Housing Stock Transfer in Glasgow—the First Five Years: A Study of Policy Implementation

Abstract

In 2003 the City of Glasgow saw the largest housing stock transfer in the UK, involving around 80 000 dwellings. Since then, the implementation of the stock transfer policy has been heavily criticised. This paper uses a framework developed from implementation studies to analyse why this policy has been susceptible to difficulties and to reflect upon the important elements of a revised analytical framework. The paper finds that the study of policy implementation must contain an appreciation of the effects of having multiple policy objectives, multiple layers of governance and multiple actors involved in policy delivery. Additional elements of a policy implementation framework are: the specification of the stages of implementation; consideration of interactions between policy objectives; the need for government to oversee and ensure the effective management of policy networks; and finally, consideration of the effects of competing political interests and perspectives.

Keywords:

• Housing policy
• housing stock transfer
• policy implementation
• Glasgow

Introduction

It is five years since public sector tenants in Glasgow saw their housing transferred from Glasgow City Council (GCC) to the Glasgow Housing Association (GHA) in March 2003, following a majority vote in favour of the move in a ballot of tenants in April 2002. The policy of stock transfer for Glasgow was meant to be the flagship of the Scottish Executive's new housing programme of community ownership (see Kintrea, 2005; Scottish Executive, 2000), but ownership of Glasgow's ex-council housing at a community level has taken longer to achieve than many wanted. This has led many observers and commentators to declare the policy a ‘failure’, such as the following:

An inheritance of four years of political bitterness resulting from the failure to split the organisation into more than 60 local associations. (Gerry Braiden, journalist, The Herald, 11 June 2007)

It is with some concern that I see the situation with Glasgow's stock transfer drifting all over the place. (Alastair McGregor, former GHA Board member, The Herald, 21 September 2007)

Given that council housing stock transfer has been taking place in the UK for around 20 years, commencing under the Conservative Governments of the 1980s and 1990s and continuing under New Labour, it could be asked why Glasgow's transfer has been so problematic.1 Malpass & Mullins (2002) have shown how stock transfer policy in the UK has gone through different phases, from locally led initiatives (early 1980s), to a centrally-controlled programme of local initiatives (late 1980s to early 1990s), to centrally promoted policy to overcome local resistance to stock transfers and to resolve financial barriers (late 1990s onwards). Interest in the Glasgow housing stock transfer stems partly from the fact that it is the product of a centrally-directed, and locally unpopular (with local authority councillors and some tenants) solution to the problem of investing sufficiently in poor quality housing stock (see Scottish Office, 1999); but also that the implementation of the policy has been left in the hands of a fragmented system of local housing governance which has been further complicated by the policy prescription itself, as we shall see. Thus, the case shows how the ‘fragmentation of local governance’ is not only an objective of policy, as argued by Malpass & Mullins (2002), but also a problem for policy. A further reason to look at this policy in practice is that most of the literature on housing stock transfers has looked at the overall, housing-system-level effects of the various programmes over time (e.g. Kleinman, 1993; Mullins et al., 1995; Taylor, 1998), rather than at the implementation of stock transfer ‘on the ground’.

In what follows, the paper first discusses how the Glasgow transfer is unique compared to its predecessors. It goes on to use a framework developed from implementation studies in public policy to review various elements of the policy and its implementation to identify weaknesses and complications that have predisposed the policy to implementation difficulties. The paper concludes by considering what the Glasgow case has added to our understanding of the important elements of policy implementation to be studied in future.

Glasgow's Unique Stock Transfer

Whilst the transfer of council housing to other landlords is no longer a new phenomenon, having taken place across the UK continuously since 1988 (see Malpass & Mullins, 2002; Mullins et al., 1993; Taylor, 2000), Glasgow's transfer has been unique in at least four respects. First, the size of transfer: GCC owned over 80 000 dwellings at the time of transfer, compared with a typical transfer among other local authorities of between 5000–10 000 units each (see Wilcox, 2001), with notable exceptions being Coventry (21 000 units), Wakefield (31 000) and Sunderland (36 000).

Second, allied to both the size of the transfer and the poor condition of the housing stock involved, Glasgow's transfer involves much larger sums of money than previous events: housing debt of circa £900 million, plus £4 billion of public-and-private housing investment over 30 years. As Gibb (2003) outlines, this involves a great deal of risk surrounding the following: future moves in interest rates; policy reforms to the Housing Benefit system which underpins rental payments; and the need to improve rental income in the context of falling social housing demand (Gibb, 2000).

Third, the Glasgow stock transfer is to be a two-stage affair, in comparison to other single-event transfers. Scottish politicians presented the policy as consistent with Glasgow's heritage of ‘successful’ community-based housing association (CBHA). The belief in the CBHA model is expressed by two of the individuals involved in the original policy formulation:

[it was] about giving tenants real choice of localised, neighbourhood-based community ownership because we believed it was real. (Scottish Executive Official)

So for us it was always about, this is a totally different model of management and ownership … This was about what works, small is beautiful. (Labour MSP)

Local ownership was not to be achieved through a series of sales to individual housing associations over a period of time, but rather through a single transfer to the GHA, and thereafter second stage transfers (SST) to local communities. A network of local housing organisations (LHOs) was to be created and supported by GHA to manage the housing stock in the interim before taking ownership.

How this two-stage process would work in practice appears to have been difficult to specify in detail. The Glasgow Housing Partnership Steering Group, set up by the Scottish Executive and Glasgow City Council in 1999 (also with local housing association representation) published a Framework Agreement in 2000 (GHPSG, 2000) which contained some key statements about SST:

From the outset, tenants will be offered advice and assistance about taking ownership at a more local level in future. There will be different solutions for different parts of the city…

Transfer to local ownership will be by means of a local ballot which can take place as soon as it is legally and financially possible after a city-wide transfer has been effected

At the end of a 10-year period after city wide transfer, those tenants who have not chosen local ownership will be invited to participate in a survey confirming whether they wish to remain within the GHA of consider transferring to more local ownership arrangements.

