Between a Rock and a Hard Place: Investigating Planners’ Learning Experiences in Amsterdam’s Fragmented Governance of Property Development

ABSTRACT

Fragmentation in terms of institutional complexity and a multitude of governance actors involved in property development changes the nature of planning practice. I investigate how planners in Amsterdam, who operate as project managers, learn from interacting with property industry actors, and how their learning experiences travel beyond the project scale to instigate wider institutional change. Planners in Amsterdam struggle as the distance between municipal policy directions and development practices ‘on the ground’ grows. I argue that putting a human face on these governance intricacies and incorporating private sector considerations in literature on institutional change in planning becomes increasingly pivotal.

KEYWORDS:

• Property development
• planning
• institutions
• fragmentation
• learning

1. Introduction

Flexible planning regulations to accommodate market dynamics and private sector interests change the fundamental nature of the planning profession. While scholars particularly criticise the dependence of public sector planning on property industry actors, the fact that planners are active participants in the construction of property markets is often neglected (Adams and Tiesdell 2010). Yet, based on planning regulations that enable room for negotiation and flexibility in decision-making processes, planners engage in ‘deal making on a project-by-project basis’ (Fainstein, 1994, p. 5). Each property development project represents a ‘bundle of relations’ (Blomley, 2017), which is in large part governed by the interactions between unique sets of governance actors. At the scale of a city, the multitude of these projects can be conceptualised as fragmented governance, with fragmentation referring to ‘the institutional complexity caused by the increasingly diverse types of actors (organisations) involved in contemporary urban development’ (Tasan-Kok et al., 2020, p. 2).

Planners operating in this fragmented governance landscape often find themselves in the position of project managers. In this paper, I argue that these planners’ learning experiences, gained from relations and interactions with property industry actors, deserve heightened academic and professional attention. I pose that learning can constitute a valuable means to connect actions, practices and lessons learnt from different projects with one another, and to instigate institutional change to guide the future behaviour of actors. Planning institutions – the formal and informal rules, norms and established patterns and procedures which frame human action and fashion behaviour (North, 1990) – not only transform rather slowly but tend to be unprepared ‘to cope with market pressures’ (Eraydin & Tasan-Kok, 2012: 2; Albrechts, 2016). The fragmented governance of property development accentuates this problem, as it requires wider institutional patterns to change based on project-specific experiences (Özogul, 2019).

Even though fragmentation has brought questions surrounding institutional change to the fore, the existing literature on institutional change in spatial planning and governance vastly disregards the intricate relationships that planners have with private sector actors (Özogul, 2019). To fill this gap, I answer the following research question: How do planning-project managers learn from interacting with property industry actors, and how do their learning experiences travel beyond the project scale to instigate wider institutional change? I operationalise ‘learning’ by utilising the learning-loop concept emanating from organisational theory (Armitage, 2008). Single-loop learning refers to the creation of alternative tactics to address concrete and context-specific challenges, double-loop learning to the adjustment of individual behaviour, and the triple-loop learning to the transformation of existing institutions on which single- and double-loop learning rest (ibid.). This concept is compelling for governance scholars as it incorporates different levels of transformation in a social structure (Pahl-Wostl, 2009) but has to date not been employed in studies on the governance of property development.

Residential property development in the City of Amsterdam serves as the case study context. I illustrate how the governance of property development in Amsterdam is fragmented due to regulatory and policy changes at different spatial scales, and the connected relations between public and private actors that are being formed at the city level. Rooted in the Dutch planning tradition of strong state interference, active land policy and extensive public landownership, the City’s experience with property development in which public actors have restricted influence is limited (City of Amsterdam, 2018a). However, housing demand, local politics, and corresponding ambitions of the City to build 50,000 new apartments by 2025 necessitates the active participation of planners in residential development projects in which private actors hold considerable powers (ibid.). I showcase how planners working as project managers for the City’s Project Management Office (PMO) learn from the concrete challenges emerging in interactions with property industry actors, and how they adapt their attitudes and behaviour in line with type of projects they are involved in.

The analysis reveals that local political changes coupled with rising land and property values boost the City’s confidence and in fact result in less regulatory flexibility, stricter regulations and tighter policy objectives in contrast to national government agendas that embrace private sector interests. Project managers’ single-loop learning occurs through creativity and experimentation to solve unforeseen challenges but can also increase their daily stress levels as they enter unchartered territory and must construct their own behavioural manuals. Double-loop learning is established through formal and informal channels; it involves changes in attitudes and behaviour, and the establishment of new skillsets to meet new conditions. A deficiency in triple-loop learning in the fragmented governance of property development, however, creates a conflicting situation for project managers. The distance between municipal policy directions and development practices on the ground is growing: Project managers are increasingly torn between the City’s policy requirements and the necessary actions in concrete development projects that require the accommodation of private developers and investors’ interests.

The paper begins by tying together literature on property, planning and learning by focusing on social relations as their common denominator. It is followed by the methodology which details the study’s methods and research approach. Then, I review planning and policy changes in the Netherlands that gave way to heightened fragmentation in Amsterdam’s development landscape and support my arguments with reflections provided by both public and private sector interviewees engaged in residential projects throughout the city. It is followed by detailed insight into learning, structured according to the three learning loops, from the perspective of planning-project managers. The paper ends with a discussion and conclusion that stresses the importance of putting a human face on the intricacies of relations between planners and private sector actors in scholarly work on property development and institutional change. Furthermore, I highlight the importance for practising planners to acquire specific skills and for local administrations to actively support the transmission of learning experiences to navigate complex relations with the property industry.

