Hard, soft and thin governance spaces in land-use change: comparing office-to-residential conversions in England, Scotland and the Netherlands

ABSTRACT

In recent years, converting office buildings to residential use became a high-profile issue in the UK and in the Netherlands. There has, however, been differentiation in the policy response between England and Scotland (planning policy being devolved within the UK), and the Netherlands. We conceptualize this differentiation through the lens of variegated neoliberalism in the forms of hard, soft and thin governance spaces. England, where planning deregulation is more strongly adopted, represents a thin governance space. Scotland, where there has been little policy change, illustrates a hard governance space. The Netherlands represents a soft governance space, where proactive partnerships between government and developers predominate. This paper characterizes these distinct governance spaces and explores their impact on housing delivery and place-making, and the impact of underlying ideologies and planning culture(s) in governing office-to-residential conversions in the three countries. Drawing on national government assessments and statistics, interviews with stakeholders, and case study data from three cities: Leeds, Glasgow and Rotterdam, we conclude that while both hard and soft governance spaces, to different degrees and with different merits, are environments that enable planning, thin governance spaces – being driven more by ideology than notions of good governance – imply weak planning and place-making.

KEYWORDS:

• Change of use
• office-to-residential
• planning
• permitted development
• deregulation
• governance spaces

The planning and governance of land-use change

Over the last decade in Europe, there has been an increased policy focus on the governance of the change of use of commercial buildings to residential use. This emerged in the context of a marked housing crisis in the UK (particularly in South East England and London) and high levels of office vacancy in the Netherlands (particularly in the Randstad and Rotterdam). In most European planning systems, changing the use of a building requires compliance with relevant land-use plans. Arguably, regulating land-use change protects uses that could be otherwise threatened by others with higher market value. Land-use planning also enables some control over the mix of uses in a neighbourhood, a critical place-making dimension considering the interdependency and synergetic potential between different uses (Beauregard 2005; Canelas 2018). The capacity to enact and implement planning regulation to manage office-to-residential conversions can thus be perceived as good governance.

Land-use change is usually regulated through planning, including development rights being held within the state and restricted for private individuals (EPSON 2018), but this was not always the case. The rise of comprehensive planning systems in the late 19th and early 20th centuries, which included the control of property development and land-use change, constituted ‘a major shift in the ideological landscape towards egalitarian liberalism’ (Davoudi, Galland, and Stead 2020, 21). If changing ideologies, on the one hand, respond to changes in material conditions, they also reflect and are ‘constructed within a discursive formation that is constrained or enabled by social and cultural contexts’ (Davoudi, Galland, and Stead 2020, 21–22). By the end of the 20th and early 21st centuries, with the rise of neoliberal ideology, and following years of welfare state retrenchment, deregulation and what Peck (2001, 452) calls the ‘slippery policy spaces’ of policy reform or the ‘thin policies’ of a weakened state, became a predominant form of public intervention.

In the Netherlands and the UK – both England and Scotland – as in many other national contexts, we have seen distinct planning reforms in recent years which appeal to the logics of neoliberalization, yet in variegated ways (Allmendinger and Haughton 2013; Peck 2013; Inch 2018; Davoudi, Galland, and Stead 2020). In particular, in both England and the Netherlands, policy changes for office-to-residential conversions have been aimed at creating a more responsive (de)regulatory environment while incentivizing the market (Muldoon-Smith and Greenhalgh 2016; Clifford et al. 2018; Remøy and Street 2018; Ferm et al. 2021). In England, since 2013, central government has experimented with changes to national planning regulations so that office-to-residential conversions no longer require planning permission, usually required on a case-by-case basis under the British tradition. These are now instead classified as ‘permitted development’, where developers have only to notify to planning authority through what is referred to as ‘prior notification’. At the same time, in Scotland, where a devolved planning system operates independently from that in England, the issue has received less attention, and full planning permission for office-to-residential conversion is still required in a process akin to the previous approach in England between 1948 and 2013. In the Netherlands, since 2010, changes included a series of measures, both central and local government-led, which involved a growing interaction between the public and the private sector, together with best-practice knowledge gathering and sharing. England, Scotland and the Netherlands lend themselves to the study of alternative planning approaches to office-to-residential conversions because there have been different policies adopted in each territory.

In recent decades governance theories have populated the debates on urban planning (de)regulation, including proposing distinct analytical governance models involving characterizing their ‘participants, objectives, instruments and outcomes’ (Pierre 1999, 372). Many of these debates have been formulated in terms of contrasting binaries including government/governance, concertation/competition, bureaucracy/market, formal/informal, participation/efficiency (e.g. Pierre 1999; Haughton and Allmendinger 2008). Pierre (1999, 389) defines models of urban governance as ‘clusters of policy objectives, cultural values, norms, practices and patterns of recurrent political behaviour’. Haughton and Allmendinger (2008) suggest the concepts of hard and soft governance spaces as landscapes of distinct administrative practices.