Thus, although not specific about the timetable for SST, the Framework Agreement implies that local transfers of ownership would happen during the early years after whole stock transfer and certainly much of it would happen within the first 10 years, with remaining GHA tenants being offered the chance to express their wishes at the end of this ten year period. However, the legal notice about city-wide stock transfer that went to tenants in 2001 was less committal about the timing but more specific about the conditions for SST:

Such SSTs will only happen if the financial and other arrangements for such a transfer continue to ensure that all the commitments given in this document are guaranteed to be delivered to those tenants transferring and to those who remain with the GHA. In addition, SST will only happen if it has a financially neutral effect on the GHA and its operations. (GCC, 2001)

Here it can be seen that SST could be proscribed by the requirement for ‘financial neutrality’, the concern being that the ‘best’ stock could be transferred to local organisations, whilst the remainder becomes unviable as a collective stock holding under GHA. The uniqueness of the Glasgow case is therefore also the fact of its complexity, as noted by Gibb (2003).

Finally, the Glasgow stock transfer is unique in its political importance. Locally, the transfer has been politically controversial with a vocal opposition in the form of a ‘Vote No’ campaign at the time and an ongoing Defend Council Housing campaign (see Defend Council Housing, 2003). In Glasgow, “the dominant housing culture was municipal” (Daly et al., 2005, p. 329), and “it is difficult to underestimate the role that council housing plays in the fabric of Glasgow life, in its recent social history and… in its politics” (Mooney & Poole, 2005, p. 30). The performance of GHA continues to be controversial with the city council's credibility on the line if things go wrong, since the transfer was seen “as something of a watershed” (Mooney & Poole, 2005, p.30).

Nationally, the Glasgow stock transfer is also politically crucial. Kintrea's (2005) analysis of housing policy development in the early years of the Scottish Parliament (see in particular Scottish Office, 1999 and Scottish Executive, 2000) led him to the conclusion that there

is one dominant method designed to simultaneously achieve all of these [high level] goals [of social justice, social cohesion, economic competitiveness and empowerment]—stock transfer or ‘community ownership’. (p. 191)

He describes the Glasgow case as “absolutely pertinent” and of “political importance”, more so since the adoption of the Scottish Housing Quality Standard (to be achieved by all social housing by 2015) “has renenergised the stock transfer debate, with more councils now accepting that this is the route they will follow” (p. 195). As a result, progress in Glasgow has become a totemic political issue:

Succeeding to transfer Glasgow's council housing was massively important both in terms of securing a sustainable future for Scotland's largest council stock, and for constructing a symbol of success for ‘community ownership’.

In the light of this level of importance, it is a very significant development that many commentators are now calling the transfer process in Glasgow a failure.

The Research

The investigation here of the formulation and implementation of the Glasgow housing stock transfer consisted of three elements. First, an analysis was made of policy documents and academic articles on the issue, over the period 1999 to 2007. Second, media commentaries about Glasgow's housing over the same period were reviewed. Many of the protagonists involved in the delivery or receipt of the policy voiced their opinions in the national press. Third, 20 policy makers and practitioners at national and city levels were interviewed. These were people who had been closely involved in the development and implementation of the stock transfer policy at various times over the period of study,2 and included staff, or former staff, from the Scottish Executive, Communities Scotland, Glasgow City Council, Glasgow Housing Association and other Registered Social Landlords (notably Community Based Housing Associations) in the city, as well as opponents involved in campaigning against the transfer. Interviewees were either identified and selected by ourselves from the documents studied, or were named to us by other key participants. The interviews were recorded and transcribed in full for analysis. Anonymity was preserved here by only giving individuals organisational affiliations rather than their job titles. The majority of the interviews were conducted in the period August to October 2006, with one interview in 2007.

A Framework for Studying Implementation Difficulties

It is clear that things have not run smoothly in the process of Glasgow's two-stage housing stock transfer, whether or not one concludes that it has been a ‘failure’. The rest of this paper identifies the causes of difficulties—within the area of policy formation and in practice—using a framework developed from implementation studies, a field which has recently been ‘revived’ (see Schofield & Sausman, 2004).

Overviews of implementation studies (see for example Barrett, 2004) identify three types of approach. The top-down approach seeks “to suggest ways of enhancing the likelihood of obtaining compliance” (Barrett, 2004, p. 254) with policy objectives made at the top. This is taken to be a rational and normative view of policy, seeing the goals of the ‘top’ as legitimate and divergence as undemocratic (Ham & Hill, 1993). Jordan (1995) points out that “more sophisticated top down models (e.g. Sabatier & Mazmanian, 1980) take a less ‘mechanical’ view, but while they see implementation as being more ‘political’, they nonetheless retain a ‘top down’ focus” (p. 8). Bottom-up approaches focus more on the behaviours of implementing agencies who have their own motivations, structural constraints and who exercise discretion given that they face contradictory demands and partially formulated initiatives (Elmore, 1979–80; cited in Jordan, 1995). Third, the ‘bargaining and negotiation models’ approach (Barrett & Fudge, 1981) examines how implementation is a product of both the top and the bottom in reciprocal power relations. The top may structure the implementation process and constrain the power of those below, but “lower level actors also take decisions which effectively limit hierarchical influence, pre-empt top decision-making, or alter policies” (p. 25).