2. Planning, Property and Learning: Institutional Change through Social Relations

Notions of learning have long played a fundamental role in planning. Faludi (2000) proposed to understand planning as a form of mutual learning altogether. Friedmann (1971, p. 320) posed that ‘knowledge about the metropolis is revealed to us only in fragments’, correspondingly requiring a fluid and constantly evolving understanding of knowledge. In his ‘epistemology of mutual learning’, Friedmann (2017) points out the limits of expert knowledge and the importance of inter-personal dialogue and face-to-face interaction between planners and other actors coming together in concrete development projects. These factors are also crucial determinants in property development, but the existing literature tends to focus on unequal power relationships between actors, being less vocal on specifying how it is dealing with new forms of interpersonal interaction.

Despite its path-dependent and context-specific unfolding, scholars have documented how property development has become a major activity in urban development with strong repercussions on project-specific governance arrangements and the positions and roles of actors within them (Tasan-Kok, 2010; Fainstein, 2008). Almost 25 years ago, Krueckeberg (1995) argued that property should be more centrally featured in planning studies. Nonetheless, Fainstein’s (1994) work on ‘city builders’ in London and New York was for a long time one of the few works explicitly linking property markets and public sector planning practices. In recent years, however, academic engagement with the intricate relations between property markets, planning practices and policy has been growing (Theurillat et al., 2015). According to Blomley (2017), the difficulty of connecting property and public sector regulation stems from the fact that much urban land is privately owned. In this respect, property represents ‘a complex of overlapping tenure rights’ (Ryan-Collins et al., 2017, p. 16), allocated, regulated and taxed by planning regulations. Thereby, Ryan-Collins et al. (2017) argue, public sectors intend to either reduce property market imbalances and their negative externalities, such as through regulating land development and measures to increase housing affordability, or achieve other objectives entirely through property tax revenues (ibid.). Adams et al. (2005) advocate for a broader perspective in property research beyond private landownership and stipulate that neither property markets nor state processes and planning policy can be considered in isolation from each other. In a similar vein, Adams and Tiesdell (2010) characterise the relationship between planning and property markets as symbiotic, not dichotomous.

The wider literature discerns different perspectives on actual property development processes, such as from political economy, Marxist or institutional stances (see Drane, 2013 for a useful overview). Institutional perspectives prove most beneficial to counter dichotomous thinking and the widespread generalisations of property industry actors in planning studies (Özogul & Tasan-Kok, 2020; Campbell et al., 2014; Adams and Tiesdell 2012). They tend to approach property as a ‘bundle of relations’ capturing the intricate interplay between public and private sector actors (Blomley, 2017; Fainstein, 2008). Moreover, they offer alternative explanations to rational choice economics by treating social relationships, structures, and processes as equally ‘important as their economic equivalents in “explaining” property development’ (Guy & Henneberry, 2002, p. 2399). The property development process is hence best considered ‘a set of coordinating performances’ (Needham et al. 2011: 163) in which ‘transactions, like social interactions are conditioned by humanly devised rules, norms and regulations’ (Adams & Tiesdell, 2010, p. 193).

Considering each development project as a unique bundle of relations also sheds light on the fragmented governance under pinning property development (Özogul, 2019). In each project, a specific set of public, private and civil society actors as well as private individuals from different spatial scales come together and interact in decision-making processes (Figure 1). Thereby, public sector planners often take on facilitating and negotiating roles (Sehested, 2009; Fainstein, 2008). Flexible planning policy creates room for these negotiations and adds to the existing institutional complexity (Tasan-Kok et al., 2020). At the same time, an institutional perspective sheds light on the fact that all development projects, regardless of their unique actor constellations and relations, are embedded in a wider governance landscape (Figure 1). Scholars refer to this governance landscape, for example, as ‘governance regime’ (Pahl-Wostl, 2009), ‘meta-governance’ (Jessop, 2016), or the ‘rhythm of established governance processes’ (Healey, 2006, p. 299). These conceptualisations have in common that they draw attention to the socio-economic context, policies, regulations, and institutions that guide and steer the interactions of governance actors.

Figure 1. Analytical framework

While project-driven urban development may have the benefit of better responding to context-specific circumstances, existing research points out concerns about disconnected, piecemeal forms of development solely accommodating private sector interests, and a lack of planning guidance and public sector control (Özogul, 2019). Dissatisfaction with this situation led to numerous scholars to call for ‘transformation’ (Healey, 2018), ‘transformative practices’ (Albrechts, 2010), or ‘transformative planning practice’ (Steele, 2011), to name a few. As Albrechts et al. (2018, p. 4) declare, ‘the transformative agenda is a modern term for structural change’. In this paper, I argue that learning can play a fundamental role in instigating these changes. Most learning theories model learning as a collective process since the opportunities to learn without interacting with others and being exposed to new ideas are limited (Wink, 2010). Hence, they share the focus on social relationships and interaction of the institutional perspectives on property development.

The learning-loop concept differentiates levels of learning that are required to achieve ‘structural change in the governance regime as a whole’ (Pahl-Wostl, 2009, p. 358). Therefore, it is especially suitable for governance studies (ibid.). I apply this concept to property development and equate loop-learning with different levels in the fragmented governance of property development (Figure 1). Figure 1 shows a conceptual visualisation of fragmented governance based on three exemplary property development projects, embedded in a wider governance landscape. Each project encompasses a unique set of actors from multiple spatial scales as well as their relationships. Civil society actors and private individuals also partake in property development and are included in the visualisation to show the complexity at hand. Yet, the focus of this paper lies on the relations between public sector actors, specifically planners, and property industry actors.