Expanding on Haughton and Allmendinger’s (2008) hard and soft governance spaces, and inspired by Peck’s (2001) idea of thin policy, in this paper we propose three prototype governance spaces, hard, soft, and thin, as a heuristic device helpful to problematize the case of office-to-residential conversions in England, the Netherlands and Scotland. We suggest that Scotland operates in a hard governance space in the management of office-to-residential conversions. This means that planners at the local authority planning control over conversions. In the Netherlands we found a soft governance space where planners play a definitive role in steering the market toward planning aims and coalescing state-market interests. In England currently, the management of office-to-residential conversions takes place in a thin governance space, where a prominently centralized governance system deliberately pursuing a deregulatory agenda has stripped local authorities of planning and place-making powers – key tools of good governance. With significant weaker delegation of powers to planners than the hard and soft governance spaces used in Scotland and the Netherlands, respectively, thin governance spaces undermine ‘good’ planning – planning concerned with housing quality and place-making.

The aim of the paper is threefold. Firstly, it explores the different governance spaces in which office-to-residential land-use changes are managed in England, Scotland and the Netherlands. Secondly, it evaluates, quantitatively and qualitatively, the varied impacts of these different governance spaces, particularly with respect to housing delivery and place-making. We do this by exploring national government policy and the role of the local state in Leeds, Glasgow and Rotterdam, second-tier cities located in each nation. Thirdly, more broadly, this article seeks to contribute to debates on the impacts of ideology and planning culture(s) in the governance of planning policy and (de)regulation. In doing so, we are conscious of the way in which comparative studies can help improve planning practice (Bourne et al. 1984) yet acknowledging that planning systems are deeply embedded in their broader socio-economic, political and cultural contexts (Nadin and Stead 2008).

In the next section, we consider the nature of the planning system in England, Scotland and the Netherlands, and any relevant recent reform before reviewing the literature on governance spaces, and the literature on the merits and limitations of office-to-residential conversions as perceived by key stakeholders. We then turn to our comparative research findings, debate and conclusion.

England, Scotland and the Netherlands: office-to-residential governance spaces

Planning is now devolved in the UK, albeit with some similarities in recent reforms and structures in each constituent nation due to the common source of the Town and Country Planning Act in 1947 (Clifford and Morphet 2015). The planning systems in both England and Scotland are infused with a dominant liberal social model, a pragmatic approach to governance and common law legal systems (Nadin and Stead 2008). Although prior permission for property development (including the change of use of a building) has been required in England and Scotland since the 1947 Act, minor developments have always been allowed to take place without planning permission, termed ‘permitted development’ (PD). The precise definition of what is PD has changed over time, defined by secondary legislation.

In England, since May 2013, central government extended PD to include office-to-residential change of use. Initially a temporary measure, this was made permanent in April 2016. Under this new policy framework, developers seeking to convert office buildings into housing have only to notify the relevant local planning authority (LPA) through prior notification. Under PD, LPAs have been unable to evaluate or influence the principle, location or design of conversions, nor can they negotiate planning obligations as they can within the planning permission process in the UK. They can only evaluate prior notifications with reference to a limited scope of technical planning matters, including transport and highways, contamination, flooding risks and noise. The extension of PD to include office-to-residential conversions in England has generated tensions between central and local government. As noted in the media, ‘when the legislation was introduced, half of the local authorities in the UK asked for exemption, which should have been a warning sign that maybe it wasn’t such a great idea’ (Wainwright 2015, online). Exemption areas retain the hard governance spaces of office-to-residential conversions in England. From the 165 LPAs in England that applied for exemptions, only 17 were granted exemption areas (DCLG 2013a), which shows the strong hand of the central government in thinning local authority governance powers.

In Scotland, planning control for office-to-residential conversion has not been deregulated in the same way. Instead, developers wishing to convert an office building into housing must apply for full planning permission in the same way as used to be the case in England, with the LPA considering the case on its merits, including all aspects of the proposed design and its impact on place, and against the backdrop of policies they have made in their local plan. Although there has been some debate about increasing the scope of PD in Scotland, including office-to-residential change of use is currently not under discussion (Beveridge, Biberbach, and Hamilton 2016; Scottish Government 2016, 2017).

In the Netherlands, as in many continental European nations, planning consent is more of an administrative process of checking compliance with strict policies and plans – for example in terms of use, height and volume – rather than an exercise in case-by-case professional judgement and flexible discretion (Booth 2003). Notions of planning are deeply embedded in Dutch society (Faludi 2005), with regulatory interventions guided by prior assessment and a strong tradition of engaging stakeholders (Nadin and Stead 2008). There is a comprehensive, integrated approach to planning, with involvement at the national, provincial and local levels. Since the global financial crisis, the Netherlands has also deregulated some aspects of planning. As Remøy and Street (2018) highlight, there have been changes to the Dutch legislation including repurposed buildings not being required to achieve new build quality, which incentivises developers to convert, rather than redevelop, buildings. However, LPAs have retained planning control over office-to-residential conversions and, as we will see later, the emphasis on promoting conversions has been on network building as opposed to simply deregulating planning and hollowing out the local state.

