go more ‘in-depth’. At the same time it is desired to cover particular topics with each interviewee. Interviewees provide the data regarding the institutions for each case. This empirical data can clarify the actual link between the adaptive building features of a case and the institutional setting. All questions are open questions, so interviewees are not prompted into some direction (See appendix for interview topic guide). For this research experts on adaptive building and stakeholders from the different cases where adaptive building has been applied are asked for an interview. The experts have the (scientific) knowledge about the topic in general and the stakeholders obviously can tell the story of a case; the involved actors, their motives and the development of the case are important empirical data, which are not always published. Covering the same topics with every interviewee safeguards the construct validity: by literally asking about their motives, the involved actors and the cooperation between them, it is guaranteed that the institutional setting is measured. This data however can be subjective because interviewees answer the questions based on their own perceptions. The interviewees are chosen based on purposive sampling (Bryman, 2008) and ‘snowballing’. Purposive sampling is used because for this research it makes sense to search for those people who have certain knowledge (e.g. the project leader of a case; a civil servant of the planning department; a developer; or a city councilor). These interviewees in their turn can recommend