In the analysis in this paper of developments in Glasgow, elements of all three approaches are combined. Figure 1 identifies the key factors sought to consider in analysing the implementation of stock transfer, or the government's community ownership policy, in Glasgow. This draws heavily upon the frameworks proposed by Van Meter & Van Horn (1975), Hogwood & Gunn (1984) and Mazmanian & Sabatier (1983) (and Sabatier, 1986). Van Meter & Van Horn (1975) focused mostly on policy standards and the characteristics and ‘disposition’ of the implementers, which Marsh & Walker (2006) in another study of housing policy implementation divided into the implementers' understanding of the policy and, second, their response to it. Hogwood & Gunn had more to say about policy theory and on the degree of dependency and consensus among agencies (see Hill & Hupe, 2002 for a review). Mazmanian & Sabatier (1983) identified all the elements mentioned in Figure 1 concerning the Implementation Agency, but in particular the need to ‘keep discretion within bounds’ and to look at efforts by the state to “structure implementation” (Sabatier & Mazmanian, 1980, p. 544). Nearly all the available frameworks mention either the effects upon policy implementation of social and political upheaval, or simply the effects of the operating environment. This factor has been included in part IV of Figure 1, but a second element has also been added here, namely the way in which policy design and delivery is affected both by national political factors and by local political influences.

Figure 1 Framework for analysis of policy implementation

Three things have also been added to the consideration of the policy domain: whether or not the parameters for preferred outcomes are set out in policy (or are all outcomes which are compatible with the broad political aim of policy acceptable?); how well considered are the practicalities and logic of implementation (although Hogwood & Gunn recommended that policy makers specify the sequencing of tasks); and how do the different objectives of policy interact over time? These things are particularly important where there are several major policy objectives, and amongst these is an objective of institutional reorganisation which is often the most difficult to achieve.

The remaining domain of the framework is that of ‘inter-organisational relations’, also part of Van Meter & Van Horn's model, but there it mostly relates to relations between levels of government involved in implementation, rather than something broader. The first element of this domain refers to support for the policy from non-governmental members of the policy network, and is included in Sabatier's (1986) conditions for effective implementation. Mullins & Rhodes (2007) remind us that the shape and structure of networks increasingly affect policy making and implementation, especially as a result of institutional changes in housing systems. Thus, two other elements have been added here to reflect the fact that stock transfer policy as enacted in Glasgow relies for its implementation upon a complicated, local network of housing agencies, where trust, oversight and management may all play a part. A network of inter-dependent actors—rendered more complicated by the creation of new agencies as part of the stock transfer policy (e.g. the Glasgow Housing Association and its constituent Local Housing Organisations)—who have different perceptions of the policy problem, different values and different desired solutions (cf. Termeer & Koppenjan, 1997) are nonetheless required to work together in a local context. What is more, some of the actors have developed stronger expectations as a result of being involved in the policy design stages (see Palumbo & Calista, 1990). In this situation, central government has a choice with regard to the role it plays in the network, namely to try to manage in a traditional sense through central oversight, or to become a network member and manage through a form of governance which induces actors to co-operate (Stoker, 1991), and which steers the network towards joint problem solving and further policy development (Kickert & Koppenjan, 1997), influencing inter-organisational decision making (Klijn & Teisman, 1997). In neither case would central government be merely a passive observer of policy implementation.

In what follows, the paper considers each of the four main areas in turn in relation to the Glasgow housing stock transfer, to see how each might have contributed to implementation difficulties.

Policy Itself

First, the paper considers the nature of the policy objectives and the degree of consensus over those, as well as the key assumptions underlying the policy. Then, there is a look at how well specified the policy was in term of the end state to be attained, the logic and logistics of implementation, and the attention given to issues of interactions between policy objectives.

Policy Objectives

It is important to recognise that housing stock transfer in Glasgow had multiple objectives, complicating any judgement about success. For Glasgow City Council, the two key objectives were debt removal and stock improvement, as these two officials testify:

I mean, community ownership from the council perspective, community ownership was always incidental. The main, I mean, the main driver of the Council was getting rid of the debt. (GCC Official)

The council's driver was investment in the housing stock, the Executive's was undoubtedly community ownership which is why we ended up…with a two-stage stock transfer… it was the only way of putting two objectives together and getting a win-win. (GCC Official)

Likewise, it is possible to identify two key objectives for the Scottish Executive: community ownership or community empowerment; and the creation of an efficient and effective social housing system:

The original policy for stock transfer in Glasgow [was] very much driven by community ownership as such, very much driven by the view that in order to make communities sustainable, local people should be involved in decision-making, shaping the thing, owning the regeneration, as opposed to just owning the houses. (Scottish Executive Official)

We were strongly committed to using the stock transfer to create a better housing system for the non-market sector of the city as a whole. (Scottish Executive Official)

In fact, the pre-transfer ‘Framework’ document listed six objectives for transfer—the four mentioned above plus two others: to provide opportunities for house purchase; and to achieve excellent standards of design, construction, management and maintenance (GHPSG, 2000, p. 2). Research (Pawson et al., 2008, forthcoming) has since indicated that the City Council's approach, and in particular its advocacy of a single-transfer model, were also informed by the desire to secure continued work and employment for its Direct Labour Organisation (DLO), which was granted a five-year contract to undertake GHA housing repairs.

It could be argued that in its early years the GHA has put more effort and resources into investment in the housing stock and improving its housing services, than into community ownership. This partly reflects the political reality wherein there was pressure on politicians to be seen to be improving the circumstances of Glasgow residents, and pressure on GHA to perform better than GCC as a landlord; but it also reflects the more detailed timescales put alongside the investment objectives in the original tenants promises at transfer. But how the agencies deal with these simultaneous goals also raises issues about the assumptions of causality and progress within the policy itself.