It is useful to situate governance reforms around planning policy and control for office-to-residential change of use within a broader context. The planning reforms seen in England, Scotland and the Netherlands, and many other national contexts, in the aftermath of the global financial crisis, appeal to the logics of neoliberalization in variegated ways (Davoudi, Galland, and Stead 2020). As Lord and Tewdwr-Jones (2014) discuss, the successive reforms in England give the impression that planning is ‘under attack’ through a sustained programme of neoliberalism. Roodbol-Mekkes and van den Brink (2015) also highlight an increased emphasis on economic development as the explicit goal of planning through neoliberal reforms in England and the Netherlands. Similarly, Inch (2018) notes that planning reform in Scotland has been permeated with managerial and neoliberal logics. Within a broad neoliberal frame, the nature of planning reforms has varied between the three nations, hence the potential role for a comparative study. By mobilizing the concepts of hard and soft governance spaces (e.g. Haughton and Allmendinger 2008; Haughton et al. 2010) and suggesting the conceptual counterpart of thin governance, we aim at advancing a heuristic device to help to identify the underlying rationales in the formation of these distinct administrative landscapes.

The literature on hard and soft governance explores institutional reconfigurations of traditional governance spaces in the context of the growing importance of global economic competitiveness and the role that space and place play in this process. Hard governance spaces, as Haughton and Allmendinger defined them, are ‘the hard spaces of governance activity, involving statutory responsibilities, linked to legal obligations including democratic engagement and consultation, all of which take time, and come with a particular set of public and professional expectations around their choreography’ (2008, 143). The authors suggest that soft spaces also ‘have a role to play, providing a series of alternative institutional spaces in which to imagine possibilities for future place making’ (pp.143), where there is greater room for stakeholder engagement and negotiation and less stringent obligations. The authors note that when discussing hard and soft spaces, their aim is not to suggest that hard spaces should be replaced with soft(er) spaces, but rather that distinct spaces can ‘create complementary and potentially competing opportunities for development activities’ (2008, 143). Similarly, Rhodes (1996, 653) suggested that there is nothing intrinsically good or bad about distinct governance forms and that choice should not be ‘necessarily or inevitably a matter of ideological conviction but of practicality; that is, under what conditions does each governing structure work effectively’.

In this paper we suggest expanding the hard and soft governance binary to incorporate thin governance as a counterpart inspired by Peck’s definition of thin policy involving ‘dispersal “out” to markets … while powers of institutional coordination and ideological control remain firmly located (albeit in a restructured form) at the centre’ (2001, 452). The threefold heuristic device of hard, soft and thin governance spaces provides a way to locate the alternative spaces of the governing of office-to-residential conversions more clearly within distinct governance space prototypes, while critically interrogating the role of ideology and planning culture(s) in the formation of these spaces. Hard spaces stand for more traditional government leading spaces, very much as defined by Haughton and Allmendinger (2008). Soft spaces stand for governance spaces with greater room for stakeholder engagement. Thin spaces are mostly market-led, dominated by market interests and knowledge sources. The thinner the governance space, the more deregulation and limited local planning control can be found. Given the importance of non-state actors in soft and thin governance spaces, next we turn our attention to the views from non-state actors on vacancy and vacancy policy.

Viewing from the private sector angle

For real estate developers and investors including banks, pension funds and insurance companies, (re)building the built environment is a critical source of capital preservation and growth. In the context of availability of development finance and stable demand for space, speculative and surplus space is likely to be created (Weber, 2015). Particularly in the office market, tenant aspiration for new and landmark buildings can be high both because continuous technological changes affect workspace needs, and because of the way corporate identity, space and narratives around innovation and competition are intertwined (Abramson 2012; Weber, 2015). As tenants move to newer buildings, older ones become vacant. Vacant buildings are largely an undesirable phenomenon as they not only significantly diminish the building’s value, they can also devalue surrounding buildings and negatively impact the neighbourhood. In the case of office buildings, capital value is generally a function of rental value. If a building lies vacant, its current use and exchange value erodes. Eroding value should be an incentive for building owners to act and deal with vacant buildings. The principle of reusing vacant buildings instead of consuming resources into demolishing and building new buildings is hard to contest (Abramson 2012).

Previous research on the reuse of existing building stock, and more specifically of office-to-residential conversions, has however, identified some of the difficulties with conversions. Focusing on London, Barlow and Grann (1995) noted that in the early 1990s, the slump in the office market was accompanied by a significant shortage in affordable housing. Yet, despite the recognition of the environmental benefits of reusing redundant buildings, there were a limited number of conversions due to a combination of planning policy, technical and financial feasibility reasons. The authors make a case for greater flexibility of LPAs’ planning powers to stimulate conversions, suggesting that ‘with good design and careful planning, it should be possible to introduce flexible standards more widely without reducing the quality of accommodation or creating a two-tier planning system’ (1995, 273). In the Netherlands, Remøy and van der Voordt (2014) found that despite good reasons for converting vacant offices into housing, surprisingly few buildings had been converted between 1999 and 2011. The authors identified three main obstacles: zoning law, which made conversions unattainable within pre-established land-use classes; building codes, with conversions unable to meet requirements; and the municipality’s unwillingness to cooperate (2014, 388). In both cases, the research suggests that in order to enable more conversions in the future, the design process of new buildings, building codes and planning should allow for greater flexibility. There are, however, further factors beyond building regulations and planning as to why people do not convert buildings; these became apparent through our case study research discussed later in the paper.