Single-loop learning encompasses new actions, tasks, and strategies to address and solve a concrete problem at hand. In my conceptualisation, single-loop learning occurs when project managers respond to concrete and project-specific challenges without repercussions on their individual behaviour or on other projects: While single-loop learning involves the experimentation with new measures and instruments, generally no doubt is cast on established routines and processes (Pahl-Wostl, 2009).

In double-loop learning, learning does not concentrate on a concrete problem at hand but leads to adjustments in behaviour and actions of actors. Thus, while strongly linked to single-loop learning, it extends to the sphere of governance actors and moves beyond the project scale. Correspondingly, in my own conceptualisation of fragmented governance of property development, double-loop learning refers to actor-specific learning. Figure 1 includes additional small arrows to represent individual governance actors’ double-loop learning on an exemplary basis. It includes reflection processes on actors’ own positions, and doubt of existing guidelines and assumptions to achieve certain goals (Pahl-Wostl, 2009). In planning studies, this level has been approached in terms of individual capacity building (Schmitt & Wiechmann, 2018) and critical self-reflection of planners (Fox-Rogers & Murphy, 2016). More concretely related to the property industry, Adams and Tiesdell (2010) for instance argue that planners should move beyond a shallow understanding of market dynamics and acquire the ability to fundamentally challenge a property developer’s financial calculations.

Transformation of the wider governance landscape that determines the frame of reference for single- and double-loop learning is captured by triple-loop learning. Learning at this level results in critical reflections on overarching paradigms and structural constraints (Pahl-Wostl, 2009). It may lead to the inclusion of new actors and formation of new relationships and changes in overarching regulatory frameworks, norms, and values (ibid.). In planning studies, this level has been approached in terms of institutional learning, which ‘reflects upon the extent to which structures and routines have been installed and how this knowledge is managed and secured for future purposes’ (Schmitt & Wiechmann, 2018, p. 28). Triple-loop learning is a continuation of, and rests on, single- and double-loop learning, and corresponds to different levels of change highlighted by institutional theorists: ‘each level is more costly to change than the previous one’ (North, 1990, p. 83).

Most scholars engaging with transformation and institutional change in planning studies take the wide spread influence of private sector actors in planning decisions as point of departure (Özogul, 2019). Thus, it is even more surprising that the role of the private sector in these discussions is largely neglected or stays at a superficial level (ibid.). To fill this gap, this paper now turns to the different levels of learning of project managers operating in the fragmented landscape of property development in Amsterdam, to explore how institutional change is instigated on the basis of their experiences with property industry actors.

3. Methodology

The research for this paper was conducted as part of ongoing work within the framework of the WHIG Project (What is Governed in Cities: Residential Investment Landscapes and the Governance and Regulation of Housing Production), which examines the inter-relationships between contemporary investment flows into urban property markets, and the governance arrangements and public policy instruments that are designed to regulate them. The analysis of this paper is twofold. First, I illustrate the fragmented governance of residential property development in Amsterdam. My argumentation is based on a thorough analysis of changes in the Dutch planning system and the regulatory framework affecting residential property production in Amsterdam. I discussed my preliminary findings on fragmented governance with policy makers from social and spatial planning departments at the City of Amsterdam during an event in March 2019. Furthermore, as my focus lies on relationship dynamics between public and private actors, I conducted 15 interviews with senior planners and planning advisors working for the City of Amsterdam (5), private developers (5) and private investors (5). The interviews allowed me to gain insight into residential property dynamics in the city as well as opinions and assessments on general relations between public and private sectors from several viewpoints.

Second, I focused on the learning experiences of planners working for the Project Management Office (Projectmanagementbureau; PMO) of the City of Amsterdam, embedded in this wider governance landscape. The PMO is an administrative unit that specialises in project management. Among its employees, approximately 40 specialised planning-project managers are working in the PMO and form an integral part of the interdisciplinary teams that administer residential development projects in Amsterdam (City of Amsterdam 2019). I conducted in-depth, semi-structured interviews (Table 1) with nine of these planning-project managers. Interview questions centred around respondents’ work experience, challenges encountered in relations and interactions with property industry actors, strategies employed to respond to these challenges, and factors influencing problem-solving abilities. ‘Learning what’ questions were deliberately left open to allow for insights without pushing respondents into a pre-defined direction. Four interviewees had a senior position meaning that they oversaw larger areas within the City which in turn were sub-divided into smaller project areas. Three interviewees worked as project managers responsible for one of these areas, and two worked as junior project managers in a project team (Table 1). Furthermore, three project managers predominantly worked in an area development project and five in an area transformation project, a difference which is explained in detail in the following section, and which turned out to considerably affect the respondents’ perspectives.

Table 1. List of interviewees in the project management office

Respondent #	Position	Work experience	Project type
R1	Senior project manager	25 years	Area development
R2	Senior project manager	12 years	Area transformation
R3	Senior project manager	11 years	Area development
R4	Senior project manager	27 years	Area development
R5	Project manager	20 years	Area transformation
R6	Project manager	20 years	Area transformation
R7	Project manager	3 years	Area transformation
R8	Junior project manager	4 years	Area transformation
R9	Junior project manager	1.5 years	Area transformation

All interviews were transcribed and subjected to a systematic content analysis. The first set of interviews with senior planners and planning advisors, developers, and investors was used to illustrate and justify my conceptual understanding of fragmented governance of property development in Amsterdam. The second set of interviews with project managers, which forms the core of the empirical analysis, was scrutinised in terms of emerging themes, topics, and issues. I ranked the latter by their importance according to respondents and their frequency of occurrence, creating the foundation of this paper’s arguments. To account for differences in terms of position, work experience and dominant project type, I indicate which respondents particularly underscored certain issues throughout the text. Additionally, I provide a selection of direct quotes that best captured some of the key themes that emerged. Thus, even though some respondents are quoted more frequently than others, their interviews did not weigh more in the overall analysis.