Policy Assumptions

The link between ownership and empowerment

In the words of the original Green Paper, “Community ownership is a way of empowering tenants…” (Scottish Office, 1999), and thereby a contribution to the sustainable communities goal (Kintrea, 2005). But there is no consensus among the protagonists as to whether or not community ownership is necessary for community empowerment. The view that community ownership is not essential can be found in several places:

It's easy to get bogged down with focusing on the negative, you know second stage transfer is not being delivered, but that's not, you know that's not what it's all about … (GHA Official)

You don't have to own a damn thing in order to have some say about the things about it that are important to you…the trick is to find a way to give people some influence over what happens locally without necessarily having to own (GCC Official)

Those holding the first view above would no doubt emphasise the priority that tenant participation has within GHA. Independent researchers have remarked on how much more GHA is tenant-driven than its predecessor, particularly in relation to policy making (Pawson et al., 2008, forthcoming), whilst government inspectors have said that GHA has made tenant participation one of its strengths (Communities Scotland, 2007, p. 37).

Some participants question what ‘empowerment’, beyond participation, is supposed to deliver for communities:

Nobody could say, what is this thing right? Community ownership it kind of sounds good so, em, is it about empowering, is it about empowerment, is it about control, is it a substitute for people who can't own personal assets; they can own community assets? Are there behaviours that change because that level of control is, that level of actual ownership is in place? (GHA Official)

Other participants believe that community ownership is necessary for empowerment, but not always sufficient.

I don't think tenants can be empowered without ownership. They can be consulted. They can participate. Unless you control the money, control how that money's spent then you don't really, you don't real, you're not empowered as such. (Communities Scotland Official)

I think that a case can be made that community ownership can often be a major stimulus of empowerment. But it is neither necessary nor sufficient. (Scottish Executive Official)

This variety of opinion on the necessity of community ownership, alongside policy vagueness about the links between ownership and empowerment and the expected benefits from them, feeds in to a lack of consistent commitment to this policy goal.

The demand for SST

As well as presuming the link between community ownership and empowerment—rather than demonstrating it—the original policy also assumed that the vast majority of communities or LHOs will want to achieve SST. Based on her work with LHOs, McKee (2007) appears to concur: “The aspirations of LHO actors are however clear: SST is regarded as the only means by which they can realise their ambitions for local autonomy” (p.329).

However, with over 60 LHOs created, of different sizes and in different community contexts, with housing stock of different quality and value, there was always going to be scope for the aspiration of community ownership to vary, both among LHO committees and among the wider population of tenants. Surveys of various types during 2007 (reported in Communities Scotland, 2007) found that most tenants had not heard of SST and over half (57 per cent) of tenants were not interested in SST for their areas. Whilst all the LHOs linked to existing community based housing associations said SST was important to them, just over half (56 per cent) of the standalone, forum-based LHOs said SST was no more important to them than managing the stock and influencing services in other ways. Only half (52 per cent) of registered tenant representative groups saw SST as important for their group. It is clear that there is scope for reality and experience to belie the presumption of uniform or widespread demand for SST.

Policy Specification

Outcome parameters

Given the desire to devise a better overall set of arrangements or ‘housing system’ for the city, it is surprising that policy makers did not give any indication as to what the end-state should look like. This is important since the existence of 60 LHOs is now a problem for SST in organisational and financial terms. Further, despite the early Framework Document noting that the local outcome could vary, critics have been unwilling to accept that independent ownership should not be achieved by all LHOs.

Policy makers appear to have been caught between competing objectives of housing system efficiency and community empowerment. A variety of local residents groups across the city were invited to say whether they had an interest in becoming an LHO, either standalone or linked to an existing housing association. This process produced a total of 78 LHOs, now reduced by mergers to around 60, and is seen by some to have had great virtues in being a ‘bottom-up organic process’ which boosted tenant involvement:

I mean obviously, the transfer, em, in terms of how it delivers at a local level has set up this LHO network which is really significant. I mean, none of the other transfers have, have done that. You know, there's over five hundred, em, individual tenants involved in this, engaging with this. (Communities Scotland Official)

Others identify the network itself as a problem:

The original model never, never imagined that there would be 60 new LHOs formed as community owners. We assumed, on the advice of consultants, that there would be some 15 to 25 new organisations created on the traditional lines. (Scottish Executive Official)

The responsibility for this situation is laid at the door of the implementing agencies:

The existence of the 60 represents a failure within GHA the first year after the ballot and of Communities Scotland in registering all these bodies. Ministers never wanted all of this and encouraged a managing down of the numbers. (Scottish Executive Official)

However, it could also be argued that the ‘failure’ lay with the politicians who did not give a clear indication of how far they wished to go with their favoured ‘small is beautiful’ scenario. Further, there was a lack of oversight and assessment of this key stage in policy development.

The logistics of implementation

“The key partners in the transfer—GCC, the Scottish Executive, and GHA – did not have a route map to follow…” (Communities Scotland, 2007, p. 41). A further deficiency was the absence of sufficient consideration of the legal, technical, financial, logistical and organisational issues involved in the disaggregation of the GHA stock. To be fair, a great deal of policy effort was put into getting the difficult first stage transfer right, but the subsequent challenge was not adequately addressed at the time of policy formulation, even though there was awareness of the impending difficulties:

What the feasibility study showed was that if you wanted to split the stock up for reasons because of the high cost of improvements, the high management costs of some of the stock, the high propensity of high rise blocks for example… we've got these pooled rents that are sharing the cost of this. If you break the stock up you lose the integrity of that and some of the stock becomes unviable because it won't meet funders' requirements. (GCC Official)

You can't take the GHA business plan and infrastructure and just (dis)aggregate that into 63 and still have everything the same and the same things happening and proportionately, you know, the same number, it doesn't work. (GHA Official)

Thus, four to five years after transfer Communities Scotland (2007) had to recognise the problems stemming from the creation of a network of 60 LHOs, and from GHA's 30-year business plan, which appears to lack sufficient flexibility to fund the multiple transaction costs of further transfers or to vary the expected price to be received from LHOs for the stock:

… neither the financial terms nor the operational detail of how to achieve SST had been worked out at the time of transfer. (p. 42)

… transfers to 63 LHOs are not possible within the existing financial envelope; … the principles of ‘financial neutrality’ limit GHA's ability to consider transfers with proposals that vary significantly from GHA's model. (p. 46)

Capacity building versus SST sooner or later?