Case study approach: a national and city level comparison

Questions have emerged over how best to govern office-to-residential conversions in England, Scotland and the Netherlands. How governments have responded in terms of planning has, however, varied considerably between each country. We investigated this through a study which took a detailed, in-depth and mixed methods approach looking at the impact of office-to-residential change of use becoming PD in England through case studies of five local authorities (Camden, Croydon, Leeds, Leicester and Reading). We then undertook comparative visits and interviews to consider the governance of office-to-residential change of use in Glasgow, Scotland and Rotterdam, the Netherlands.

Focusing on the discourses, policies and practices shaping office-to-residential conversions, this research included 30 interviews with a range of stakeholders (particularly local planners but also local politicians, developers and their agents) where interviewees were asked about their perceptions and experiences with regard the governance of office-to-residential conversions. The research also included site visits to 568 converted buildings in 2017, and desk work including identifying and analysing relevant qualitative and quantitative government data on conversions, and local planning databases for case conversions submitted under full planning applications or prior notifications. Through site visits and planning databases, information was collected on each case including number of units, typologies and conversion status. The researchers collectively visited the first few buildings together to test and calibrate the data recording and observation methodology. Data analysis combined conventional and directed content analysis and descriptive statistics. For the conventional and content analysis we draw on the governance literature to identify defining characteristics of distinct governance spaces and their underlying ideological and planning culture(s) assumptions. None of the identified alternatives is necessarily good or bad (Rhodes 1996). However, the different paths chosen by the different countries seem to have different practical results.

Combining qualitative and quantitative data sources and methodologies enabled greater insight into the subject than would otherwise be obtained by using either type separately. More specifically, this mixed methods approach enabled us to explore both process and outcome in the governance of office-to-residential conversions. The full findings of that study are reported in Clifford et al. (2018). In this paper, we present a new analysis and comparison of our data between England, Scotland and the Netherlands and between the cities of Leeds (selected from the English case studies as a large secondary city with a similar population to the Scottish and Dutch cities), Glasgow and Rotterdam.

A national level comparison: England, Scotland and the Netherlands

As noted in the introduction, the approach to promote office-to-residential conversion in England included the deregulation of planning control in May 2013, which depleted planners’ powers. This anti-regulation approach was evident in the impact assessment produced alongside the expansion of permitted development (PD), which suggested the planning system was a key reason there would be vacant offices when there is demand for housing and that ‘in the absence of controls the market would allocate land according to its most valuable use reflecting the underlying demand’ (DCLG 2013b, 5). It was also claimed that conversions in unsuitable locations, such as industrial estates, were unlikely as housing providers would find such sites unattractive (given the apparent discipline of the market). There is no mention of housing quality in the impact assessment, and the only mention of design is in relation to the savings that developers could achieve by not having to work-up the full design details and enter negotiations over design associated with planning permission (DCLG 2013b). The policy was reviewed in 2015, heralded as successfully delivering thousands more homes, and made permanent in 2016.

In Scotland, there has been no particular policy attention to office-to-residential conversion and the government has not sought to proactively promote such change of use or to reduce related regulatory planning policy. The traditional full planning permission governance applies, which does allow LPAs to consider issues of design quality in some detail, providing the opportunity for a more comprehensive consultation and negotiation in relation to the permission.

In the Netherlands, although there has been some deregulation related to building regulations (already mentioned) and the revision of local plans, this has been part of a much wider package of measures to promote conversion of vacant buildings to residential use, much of which has involved a soft governance approach. The approach taken has included a taskforce established in 2010 by the then Minister of Housing, Spatial Planning and the Environment (Remøy, De Jong, and Schenk 2011). This included research, dissemination, lobbying, partnership working and monitoring. Key stakeholders involved include local government, developers, housing associations and finance bodies. In June 2012, a covenant was signed to agree actions between these stakeholders to deal with office space vacancy, mostly aiming to stimulate the market into embarking on office-to-residential conversions. Best practice toolkits on conversion have been published, and the minister met with housing associations to encourage them to start converting vacant office buildings into housing rather than consistently pursuing new build construction. Local authorities were required to produce a spatial vision indicating which office buildings they thought should be retained as offices and which should be converted to other uses.

The comparisons between the three case-study locations – England, Scotland and the Netherlands – must of course consider that, first, the nature of the planning systems is somehow distinct. In England and Scotland development management is based on a discretionary system where compliance with national and local strategic objectives is accompanied by a subjective judgment of the merit of the individual application. This contrasts with land-use/zoning planning system in the Netherlands, where planning permission is partly covered through certifying compliance with the binding zoning / local plans, partly through the building regulations permit and partly through an environmental permit. Additionally, a key difference between England and Scotland is that as planning policy in Scotland is a devolved matter, the 2013 and 2016 legislative changes that extended PD to office-to-residential conversion in England do not apply in Scotland. Office-to-residential conversions in Scotland still have to go through full planning application.

Second, in England, the issue of office-to-residential conversion policy change has been approached from a motivation to deal with housing shortage, albeit in a context of some surplus vacant offices in some locations. The Netherlands has approached the issue from a motivation to deal with a surplus of vacant office space, albeit in a context of some housing shortage (particularly around the Randstad region). In Scotland, the issue of office-to-residential conversion does not stand high on the agenda despite a significant housing shortage.