4. Dutch Planning and Amsterdam’s Fragmented Governance of Residential Property Development

The City of Amsterdam is located in the Province of North Holland, and is with 854,316 inhabitants in 2018, the largest city of the Netherlands. The current political and discursive focus of property development strongly focuses on the production of housing. Housing prices are rising rapidly: between April and July 2018, apartments cost 450,000 euros on average, which is more than 5,000 euros per square metre, while the average price for the country lies at 283,000 euros (City of Amsterdam, 2018a). Coupled with an expected population growth of approximately 15,000 people per year, Amsterdam is in urgent need for new housing and residential property production has become strongly politicised. Housing was a key component of the municipal elections in 2018, and the coalition accord of the elected, left-leaning government underscores the importance of a ‘just city’ connected to the housing market; the plan is to construct 50.000 new apartments by 2025. Reaching this objective, according to the City, requires new forms of interaction with property industry actors but is lacking experience in this respect (City of Amsterdam, 2018a).

This lack of experience can only be understood in relation to the changing nature of Dutch planning in the last decades. Internationally, Dutch planning has long held the reputation of being a planners’ paradise and been praised as a desired model (Buitelaar & Bregman, 2016). Active land policy, a well-established Dutch planning strategy in which municipalities prepare land for development and then hand it over to private sector actors in line with their policy objectives, plays a fundamental role in this perception. The traditional active land policy, resting on extensive public land ownership and agreements with property industry actors, has been described as entrepreneurial before the ascendance of influence of private sector actors, due to the close collaboration with the private sector (Louw et al., 2003). However, since the 1980s, the types of relationships between public and property industry actors have shifted steadily. There has been a shift from top-down public sector led planning towards more flexible approaches embracing private sector interests in property development projects (Buitelaar et al., 2011).

In the 1980s, central government transfers to local municipalities reduced and public–private partnerships were more actively utilised to finance particularly larger-scale urban redevelopment (Hendriks & Tops, 1999). In Amsterdam, extensive development schemes such as the South Axis or the IJ-bank development, started to reverse the logic of active land development with public sector planning following ‘private preferences for development’ (Hoetjes et al., 2006, p. 181). Furthermore, in the 1990s, land and housing markets were deregulated. A new national policy, the Fourth Memorandum on Spatial Planning Extra (Vierde Nota over de Ruimtelijke Ordening Extra) of 1991, restricted areas for residential property construction. Coupled with growing demand for owner-occupied housing, the policy deliberately created scarcity of land, and increased commercial developers’ interest as ‘(t)he development gain to be enjoyed by buying unserviced land became big enough to compensate for the risks’ (Needham, 2007, p. 193).

Nonetheless, the most common practice for commercial developers was to purchase land and then agree to sell it ‘to the municipality, against a price more or less similar to their costs in acquiring it’ (van der Krabben & Jacobs, 2013, p. 780). Once it has been serviced, land was sold back. Yet, municipalities retained their position as active public land developers. Simultaneously, ‘in the social rental sector the land development subsidies disappeared’ (Louw et al., 2003, p. 361). Social housing associations were privatised in 1995, and consequently became more heavily intertwined with private property market dynamics, investing in commercial properties, and adopting more business-like and negotiation-based relationships with local municipalities (Conijn, 2005).

In 2008, the Dutch planning system underwent a major reform. The New Spatial Planning Act (Wet Ruimtelijke Ordening) which determines the structure of the Dutch planning system reshuffled powers, responsibilities, and expectations between governmental layers (Gerrits et al., 2012). As the first substantial change since 1965, the major revision to the planning law replaced the hierarchical structure with spatial visions, which are only binding on the level that develops them, aimed to simplify regulatory complexity and ‘stimulate a proactive, “hands-on” and development-oriented planning’ (Tasan-Kok et al., 2019, p. 10). Additionally, the aftermath of the 2008 economic crisis put a hold on major redevelopment schemes, and in 2010 the national government enacted a Crisis and Recovery Act (Crisis- en Herstelwet) to spur market activity and speed-up planning decisions following economic downturn.

In Amsterdam, the crisis led to ‘a deep reform of its consolidated planning institutions’ (Savini, 2017, p. 857), and development patterns of piecemeal approaches to curb financial risks. The public-sector driven approach to development and large-scale active land policy in which local administrations took on considerable financial risks became fragile, and gave way to smaller, organic development projects in which the government took on the role of ‘enabler’ (Buitelaar & Bregman, 2016). Public and private actors interviewed to gain insight into Amsterdam’s governance of residential property indicated the far-reaching repercussions of the 2008 economic crisis on fragmentation. Strategic planners working for the City explained that nearly all large development projects were put on halt, and how the City tried to accommodate property industry actors’ interests to keep at least some level of construction going. They said that they were aware of the repercussions of downscaling housing production, as population numbers and housing demand continued to rise during and after the crisis. Nonetheless, financially, they described their hands as being tied. Interviewed property industry actors working in the construction, development, and investment of housing, remembered how many larger firms went bankrupt and smaller and more diverse actors started to emerge on the development landscape. Furthermore, they described the welcoming attitude of the City and the flexible regulations compared to today’s situation.