There are a number of other ways in which the stock transfer policy lacked sufficient consideration of the logic of the implementation process. Policy makers did not reconcile the dilemma between moving to SST in the short-term and the need for capacity building among LHO committee members for them to be able to take on the more strategic role and responsibilities of community ownership. Thus, a political supporter of SST can say:

All the evidence is that it takes time to acquire the confidence…that process would allow more sophisticated ways of making decisions to be made because people wouldn't be saying ‘I do want to hold onto my multi’, they would be, ‘what are the other options for me. (Labour MSP)

Paradoxically, the same person was also highly critical of the lack of SST to-date and called it ‘a monumental disaster’, and yet McKee & Cooper (2007) indicate that LHO committees are not prepared yet for the responsibilities to come and find the prospect somewhat stressful. There does not seem to have been a clear process set out by which to develop and gauge the capacity of LHO committees beyond the initial training they received when the LHOs were first set up.

Policy Interactions Over Time

The policy has also lacked a consideration of various interactions between its different elements, and this is especially important over a 10-year period during which the move to SST was expected to happen. First, LHOs would experience service provision by the GHA and this may shift them from a position of disaffection felt previously with GCC. The latest GHA survey shows that approximately three-quarters of tenants are satisfied with GHA and its services; think GHA cares about them; and that GHA listens to them, with all these indicators rising dramatically since transfer (GHA, 2006). Moreover, GHA used a wide range of methods to consult and involve tenants and residents both individually, and through representative groups and intermediary organisations (GHA, 2005). The net effect of these developments upon community empowerment and the desire for SST among residents (as opposed to LHO committees) does not appear to have been considered by policy makers and yet their effects may be considerable given that SST depends upon a ballot of all tenants within each LHO area.

Then there are the effects of the housing investment and regeneration programmes implemented by GHA. Whilst relying upon the model of the CBHA movement as a template for community empowerment, policy makers have departed from what many see as a key element of the success of the CBHA model—control over their own investment programme was crucial to community involvement and empowerment:

… they're losing the prospect of the capital investment works providing the focus and the catalyst for local action. I mean, that's not, ‘cause that's how housing associations were formed… determination in improving the environment and need to have control of the resources at your disposal to give to the committees round the tables… (CBHA Director)

The GHA approach to-date has been closer to GCC's original conception:

So, the Council's proposal was, keep it as an entity for ten years, complete the investment programme, complete the demolitions and then, and run it through local management groups and then, at the end of that period, transfer the ownership. (GCC Official)

The fact that the GHA is making the big decisions about the scale, timing and nature of housing investments and on the demolition of high-rise blocks means that a potential source of community empowerment is foregone.3 McKee identifies the major weakness as:

While individual tenants’ choice is permitted in terms of colours and designs … in terms of the bigger picture local preferences with regards to the style and cost of the modernisation works are being sacrificed for bulk procurement and standardisation. (McKee, 2007, p. 328)

Although McKee may understate the degree of local influence upon investment decisions, once again suggestions can be seen of a conflict between objectives of economy and efficiency on the one hand and empowerment on the other. A further risk in this approach is that with much of the investment programme preceding the SST stage, many tenants may no longer see SST as a key objective as time moves on.

The investment and regeneration programmes will further complicate SST by raising the price to be paid by the receiving landlords for the stock, which will then be in better condition. This could take prices beyond the reach of LHOs without CBHA asset backing, and/or result in disagreements over whether the asking price reflects the sustainability of the stock after refurbishment (where LHOs did not make the investment decisions).

Finally, as Kintrea (2005) points out, preparations for the stock transfer policy involved strengthening regulatory policy over the social rented sector, and curtailing some of the freedoms housing associations enjoyed over housing allocations. The implications of such strong regulation of the RSL sector have not been spelled out, so that many LHO committee members still expect more freedom than they will have in practice after SST.

Implementation Agency

According to theory, effective policy depends upon having an implementation agency that is sufficiently skilled and committed for the task, has adequate resources, and whose room for discretion and autonomy can be curtailed. Each of these presumptions is problematic in the GHA case. Much of the difficulty and the source of much suspicion comes from the fact that the GHA has been multi-tasked, having to dramatically improve housing services to tenants, undertake a large investment programme, and arrange for the onward transfer of the stock to the LHOs. GHA in its early years prioritised the first two objectives over the third and this has led to scepticism and criticism from all sides:

Many within the LHOs, however, believe GHA is not a willing seller and is content to hold what it has. (Gerry Braiden, The Herald, 11 June 2007)

I worry that [the break up of GHA] won't happen now because this beast is in our midst and the bigger it becomes and the more solid it becomes, the more difficult it, when it comes to deal with it because if you split it up. (GCC Official)

The fact that the inspectors (Communities Scotland, 2007, p. 42) gave a positive report on GHA's efforts with regard to SST, as well as concluding that SST was not possible as originally conceived, has led some participants and observers to criticise Communities Scotland as a weak regulator:

Four years later not one house has been transferred … But instead of taking GHA to task and admitting its own role in its failure, the regulatory body is now asking GHA to come up with its own template for the future and, on GHA's say so, implying that second-stage transfer is unworkable. (Anne Johnstone, The Herald, 20 September 2007)

Two fundamental problems with the organisational arrangements have been identified by the critics. First, that of regulatory capture, i.e. that Communities Scotland is too close to GHA:

… the regulator Communities Scotland has been compromised in coming to its judgement about GHA's performance… there is a close relationship between these two organisations that is unhealthy… now would be a good time to transfer the regulatory function of social housing providers to Audit Scotland, which is at truly arm's length from government and its policy-making ambitions. (Douglas Robertson, The Herald, 22 September 2007)

Second, that the relationship between GHA the LHO network, which was not set out clearly in the original policy, is inverse to that desired:

I've a sense of the centre of GHA being completely unaccountable now to the local arms, just because of the structural thing about them and, and the way in which the whole thing delivers and, of course, that wasn't what was intended but the centre of GHA has got bigger and bigger and bigger. (GCC Official)

With regard to the third element of agency, sufficient resources is a key difficulty, whereby “The major stumbling block is the funding gap between the prices GHA needs to transfer at and what the LHOs can afford to pay” (Communities Scotland, 2007, p. 44). This gap has been estimated at anywhere between £85 million and £507 million, with the higher figures pertaining to larger numbers of transfers, so the issue of whether there are 30 or 60 local SSTs matters. Again, Audit Scotland has been called upon to give an independent view.

It seems a partial judgement for Communities Scotland (2007, p. 43) to blame GHA for not having done detailed financial analysis early enough to identify the funding gap. Whilst it might be argued that policy makers could not have known the exact size of the gap, it begs credibility to argue that they should not have anticipated that there might be a funding gap at all, that they would be surprised by its occurrence, and that there would be insufficient flexibility within the GHA business plan and funding arrangements to make any response to it. Indeed, as will be discussed later, it has become evident that some of the parties were aware of the funding gap, but not all.

A similar point can be made about the key requirement of ‘financial neutrality’ for SST, where Communities Scotland (2007) says:

GHA has been criticised for an over-rigid interpretation of the ‘financial neutrality’ element of its commitment for SST. This term was not defined at the time of the original transfer, and it was not until mid-2005 that it was clarified further. (p. 45)

Again, this seems a crucial shortcoming of policy making since SST hinges on the use of this criterion. The paper will return later to the question of the origins of these missing steps of the route map to SST.

Inter-organisational Relations

Inter-organisational relations at a local level have been hostile, with a great deal of mistrust. Glasgow City Council are accused alongside GHA of not being committed themselves to the whole process:

Why have we got into such difficulties? It's because it's not clear, it is not owned, what you have now is the most appalling standoffs between the parties who are all shouting at each other. You can't make it happen that way… there is no ownership of the philosophy [of community ownership]… it's a monumental failure because, in my view, because the City Council stance didn't own the vision. (Labour MSP)

As a result, there are suspicions among the various parties that GCC were aware of many of the difficulties of splitting the stock up at SST, but were not forthcoming with the other parties. Ministers rejected the council's proposal for a Glasgow Housing Trust early on as not providing enough empowerment for tenants, so councillors and officers have not been favourably disposed to a large and independent GHA.

Relationships between the council and the CBHAs were also affected by the stock transfer process. Ministers turned to the CBHA movement for advice and assistance, and CBHA directors were involved in the creation of the original Framework Agreement, the establishment of GHA and the creation of the LHOs:

The other big part, we were involved in setting up local housing organisations which was an unbelievably difficult process … the Council were, officers in particular, annoyed that CBHAs were being brought in. (CBHA Director)

From championing the city's housing association movement in earlier years, GCC officials now saw CBHAs as rapacious housing developers involved in the ‘bottom-up’ process that created the LHO network:

Everybody and anybody was invited to bid for any stock in the city… you got empowered, well funded, well resources organisations with professional staff, competing with tenants associations with no staff, dependent on grant from Communities Scotland… I just thought it was appalling… against everything we were trying to do… It's nothing but self-interest in their approach to this… a lot of the existing housing association directors… are not, their interest is not in the tenants it's about their own organisation or the capacity of their own organisation. It's about the ability to continue to develop in new ways. (GCC Official)

Council officials were now prepared to question the cost-effectiveness of the CBHA model and its contribution to community empowerment, although also to acknowledge that CBHAs are ‘politically influential’ and ‘seen as being a success’ For some, GCC's negative views on CBHAs are not attributed so much to the behaviour of CBHAs as to the original stance of the council as a reluctant seller.

However, like the city council, CBHAs did not want GHA to appear in their midst: a registered social landlord like themselves but much larger, with significantly more funding (but also debt). They were bitterly disappointed that the housing stock did not come directly to them rather than through a two-stage process. Having put a lot of effort into helping ministers to establish GHA and the LHOs, CBHAs are increasingly frustrated that SST has not happened yet to give them a return for their efforts. Relations between the CBHA movement and GHA are very bitter, with GHA being described as ‘dysfunctional’ and ‘guilty of maladministration’.

A large element of the CBHA animosity towards GHA stems from the bureaucratic process put in place by GHA to assess the suitability of LHOs and CBHAs as receiving landlords under SST. CBHAs see this process as cumbersome, as well as inappropriate in that a fellow housing association, with less experience than themselves, is assessing their own competence and performance:

Now the information they ask for is nonsense. They are judge and jury over your performance … Four Gateways and we only know what one of them is. Now if you think this is ‘The Prisoner’, you're right okay. Three more Gateways, four more Gateways, I don't know what they are. I'm just telling you that they are really heavy duty in SST. I have to say that they're running a parallel universe, right. It's just not happening, right. In the meantime, they've cut up this decision into somewhere between five and seven. (CBHA Director)

However, in its inspection report, Communities Scotland (2007) concluded that “it is legitimate for GHA as prospective seller of stock to assess the suitability of purchasing organisations” (p. 44). This is unlikely to placate the CBHAs, who would ask why Communities Scotland cannot perform this task. So there is an issue about what the appropriate roles should be for the different parties, especially the regulator.