National official statistics enable us to explore some of the potential practical implications of the distinct governance spaces within which office-to-residential conversions are being managed in the three countries analysed. In England, data from MHCLG shows new dwellings created from change of use of buildings (Table 1). Data from the Scottish Government shows new dwellings created from conversion of non-domestic buildings (also Table 1). In both cases, the buildings converted do not have to have been vacant before conversion and this data includes a wider range of original building uses other than just offices. In the Netherlands, data from the Ministry of the Interior and Kingdom Relations, shows the number of new dwellings created through the conversion of vacant non-domestic buildings, covering a broadly similar period (Table 1). For a relevant comparison of these official statistics, figures are also presented in the context of the population size of each country (Table 1). The population of England in 2017 (our point of comparison) was recorded as 55,619,400 and the population of Scotland as 5,424,800 (ONS 2020). The population of the Netherlands was recorded as 17,081,507 (CBS 2020a).

Table 1. New dwellings from change of use in England, Scotland and the Netherlands.

Year	2011-12	2012-13	2013-14	2014-15	2015-16	Total	Total per 1000 inhabitants
England^a	12,590	12,780	12,520	20,650	30,600	89,140	1.6
Scotland^b	669	568	532	659	696	3124	0.6
Year	2012	2013	2014	2015	2016	Total	-
Netherlands^c	3755	7520	6395	7825	8095	33,590	1.9

^a Source MHCLG (2020).

^b Scottish Government 2019, 2020.

^c Sent to authors by Ministerie van Binnenlandse Zaken en Koninkrijksrelaties.

The comparison suggests that there is a significant difference in the number of new dwellings created through change of use between England and the Netherlands (high and increasing) and Scotland (low and stable). When adjusting for population size, delivery per thousand inhabitants is highest in the Netherlands, showing a total number of units for the relevant period per thousand of inhabitants of 1.9, compared with 1.6 in England and 0.6 in Scotland (Table 1). In Scotland, the number of new housing units created through change of use adjusted to population seems to have been less than half that in England and without the increase in the number of conversions seen in both England and the Netherlands over the same period. It is important, however, to look beyond just the headline statistics of the number of housing units created and consider the broader impacts of the different governance spaces in real places. We explore this now by considering conversions in three case study cities within the three countries being considered.

A city level comparison: Leeds, Glasgow and Rotterdam

Alongside considering national level policy, our study also included a detailed view at the municipal level. Our local case studies included Leeds (England), Glasgow (Scotland) and Rotterdam (the Netherlands). All three sites are significant cities within their national context, but not the national capital, and all three have been sites for post-industrial, socio-economic transformation. Table 2 summarizes some comparative data on the three cities.

Table 2. Summary of key indicators for Leeds, Glasgow and Rotterdam case studies.

Case studies/Indicators	Leeds	Glasgow	Rotterdam
Population^a	784,800	621,000	634,660
Office vacancy rate^b	4.1%	8.3%	18.6%
Residential market^c	Requirement of 4600 units a year	Requirement of units additional to a 5500 backlog	Requirement of 4500 units a year
Implemented units	715/1198^d	133/360^d	1210^e
Space Standards compliance^f	43%	96%	N/A but believed 100% as this is a building code issue in the Netherlands
Access to amenity space^f	1%	40%	-

^a Data for 2017 from NOMIS (2020) for Leeds and Glasgow, and CBS (2020b) for Rotterdam.

^b Data for Leeds from Bradley-Mason (2017), for Glasgow from JLL (2017) and for Rotterdam from Knight Frank (2017).

^c Data for Leeds from Leeds City Council (2016) for Glasgow from Glasgow City Council (2017a) and for Rotterdam from De Nederlandsche Bank (2017) where it is stated that 36,000 added units are expected by 2025.

^d Implemented units (2013–2017) over cases visited. Data collected by authors through site visits in 2018.

^e Created units from the conversion of vacant buildings (including but not just offices) (2012–2015) (CBS 2016).

^f From desk-research on a sample of cases and included analysing all relevant plans and documents on the relevant LPA planning database. From the plans it was possible to note or calculate the space standards of each unit and whether there was any provision of private or communal amenity space (such as balconies or roof terraces).

Leeds

Leeds is a university town and commercial hub for the broader region, also known as one of the ‘Big 6’ cities in terms of its office market, one of the largest in the country outside London (JLL 2015). The overall office vacancy rate in Leeds in 2017 was 4.1%, with grade A office space in greater demand with a vacancy rate of 1.8% (Bradley-Mason 2017).

The management of office-to-residential conversions in Leeds illustrates a case of a thin governance space, where planners depleted of planning powers were relegated to the back seat, as passive observers of market driven changes. Accessing the number of units coming forward through PD, their location, typology and quality, we compare how these matched, or not, policy aims and objectives.