In the years that followed, the land market recovered but the housing shortage intensified. Following scandals of mismanagement, fraud and land speculation of social housing corporations, amendments to the Dutch housing law (Woningwet) in 2015 restricted their activities to the supply of affordable housing. Simultaneously, a new Landlord Levy Act (Wet Verhuurderheffing) instated in 2014 disproportionately affected housing corporations (Priemus, 2014). In Amsterdam, the percentage of social housing units owned by housing corporations reduced from 50% in 2008 to 39% in 2018 (City of Amsterdam, 2018b). Although agreements have been made to slowdown the process to give priority to middle-income and affordable housing (AFWC, 2018), waiting for a social housing unit in Amsterdam can take up to 10 years. And as housing corporations are not allowed to produce affordable housing for middle-income groups, there is a perpetual lack of rental housing for households earning between 39,055 and 61,707 euros annually (City of Amsterdam, 2020). Meanwhile, annual rent increases of existing units in Amsterdam are among the highest in the country (Statistics Netherlands, 2019).

Rising rents can be attributed to international as well as rural to urban migration, and to the growing interest of both foreign and national investors in the residential buy-to-let and built-to-let market (Nijskens et al., 2019). Public sector interviewees exhibited mixed views on investors, ranging from describing investors as greedy, to complaining that Amsterdam is in the hands of financial capital but also to underscoring the importance of institutional investors to produce more middle-income rental housing. Meanwhile, the national government, for instance, actively promotes the housing and mortgage markets as investment opportunities, including at international real estate fairs. Private sector interviewees explained that in fact, the close-knit development community has become more diverse in the last years, with different types and particularly more international investors entering the development landscape.

The City of Amsterdam rolled out its ambitious housing plans, but with housing shortages not only prevalent in Amsterdam but throughout the Netherlands, regions, provinces, and the national government also develop numerous spatial plans, housing visions, and cooperation agreements with property industry actors. Private sector interviewees expressed concern about the numerous and fast-changing policy objectives and regulations, making property development extremely complex. They were not against public sector interference and even restrictions on their development practices, but rather wished for more leadership and clearer, long-term policy directions based on which they could orient themselves in their development plans for the coming years.

When the City of Amsterdam took a strong stance on steering development practices, however, conflicts started to emerge. Following the local elections in 2018, the local administration introduced the 40–40-20 rule, stipulating that new development projects must include 40% social housing, 40% for middle-income groups and 20% for high-income groups. In addition, the homes must remain in the medium-priced segment for at least 25 years and the rent increase has been limited. According to interviewed developers and investors, these rules are considerably diminishing the profit margin, and turning residential development in Amsterdam unprofitable. Correspondingly, they said, the City is not making optimal use of the development potential of institutional investors to produce the urgently needed housing for middle-income groups, and property industry network organisations and the City entered negotiations.

While residential property production continues, a clear differentiation was made by public sector interviewees in terms of area development and area transformation. Area development captures development projects that follow the more traditional approach of Dutch active land policy. Approximately 80% of the land in Amsterdam is publicly owned, and the local administration leases much of it to private parties through approximately 200.000 contracts. However, throughout the Netherlands, municipal land available for housing development is diminishing (Buitelaar, 2010). This is also the case in Amsterdam. The City, therefore, recognises that area transformation is necessary to reach its housing ambitions (City of Amsterdam, 2018a). Area transformation is a development approach that remains embedded in a public-sector strategy for a specific part of the city, but its implementation is more dependent on private landlords and property industry actors’ willingness to cooperate. Thus, while both area development and transformation could be classified as public–private partnerships, the nuances between the two are pivotal. Diverse types of actors and organisations, and their relations in area development and area transformation projects affect planning-project managers learning experiences in Amsterdam’s fragmented governance of property development.

5. Property Development at a Crossroads: Incorporating New Learning Experiences or Returning to Old Principles?

The Project Management Office (PMO) is subsumed under the cluster Space and Economy (Economie en Ruimte), one of the four major clusters of Amsterdam’s local administration. Its approximately 480 employees are specialised in complex multidisciplinary and integrated projects in the social, physical and economic domains in collaboration with other municipal departments, districts and neighbourhoods, and other municipalities and the region (City of Amsterdam 2019). The PMO does not develop policy independently, but project managers are internally hired to work under the direction of other divisions. The approximately 40 specialised planning-project managers work under the leadership of the department of Land and Development (Grond en Ontwikkeling), and the department of Space and Sustainability (Ruimte en Duurzamheid). The analysis investigates these project managers’ learning experiences and showcases how Amsterdam’s property development finds itself at a crossroads signified by two opposing processes: one going into a private sector-driven direction, and one intending to return to the local administration’s traditional strong position in urban development.

5.1 Single-loop Learning in Response to Unchartered Territory in Transformation Projects

The development of alternative strategies and tactics in the form of small, incremental changes in response to emerging challenges is encompassed by single-loop learning (Pahl-Wostl, 2009). A main difference in the importance of single-loop learning within the Amsterdam context emerged between area development and transformation projects. Each development project by the PMO has a project manager who leads a multidisciplinary and multidivisional team that always includes one technical person responsible for technical questions pertaining the development, one urban planner or designer, and one person specialising on land prices and leasing (R8). Structures and procedures in area development were described as well established, and interviewees stated that unexpected events or occurrences occur rather seldom. As one interviewee explained, ‘you write down what you want’, organise a tender which explicates the rules and standards of the developments and then select a developer, which requires no one-on-one or negotiations with developers before the start of a project (R4). Based on the tender, a developer is selected, and a leasehold contract provided. Even though there remains some risk that the developer drops out or cannot fulfil an agreed obligation, the risk for the City remains rather manageable. If a project does not work out, the developer is losing a fee that was paid at the beginning of the leasehold contract, and the whole procedure starts from the beginning (R1).