Underlying all this bitterness though, is the fundamental mistrust of GHA as a transitional body. This affects the parties’ interpretations of most of GHA's actions. A very important case in point is GHA's introduction of Local Shared Services (LSS) over the period 2005–2007: this has involved dividing up some of GHA's central services (e.g. financial management, organisational development, investment) into more local service units around the city (GHA, 2006b). Whilst this has been seen as “a positive step in the disaggregation of GHA's central services” (Communities Scotland, 2007, p. 69), others see it as a means of ‘locking-in’ LHOs to having their future services provided by GHA. The LSS development has also suffered from a ‘bottom-up’ approach: LHOs were asked to form their own LSS groupings and this resulted in seven groupings, with two dozen LHOs choosing not to participate (presumably aiming to get their central services from CBHAs or GHA centrally). The resulting configuration is not ideal:

We found the geography of the groupings is not necessarily the best for delivering services efficiently; this is a missed opportunity for GHA to achieve a strategic fit for delivering its service. (Communities Scotland, 2007, p. 69)

This is potentially an important issue for the future of the social housing system in the city—one of the original key goals of the stock transfer policy—if a large proportion of the social housing providers are configured in a way that is sub-optimal and not consistent with other emerging spatial governance arrangements such as Community Planning. One is left wondering where the oversight of the policy was during this crucial stage? Some independent assessment of the suitability of LSS to serve the ultimate policy goals should have taken place.

Social and Political Factors

Finally, the paper considers the role of social and political factors in the delivery of housing stock transfer. First, there is a look at political changes that have mostly occurred at the national level. Second, there is an examination of how key decisions and the attitudes of key players have been influenced by political considerations, both at the national and local levels.

Social and Political Change

There have been no dramatic social or economic changes in the city of Glasgow since the initial stock transfer in 2003, but there have been political changes at the national level of three types.

First, there have been several different ministers responsible for housing. Some argue that if the energy and drive of the early political champions for the policy had been retained, progress to SST would have been quicker. Subsequent ministers have had different priorities and perhaps less local knowledge, confidence and appetite to intervene in Glasgow's politics to try to sort things out.

Second, the national housing agency Scottish Homes was changed early under the Scottish Parliament from being a quasi-independent body to become part of the Scottish Executive itself. At this time, it lost some of its most experienced staff (often with RSL experience) and some would also say, ‘lost some of its bite’. Certainly their role in the stock transfer process has been heavily criticised:

Well second stage transfer, why it's no happened. What's the role of the regulator there, of Communities Scotland there?… you also get the impression that, em, if the regulator does occasionally raise its head above the parapet then they just get so scared of the GHA or the GHA immediately goes running to a QC… the regulator then backs off and says, I'm terribly sorry. (CBHA Director)

The third national change has been the advent of an SNP minority government within the Scottish Parliament since May 2007. They committed themselves early on to a review of the GHA situation, but this has yet to produce a notable outcome. In their first major housing policy statement, the new Scottish Government avoided use of the terms ‘community ownership’ and ‘community empowerment’, placing more emphasis on giving tenants a greater say, and accepting that there would be a variety of outcomes for communities in Glasgow:

We want to see tenants being given a greater say in the management of the houses and their neighbourhoods, including through second stage transfer where that is what tenants want and where it is sensible and financially achievable. (Scottish Government, 2007, p. 56)

This is not quite what the policy's supporters want. The CBHA movement has sought to use knowledge of the SNP's preferences (particularly for community land ownership in rural areas) to argue for stronger community rights to SST, attempting to reverse the power relation from seller to buyer:

The Scottish Parliament should approve legislation that will give communities the right to purchase the stock which is currently the subject of a management agreement approved by ministers under statutory order. (Lyn Ewing, Glasgow and West of Scotland Forum of Housing Associations, The Herald, 22 September 2007)

Surprisingly, for many observers, the SNP-led government has been fairly cautious and silent on the matter, and in its inspection report, Communities Scotland noted: “GHA has been reluctant to commit itself to an alternative direction on SST in future until the policy objectives of other key stakeholders—mainly Scottish Ministers—are clear” (CS, 2007, p. 47). But perhaps reflecting uncertainty among SNP Ministers, the report went on: “However, GHA clearly has a central and legitimate role to provide leadership on the future strategy for the ownership and management of its houses” (CS, 2007, p. 47). This somehow has to be married with giving tenants a greater say.

However, the SNP Government has been clearer on other issues and has announced the abolition of Communities Scotland, with its regulatory function (defined as setting standards, and measuring the performance and value for money of landlords) being reassigned to have ‘greater independence’. This still leaves open the question of who monitors the implementation of a policy such as stock transfer—if it is now left to the core civil servants in the Scottish Executive, this is more likely to be done in a ‘softly, softly’ manner rather than pro-actively since experience suggests that civil servants do not like to get involved in local political issues.

The Politicisation of Policy

Lowe (2004) has reminded us recently that in respect of welfare provisions, ‘politics matters’ and that housing, more so than some other areas, is a policy sector very much in the public domain and subject to conflict (p. 28). Yet, the housing policy network is, as Lowe describes it, becoming less vertically integrated with a more extensive network that the centre is less able to ‘steer’. Given Glasgow's municipal culture and the fact that in the UK, “in housing affairs local politicians have been both more confident and disposed to intervene” than in other areas of service delivery (Cole & Furbey, 1994, p. 122), the politicisation of housing stock transfer should come as no surprise.