In Leeds planners could only observe the output of the market driven interests in shaping office-to-housing conversion. Between May 2013 and May 2017, prior approvals (the term used for the deregulated planning process for office-to-residential conversion) had led to the delivery of 731 housing units, with a further 467 approved but not yet started.¹ Planners considered these numbers low given the municipality’s overall housing requirement of 4600 dwellings a year. Conversion of redundant office buildings in areas of low demand, was welcomed in policy, considering there were newer office stock available in the city. Yet planners worried about conversions of office buildings in the middle of still functioning industrial estates on the periphery of the city, which they could not control. Of all implemented units in Leeds, 18% were in industrial estates in the outskirts, a real reason for concern considering the lack of residential amenities. As a planner noted, ‘I think the government really has to think carefully about that kind of thing because we have no way of controlling that and yet, late at night, you wouldn’t be wanting to walk down there, … you’re cheek by jowl with car repairs, scrap yards’. More generally, planners expressed concern about whether the neighbouring uses negatively impacted residents’ quality of life, their access to green space and questioned whether a sustainable community was being built.

In terms of typology of units coming through the PD route in Leeds, 70% were studio or one-bedroom apartments. In the city centre, many units were used for student accommodation and short-term rental apartments, rather than delivering family housing. Such provision did not match the housing demand profile in the city, particularly in need of affordable family housing. There were also concerns about housing quality. Leeds can be perceived as a tale of two cities. On the one hand, in the city centre the quality is generally high, in terms of unit sizes, layout, finishes and proximity to amenities. On the other hand, in terms of suggested national space standards (which are applied through local plan policy and planning permission and have not been applied for deregulated conversions), only 41% of the new housing units created through PD in Leeds would meet these standards. Many were significantly smaller: ‘studio apartments’ of about 20m² each were not uncommon, without compensating through amenity space (such as a balcony, roof terrace or communal spaces in the building). Planners described internal layouts as very poor and with limited daylight which we confirmed with our site visits and floorplan analysis.

Many of these concerns are common across many different local authorities in England (Clifford et al. 2018, 2019; Ferm et al. 2021). They suggest that the deregulation of office-to-residential conversions results in market adhocracy, showing significant limitations in terms of delivering policy objectives, as defined both centrally and locally. Next, we explore the case of the regulation of office-to-residential conversion in Glasgow, which we define as a hard governance space.

Glasgow

Glasgow is the largest city in Scotland and the fastest growing major city economy in the UK after London (Scottish Business News 2019). Glasgow’s Housing Strategy predicts a population growth, which means additional units are needed adding to its current housing backlog of over 5500 units (Glasgow City Council 2017a, 2017b). The office vacancy rate in Glasgow is 8.3%, with grade A office space in short supply and a vacancy rate of 1.7%, as much of the vacancy comprises older and poorer quality stock (JLL 2017).

The management of office-to-residential conversions in Glasgow illustrates a case of a hard governance space, a space where planners retain planning control over office-to-residential conversions. Local planners, politicians and civic society stakeholders interviewed were aware of the planning deregulation introduced in England and were universally supportive of the fact the Scottish government had not adopted the same approach.

In Glasgow there had been 77 full planning application, corresponding to the creation of a total of 564 housing units for change of use over the 2013–2017 period analysed. Without the introduction of PD, the number of office-to-residential conversions in Glasgow still increased in 2013–2017 when compared with 2009–2013 where there were 40 full planning applications in a total of 155 units. This suggests that even without PD, there may have been an increase in conversions seen in England anyway, as the economy recovered after the financial crash.

Interviewees expressed concerns over a loss of occupied office space, and a lack of capacity for a nuanced assessment over the location of the conversions if PD were introduced. As a planner questioned, ‘just exactly what kind of amenity, what kind of living environment are these people going to have, even from a security point of view?’. A councillor added, ‘I think when it comes to extension of PD rights, that they’re always trying to make things faster, no matter what kind of quality you think you’re adding to the process’.

From the 50 schemes visited in Glasgow, we found a higher quality of conversions under full planning permission compared with the 414 PD conversions we visited across our five English local authorities (including those in Leeds). Additionally, from the five examples of conversions in Glasgow we explored in detail, the results show that 96% of all units converted (135 out of 140 units) would comply with the Nationally Described Space Standards, and 40% of housing units had access to amenity space. In terms of typologies, the mix included larger flats, 77% with two or more bedrooms. In summary there was a lower number of conversions seen in Glasgow than Leeds, but those delivered were of higher quality, included larger typologies and none were in unsuitable locations for residential use. As conversions in Glasgow must go through full planning permission, planners have a degree of control over location, typology and quality of the units, which is absent in the context of the thin governance space utilized in England.

Rotterdam

Rotterdam is the second largest city in the Netherlands and is known for having one of the highest office vacancy rates in Europe. In 2017, the office space vacancy rate in the city had fallen to 18.6% following a peak vacancy rate of more than 20% in 2014 (Knight Frank 2017). Most of Rotterdam’s office space comprises low-profile aging buildings (CBRE 2012). Given the high vacancy rate, the issue of office conversion to other uses (office ‘transformation’ as it is referred in Dutch) has gained some attention in the Netherlands generally and in Rotterdam specifically.