The situation, however, looks very different in area transformation projects, where interviewees saw unexpected things happening all the time (R5; R8). They explained how the property market is very dynamic and that there is a general lack of experience of collaborating with property industry actors in new constellations. It was described as leading to primarily ‘learning-by-doing’ (R2; R3; R5; R7). The lack of complete or partial public landownership, for example, was considered as a major challenge in this respect. It requires interaction and the development of working relations between a variety of actors with different needs, interests and personalities, ranging from large private developers to public housing corporations as existing property or landowners (R8). R2 recalled a case, for instance, of repeatedly having conflicts with a specific housing corporation but passed it on in the administrative hierarchy where the issue would be put on the agenda and addressed with the help of a mediator. To address challenges emerging in negotiations with private sector actors directly and spontaneously, a ‘4 eye principle’ is followed (R5). This means that a project manager never enters a negotiation alone, which was also considered as an instrument to prevent corruption and fraud (R5).

During negotiations, the inclusion of a representative of the Land and Development department of the City, which is ultimately responsible for the land lease in case of public landownership, was seen as crucial leverage (R6). In transformation projects, the main advantage for the City was considered a zoning plan that did not yet allow the designated or envisaged function, providing the City with influence despite lacking landownership (R6; R9). R3 drew a ‘carrot and stick’ analogy in which desired rezoning could be utilised in negotiations and push for a fulfilment of the City’s policy objectives. When facing very concrete problems, interviewees reflected positively on the institutional channels at the PMO through which they could seek help and support. For instance, project managers can request an ‘InterVision’ which provides external feedback of other project managers in the search for possible solutions to concrete problems (R4). Even more valued, however, were informal discussions with colleagues at weekly lunches or during coffee breaks, where information could be exchanged; though R2 admitted that in a hectic work environment, sharing experiences does not always have priority.

In area transformation, no established manual and procedures exist but project managers stressed the value of trial-and-error and testing out new ideas (R4; R5; R7; R9). The freedom was highly valued, and as R7 stated, ‘it makes it fun, because your boss doesn’t know any better than you do what to do.’ Similarly, R6 explained, ‘sometimes we just do something, it is a best guess and check later if it works but there is no one from higher up that tells us what to do.’ This freedom, however, was also connected to a downside, namely, the stress and pressure that is put on the project manager to fulfil policy objectives which were particularly addressed in relation to the ambitious housing production goals (R4; R6; R9). R7, for instance, described the ambition to build 50,000 houses by 2025 as enormous, and confessed: ‘we are worried all the time because if we don’t pull this off, we have a big problem.’

Thus, single-loop learning in transformation projects occurs on a regular basis and is strongly connected to trial and error, and experimentation. In most projects, solutions to emerging problems fall back on mechanisms and tools of municipal control of land lease and zoning plans. When these conditions are not met, however, project managers exhibited a certain degree of insecurity. The single-loop proves beneficial as it draws attention to these differences by uncovering the reactions of project managers encountering challenges in property development.

5.2 Double-loop Learning as Institutional Queries and Critical Self-reflection

In double-loop learning, actors develop alternatives in behaviour as well as more critically reflect on established procedures and their own positions (Pahl-Wostl, 2009). In fact, most interviewees saw an adjustment of their behaviour based on their experiences and face-to-face interaction with private developers. Again, the frequently highlighted learning-by-doing approach plays a fundamental part in this and is also reflected in the set-up of the PMO in terms of internal career trajectories. In general, the PMO only hires new project managers in junior positions, and through mentoring and annual performance reviews, including a detailed assessment of skills and the development of a personal trajectory and learning plan, they can progress to higher positions (R2). Furthermore, interviewees described a rather elaborate system of courses they are obligated to follow but can also choose on a voluntary basis to enhance their skills. These skills were described as rather generic, such as stakeholder engagement and not specifically targeting interaction with private sector actors (R7).

A common thread in the conversations was that no form of education or course can prepare a planning-project manager for real-life situations. A good project manager, according to interviewees, has to rely on gut feelings, intuition and instincts, assessing the body language or the personality of a property industry actor in order to decide which strategy is best to approach and negotiate (R1; R5). For instance, R5 elaborated on the hierarchical set-up within the City, which is known to private developers and who are hence not always taking project managers seriously if they are not operating at the top level. Therefore, R5 underscored that depending on the personality of the developer, it is crucial to make the City’s position clear to be taken seriously. Though, as R5 added, ‘strict’ behaviour can only be practised in a certain way, and especially at the beginning of a project, flexibility is important. Interviewees with 10 or more years of work experience reflected on the shift that occurred in terms of the importance of technical skills in the past, and soft or ‘people’-based skills in the present in order to establish long-lasting relationships and connections between different parties in the development process (R2).

While age and years of work experience as project manager did not play a crucial role in differences in perception, the fact whether the project manager worked in a rather traditional area development or a transformation project appeared to be correlated to the reflection on their own position. Interviewees working in area transformation were more critical about the established procedures and institutions, and more strongly reflected on their own role as a negotiator instead of decision-makers. R4, for example, working in a traditional area development, made it clear that everyone has a main task, the City, the developers and all other parties involved in the process, stating that ‘if everyone sticks to their core business, meaning, everyone knows what they are good at and what they are experienced at – for example, us as a City know what is best for a particular development, what is required, we make the guidelines for the project, etc. and the developers make the plans, it will be better as a whole.’ R1 pointed out that if the City would listen to everything that people wanted, no housing would be built.