The management of office-to-residential conversions in Rotterdam illustrates a case of a soft governance space, which includes planners adopting a network approach to steering the market. According to our interviewees, despite the national level work, up to January 2014 there had been little local urgency around vacant office transformation. This changed following the 2014 local elections, when the municipality, without changing regulations (in planning terms, the local plan has apparently supported mixed use for parts of the city centre since 1988) or deregulating, instead appointed an official to lead on office transformation while involving market players. The appointed Alderman, an elected member of the city council responsible for planning, would proactively approach owners of vacant buildings to discuss their potential for conversion. The official also acted as a single point of contact to help developers navigate through the processes and regulations required to deliver transformation. A range of advice and publicity materials addressing the issue were produced, including a website (Gemeente Rotterdam 2017a).

By January 2015, a total 56,000m² vacant office space had been converted to other uses, 81,000m² by January 2016, and 116,000m² by January 2017 (Gemeente Rotterdam 2017b), suggesting the effectiveness of the soft governance space. Initial conversions were predominantly for medical use and schools (the target of conversions were to any productive use, not just housing) but increasingly the focus became conversion to residential use, with 40,000m² converted into 600 apartments in 2016–2017. It should be noted that the conversion figures include office buildings demolished and replaced by buildings with alternative use, believed to be about a third of the total figures (the rest being change of use of the vacant buildings). Figures from the Central Statistics Bureau show 1210 residential units created from the conversion of vacant buildings (including but not just offices) in Rotterdam between 2012 and 2015 (CBS 2016), which is higher than Leeds or Glasgow (see Table 2).

The steering role of the municipality in relation to change of use was apparent. The combination of planning, environmental and building control regulations and permits continued to allow the council to have a strong influence on location and quality issues. Residential one-bed apartments had to be at least 50m², and if less than 60m² the developer had to provide justifications for smaller floorplates and compensate through additional access to amenity space. There is also close attention paid to sustainability issues and ensuring the building will be fit for occupation for at least 30 years.

Interviewees in Rotterdam suggested that planning regulations were not the only barrier preventing vacant office space conversions to residential use. There were cases of building owners having unrealistic expectations about the potential to attract tenants (‘hope value’), absentee landlords (particularly overseas investors) who were hard to contact and discuss conversions with, and a conservative approach from developers who had concentrated on new build in the past rather than conversion. The latter continued to repeat what they had done before, rather than trying new approaches. These cases were perceived as important as building control and planning regulation in explaining why redundant office buildings could remain vacant so long. The soft governance space where the management of office-to-residential conversions was now taking place, involving a clear steering role of the municipality within the regulatory system, emerged as an alternative space for removing perceived and real barriers to office-to-residential conversions.

National differences and local sensitivities

Whilst across the whole plethora of planning reforms there are variegated approaches apparent even within single national contexts, in relation to the regulation of office-to-residential conversions, we identify three prototype governance spaces from our three case studies. The thin governance space used in England, includes a nationally defined ‘one size fits all’ strategy of deregulation, involving the transference of power from individual LPAs to the initiative of individual market actors. Under PD, English LPAs with reduced control over the principle, design and location of conversions, see their previous plan-making, policy-making and decision-making powers curtailed. In stark contrast to England’s thin governance space, Scotland still operates within a hard governance space in the management of office-to-residential conversions. This involves maintaining the inherited British tradition of discretionary planning where market initiative is regulated by LPAs through their planning and place-making powers. In the Netherlands there has been an emphasis on soft governance spaces. Drawing on planners’ local knowledge and planning powers, LPAs are responsible for producing a spatial vision including which office buildings could be converted and which could be retained as offices. Within this regulatory framework LPAs then steer the market by proactively identifying and encouraging developers to consider conversions. Comparing the management of office-to-residential conversions in the three countries, this study suggests that soft governance spaces offer a real alternative to both hard and thin spaces, one that retains planning vision and control over place and place-making, mobilizing the market to achieve both property market and planning objectives in the spirit of good governance.

Implications from these distinct office-to-residential conversion governance spaces can be assessed in terms of the quantity and quality of the residential units delivered, with government data suggesting distinct numbers and trends between the three countries. In England and the Netherlands numbers are high and increasing, with the Netherlands delivering higher numbers once adjustment is made for population (Table 1). This suggests that the Dutch soft governance space carries greater potential for delivering both higher numbers and better quality in planning terms than the English thin governance space. In England, deregulation of change of use undermines LPAs’ place-making powers, compromising delivery of the desired type, size and location, and has delivered less units per capita that alternative governance spaces. In Scotland, numbers are low and stable suggesting that a hard governance space might not produce a high level of reuse of vacant office space. The low conversions in Scotland are nevertheless accompanied by maintaining a level of appropriate control related to amenity and local service provision, and a wide variety of potential barriers and inertias could help further explaining the low conversions in Scotland. Overall, data suggest that it is possible to deliver high and growing number of units through conversion while retaining planning control.

At the city level, our site visits and analysis showed a distinct bleaker picture in Leeds compared to Glasgow and Rotterdam. In Leeds there were some good schemes in the city centre, mostly targeting students and the private rented sector. However, what dominated were schemes with extremely small unit sizes, strange internal layouts, poor natural light, limited access to amenity space and examples of inadequate locations such as industrial parks. This was different to Glasgow, where the lower rate of conversions had nevertheless produced better-quality housing, without unplanned negative externalities for local infrastructure. In Rotterdam there had been a higher rate of conversion than Leeds and Glasgow. Local state proactivity appeared to have delivered the right sort of conversion, of the expected type, size and quality and in the places decided by the local authority.