Interviewees who faced more challenges in their every-day work in transformation projects, on the other hand, admitted that the City sets so many goals and objectives which cannot always be met. R6 explained that the City is still trying to fit into their new role as facilitator instead of as decision-maker, and ‘only slowly accepting that they are “not god” anymore’. R2 described his personal development in this respect as shifting from technical skills to the creation of long-lasting relationships which means giving up on average 50% of the initial demands made by the City in an actual transformation project. Similarly, R7 described her personal development trajectory of having become much more professionalised whereas in the beginning she was very optimistic and thought everything is possible. Simultaneously, the way of describing her ‘professionalised’ behaviour was not connected to frustration or a sense of defeat but simply as a different approach to working which requires a different skill set than in traditional area development.

A major problem in this respect was seen in the existing courses and training which transferred a skill set related to area development and did not catch up on adjusting to the new conditions though things were slowly starting to change (R6; R7). On the basis of the initiative of several project managers, the ‘learning house’ organisation within the PMO with around 130 active people to develop, share and spread knowledge, for instance, formed a new group to exchange and share practices from transformation projects (R5). There was widespread consensus that the expertise of project managers needed to grow in relation to market-based and market-driven development initiatives, sharing, for example, instruments and mechanisms on how to get a public school financed within a private-led development without a tender procedure (R7; R9). Thus, double-loop learning in the PMO is encouraged through formal channels in terms of the development trajectory of junior project managers.

As analytical lens, double-loop learning provides a more nuanced picture of actors in property development and their internal struggles. It draws attention to the channels through which abstract connections between different projects are established through the sharing of learning experiences. The difference in self-reflection and adjustment of behaviour is closely related to personal exposure to landownership patterns and new forms of property development. Project managers working in transformation projects provided critical evaluations of existing procedures and underscored the need for enhanced mechanisms to share learning experiences and support one another. Others, however, believed more firmly in traditional role divisions between public and private sector actors and favoured returning to old principles. Thus, embodied experiences represent the basis for critical reflection of established procedures. The fact that those involved in area development have difficulty sympathising with the experiences of those working in area transformation slows down the establishment of formal strategies to share experiences and learn across development projects.

5.3 Rising Land and Property Values as Impediment to Triple-loop Learning

Triple-loop learning, culminating single- and double-loop learning, results in transformation of established procedures and routines, and may involve the implementation of new approaches in line with ‘corresponding efforts to change structural constraints’ (Pahl-Wostl, 2009, p. 360). Interviewed project managers recognised the need for improvement and overarching change as ‘the market moves fast, and the policies are still “behind”’ (R4). At the same time, however, it was stressed that trying to follow the speed of the market is not always desirable and instead suggested that the City should slowdown, re-evaluate and develop a carefully thought-through long-term vision (R4). Being explicit about the need for improved learning and exchange of experience in transformation projects, reasons for their absence were ascribed to the City’s comfortable position. R2 explained that ‘the City has a lot of money so they can also do a lot of things very carefully’ and R6 stated that ‘more and more private parties want a piece of the cake.’

The current situation where land and property prices were rising and the demand and interest of private sector actors in urban development was high, was contrasted to the situation a few years ago. R6, for instance, recalled that in the beginning of her current project briefly after the 2008 crisis hit, they were almost ‘begging’ private developers to build housing, which resulted in a lot of flexibility to accommodate private developers’ interests. In contrast, now with a changing market situation, the City finds itself in a different position, which R6 saw for instance in demands that can be put on private sector actors. Similarly, the first contracts were very flexible and gave a lot of room to the private sector to make their own decisions, while now with increased market interest, regulations are becoming tighter.

Nonetheless, policy demands were often described as ‘unworkable’ (R7), and the general difficulty of a project manager as having to combine the wide array of policies and regulations from different departments that are applicable when it comes to property development projects (R9). Every week, according to R1, there are a lot of changes and new ideas but when you want to make ‘housing production like a machinery’, constant change is not good. With trust being a crucial element in the interactions between project managers and property industry actors, too much flexibility in terms of changes on behalf of the City was perceived as counter-productive to establish long-lasting relationships (R7; R8).

Regarding the question whether project managers’ opinions are sought after by policymakers in the policy development process, answers were very mixed. Two senior project managers (R2; R4) reflected positively on the consultation process and stated that there was some continuous dialogue. The rest, however, were more critical and stated, for example, that 90% of the time, they are simply informed and even if they are consulted, they doubt that anything is done with the provided input (R6). Measures to bridge the gap were highlighted, for instance by taking policymakers on field trips to the project areas. The objective is to give policymakers real insights from the actual work being done as ‘the City should really learn how it works on the other side of the table’ (R7). Still, the ‘gap between policy and reality is too big’ and it was suggested that policymakers participate in project management team meetings in order to be ‘confronted with the practical reality of issues’ (R5).

R1, responsible for an area development project, gave the example of structural change based on learning experiences on the ground. He described how 20 years ago, it took a long time to discuss contracts with developers as land was leased to a developer, the City set up a contract and then the discussion started. With the introduction of the tender procedure (bouwenvelop), however, this negotiation would be cut shorter as the City clarified its objectives beforehand. He was confident with the approach and did not see it changing in the future. R7, in contrast, working in area transformation is hoping for a wider shift in the City and an increased sharing of practices, finding it difficult to work closely together with a private party and demanding all kinds of standards, and regulations such as the 40–40-20 rule to be upheld, but developers saying that it is not profitable enough: ‘So we get a little bit discouraged. The most challenging part is that we, as City, try to fit into the new role but throw down a lot of standards’ (R7). The project manager’s task was described as having to find the balance in a project so that alderpeople approve it, ‘but also to make the developers happy’.

Triple-loop learning is the most difficult one to achieve and instead of incorporating experiences and transferring learning practices of project managers in transformation projects, it appears that the gap between policy and implementation within the local administration is currently widening. Nonetheless, the triple-loop lens provides a valuable and hopeful perspective on the case at hand. It uncovers the channels that fail to translate experiences ‘from the ground’ and the limits of project managers as governance actors in the wider institutional arrangements. Simultaneously, however, it connects small, incremental changes and individual efforts and reflections to wider institutional change. It highlights that the actions of project managers and their creation of linkages through sharing experiences, strategies and skills beyond the project-scale are a fundamental exercise that should be enhanced in order to learn how to implement public policy objectives in changing circumstances.

6. Discussion and Conclusion

This paper investigated planning-project managers’ learning experiences in Amsterdam’s fragmented governance of property development. The phenomenon of fragmentation is by no means a novel one and I acknowledge that all governance landscapes are fragmented to some extent. However, even for Amsterdam’s local administration that is known for extensive public landownership and the power and control of spatial planning matters, the degree of fragmentation is intensifying. Both public and private interviewees involved in Amsterdam’s property market testified to this fragmentation, its intensification and challenges. I particularly focused on planning-project managers and revealed how fragmentation creates a conflicting situation for them. Municipal policy directions and some project managers aim to return to more traditional ways of development with the public sector in the lead. Development practices in area transformation projects, however, requiring project managers to adopt new forms of cooperation and compromises with property industry actors.

The purpose of my argument is not to advocate for institutional change allowing for more flexibility and influence of the property industry, or to embrace private sector interest more fully in urban development. Rather, my intention is to highlight the complexities of dealing with property development dynamics within a public administration, and particularly point out the challenges of project managers working on transformation projects, who find themselves between a rock and a hard place. Their dilemma is to fulfil spatial policy objectives involving compromises with property industry actors in a tightening regulatory and policy framework. The tightening framework makes the process more difficult and may even deter private sector actors engaging in housing production if they deem it not profitable enough. Both not delivering policy objectives, but also not upholding the municipal government’s standards, are not an option. While this specific situation is unique to the Amsterdam context, seemingly irreconcilable situations as such are not uncommon to many actors in complex governance landscapes.

Thinking in terms of fragmentation sheds light on the intricate relationships between governance actors. With this paper, I aim to contribute to the growing number of institutional studies on planning and property development that lament the over-simplification and dichotomous differentiations between public and private actors. By showcasing differences within Amsterdam’s local administration up to the point of internal struggles and reflections of individual project managers, I acknowledge both the struggles and efforts of public sector actors and put a human face on them. Contemporary governance landscapes are extremely dynamic and traditional role divisions are not applicable anymore.

Operating in a fragmented governance landscape, planners need to be trained to understand and disentangle governance complexities. Despite working on a project-by-project basis, they need to be trained to ‘see the forest for the trees.’ Single-loop learning of project managers in Amsterdam underscored the importance of taking land, property and their relationship with public-sector planning interventions into account. Correspondingly, I concur with Adams and Tiesdell (2012) that specialised property market knowledge and capacity building with private sector actors are indispensable in this process. Furthermore, double-loop learning of project managers revealed the importance of social and inter-personal skills in both negotiations and building lasting relations with the private sector. The contemporary planner cannot operate as a technocrat but needs to be able to collaborate with, and mediating between, various governance actors. The general exchange of information within Amsterdam’s PMO is well established, yet single- and double-loop learning was strongly linked to exposure to new development dynamics and approaches. Thus, there is room for improvement especially when it comes to an explicit focus on exchanging learning practices in relation to inter-personal interaction with property industry actors, within and particularly beyond the PMO.

The learning-loop concept proved beneficial to operationalise the abstract notion of institutional change and link it to both project- and actor-specific solutions and behaviour. In Amsterdam, triple-loop learning can be actively improved through targeted actions that bring policymakers and project managers together, and addressing issues and concerns, considering a two-way stream between policy-making and implementation. Theoretically, incorporating learning experiences with property industry actors is an element that is largely neglected in the existing body of literature in planning studies that advocates for institutional change. I contend that this thematic area deserves much more attention as private sector actors are indispensable in processes of urban change. My primary focus on planning-project managers provides just one public-sector perspective, and further research should explicitly focus on perspectives, strategies, and opinions of property industry actors to gain a more comprehensive picture of the fragmented governance of property development.

The employed single-case study approach has the advantage to generate in-depth insights but limits generalisations. The example of residential property development in Amsterdam illustrates institutional complexity, but future research should focus on comparative studies to further explore the interplay between planning institutions and property markets. There is no blueprint solution for governance actors to manoeuvre out of challenging situations, such as the one of project managers in Amsterdam. Nonetheless, planning organisations should be proactive in developing channels to share learning experiences and discuss fast-changing conditions that shape planning practice instead of being stuck in old ways of thinking. I believe that it is the planning scholars’ task to support planning practitioners in this pursuit by more strongly featuring private sector dynamics in theoretical contentions on institutional change in planning and governance.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

The project What is Governed in Cities: Residential Investment Landscapes and the Governance and Regulation of Housing Production (https://whatisgovernedincities.eu/) has received funding from the Open Research Area for the Social Sciences (ORA) under grant agreement no. 464.18.113.