Arguably, England’s permitted development rationale is one of sheer numbers and not one of quality or planned delivery and will leave an unplanned legacy for the decades to come. A centrally imposed deregulation of planning in England undermines its discretionary potential, offering a rigid (de)regulatory approach. Results from our study suggest that permitted development inadequately deals with the different built environment issues experienced by distinct local real estate markets. It seems problematic to subject high demand, expensive residential markets such as those in London and lower performing markets, such as Leeds, to the same policy – one that essentially deregulates planning and leaves matters to the market. Likewise, it is hard to make a case for this national policy to apply similarly to places with high and low office vacancy rates. This ‘one size fits all’ national policy, appealing as it may sound to central government, is hardly understandable in the context of such diverse office and residential markets across the country.

Conclusions

In this paper, we have advanced the concepts of hard, soft, and thin governance spaces, drawing on Haughton and Allmendinger (2008) and Peck (2001), as a heuristic device to further our understanding of the governance of offices-to-residential conversions in England, Scotland and the Netherlands. Whilst urban governance involves complex, and even sometimes contradictory, processes and scalar differentiations, we argue that these three prototype positions help to identify both the discursive conditions of policy process, and the material conditions of policy outcome. Through this heuristic device, we explored how traditional governance binaries including state/market, hierarchies/networks, regulation/deregulation, bureaucracy/market, were mobilized and legitimized, and ultimately how they clustered and appear to stabilize in distinct governance spaces.

The hard, soft, and thin governance spaces, at a high-level range from more to less state, but each space has their distinct formulation. Typical hard and soft spaces both reflect local and embedded responses, yet with soft spaces enabling greater proactivity and local discretion to cater for the individual needs of places and players. In contrast to hard spaces, soft and thin spaces are more market-leaning, and offer alternative spaces to the hierarchies and bureaucracies that populate harder spaces, yet they differ from each other in the way they consider planning regulation. Soft spaces enabled soft planning tools to be negotiated at the local level, whereas thin spaces filter into practice their market-leaning ideology by removing local discretion. This suggests that similar ideological backgrounds can have different expressions on the ground based on national ideology and local planning culture(s), providing further evidence of the different trajectories that neoliberalism can take in different places leading to its variegated forms (e.g. Peck 2001; 2013). Within the realm of planning reforms that ‘purge the system of obstacles to the functioning of “free markets”’ (Peck 2001, 445), there are variegated processes and outputs in the management of office-to-residential conversions.

The policy informed scepticism about the value of planning regulation (Lord and Tewdwr-Jones 2014) currently experienced in England is not new. As Rhodes (1996, 654) noted in Britain ‘the ideological preference for less government was stated loudly and often’. It contrasts with Scotland’s more traditional approach despite its ‘open for business’ neoliberal reform (Inch 2018) and with places such as the Netherlands with a long identified strong planning culture (Faludi 2005). The network approach seeking consensus building, more typical of the Netherlands than England and Scotland, contrasts with anti-regulation bias approach to planning reform in England. As Davoudi, Galland, and Stead (2020) have shown, change is legitimized and embedded through strategies of rhetoric persuasion. In the case of the management of office-to-residential conversions in England, ideology is revealed in government documents emphasizing the successful delivery of ‘thousands more homes’ and ‘breathing new life into neighbourhoods’ (Gov.uk 2015, online) through the deregulation of planning, and making it apparently self-evident that planning was the sole barrier to building conversion. This is a worrying trend because of the negative impacts planning deregulation can have in the built environment, and ultimately in the way people live.

The management of office-to-residential conversions in England, is currently unfolding in a thin governance space, which includes the hollowing out of local state knowledge and powers through a nationally imposed deregulation and the transfer of powers to market initiative. On the other hand, the soft governance space formulated in the Netherlands enables a form of local empowerment, relying on local knowledge and local planning control. Both soft and thin governance spaces reveal a strong market ideology, yet the thin space found in England shows a marked anti-regulation bias, depleting planners of most hard and soft powers, and weakening planning and place-making.

Overall, this paper calls for an increased focus on the continuing emerging forms of governance spaces, their merits and limitations, and the impacts of ideology and planning culture(s) in the formulation of governance spaces, including planning policy and (de)regulation. In reflecting on the governance of land-use changes, our findings suggest that to ensure good quality housing and place-making, solutions should rely on planning and not just on markets. It also suggests that it is possible to shape markets while preserving local state land-use planning and place-making powers. Soft governance spaces not only seem to provide the more adequate response to office-to-residential conversions in planning terms but also to other changes of use with a potential impact at the neighbourhood scale. In the near future we may see the issue of the management of office conversions rise up the urban policy agenda if the ongoing trend towards remote working, drastically accentuated by the COVID-19 pandemic, is to continue.

Disclosure statement

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

Additional information

Funding

This work was supported by Royal Institution of Chartered Surveyors.

Notes

1 Different from other LPAs, data collected and provided by Leeds, covered schemes of five or more units only.