Whilst some key difficulties with Glasgow's stock transfer stem from the technical and financial challenges involved in trying to improve and disaggregate a large housing service suffering problems, a key lesson for policy makers is that of the need to give more explicit attention to the political issues surrounding policy, and especially to how the politics of the situation influences the stance, behaviours and decisions of key actors.

At a national level the political thrust towards ‘community ownership’ has had a number of repercussions. The link between ownership and empowerment has been poorly specified, with confusion then ensuing as to whether community ownership is a means or an end in itself. In people forming judgements about the policy, community ownership has been at the forefront (as an end) much more so than the other policy goals.

The political emphasis upon community ownership has had two other important consequences. First, the ‘bottom-up’ and ‘small is beautiful’ ideology led to the instigation of a process that produced potentially 60 SST recipients, well beyond the original policy makers' intentions, and now seen as a major difficulty. The same approach helped produce a set of arrangements for ‘shared services’, which is seen now as sub-optimal in strategic terms. Second, GHA has felt under pressure to be ‘getting on with’ SST as a result of which it raised false expectations that it could substantially deliver SST over four years to 2008, now the cause of severe criticism (see Communities Scotland, 2007, p. 43).

National political changes, of minister and in the treatment of the regulating agency, have contributed to a lack of overall control of the policy. At times, it has seemed as if no one wanted to take charge or responsibility for the implementation of national policy in Glasgow until forced to do so by political pressure from beneath. Key decisions were made by the parties locally without any independent assessment of their potential effects.

Spanning the national and the local levels is the unresolved issue of the ‘funding gap’ within GHA's business plan. At the national level, the Scottish Executive opted to use an unusually low discount rate in the initial pricing of GHA stock, but they should have recognised that this could cause a problem of high stock valuations for LHOs and CBHAs later on.

At the local level, it is reported (Pawson et al., 2008, forthcoming) that GCC, through its consultants, knew before the first ballot that there would be a funding gap for SST as a result of this approach, but this information does not appear to have been transmitted or to have been recognised by all the other parties, until it emerged in analysis for GHA and Communities Scotland in 2005. Silence on this issue served the interests of the two main protagonists: it reduced the Scottish Executive's financial commitment to the GHA Business Plan; and for GCC, it protected the amount of funding within the GHA Business Plan available for housing stock investment. Thus, different political interests at both levels influenced the fact that an extremely important problem was created for the future but then not addressed until it could be ignored no longer.

At a local level, the stock transfer policy has faced the strong municipal culture of the city. This was always going to lead to vocal opposition to what was seen by some as the ‘privatisation’ of council housing in the city. This fed into the decision to have a two-stage transfer process so as to avoid multiple, local ballots having to be fought by ministers across the city at the same time, and with potentially variable outcomes. However, this two-stage process may turn out to have been more trouble than it was worth. But the alternative of the municipal culture was fierce criticism of the council as a monolithic, poorly performing housing provider. This fed into early rejection of the option of dividing the city's housing up into a smaller number of units as this was seen as too similar to GCC's own operational structure. Into the midst of this situation, politicians created GHA as a large housing provider without the public sector and dependent on private lenders—a sure-fire recipe for criticism that has not ceased since.

Local inter-organisational politics, as exemplified by the Glasgow scenario, has also been under-appreciated by policy makers. The long-standing consensus that community-based housing associations are highly performing and empowering organisations is now breaking down as the basic assumptions of policy are being questioned. There is underlying animosity between the key partners who have to work together to implement the policy: the city council, GHA and the community based housing associations all question each other's motives. The city council now sees itself, and to some extent its tenants, as the victims of the housing associations' predatory behaviour.

The city council and the housing associations are suspicious that the GHA is seeking to be a permanent feature of the local institutional landscape. There is potentially a fundamental and perhaps naïve flaw in a policy design which creates a large, new organisation in the midst of all the pre-existing social landlords, hands most of the power to it, and expects it to both develop organisationally in order to improve housing service performance but also to design and implement its own demise through stock sales.

Another key lesson is that for a programme like this to work over a period of ten years or so, there needs to be strong and independent oversight of the process: to assess decision making, to arbitrate in disputes, and to step in where delaying tactics are alleged. The stock transfer process in Glasgow has lacked such a presence.

Conclusion

In relation to the case study presented here, it can be seen that the implementation of housing stock transfer in Glasgow has been hindered by three key deficiencies. First, there have been weaknesses in policy formulation, most particularly a lack of attention to the balance and interactions between competing policy objectives over time, as well as insufficient consideration of important financial and organisational elements of the implementation process. Second, the policy has lacked consistent and strong oversight by and on behalf of ministers, which has allowed the policy to drift and key decisions to be made without proper assessment of their impacts upon the ultimate policy goals. Third, there has been a lack of appreciation of how the politicisation of the policy at national and local levels has influenced the course of its implementation, and at the same time a lack of awareness of how the local politics of housing within the city affects working relations with the key implementation agency, GHA.

The study of policy implementation must therefore contain an appreciation of the effects of having multiple policy objectives, multiple layers to the governance of policy, and multiple actors involved in policy delivery. This echoes the suggestion by Hill & Hupe (2000, p. 43) that the study of implementation might have to be varied according to the type of policy issue concerned and the institutional context within which it occurs. The crucial elements of the framework here for studying policy implementation therefore turned out to be as follows: the extent to which policy formulation had considered issues of implementation; inter-organisational relations in the local arena and the degree of oversight of this policy network; and the effects of political considerations and competing political interests and perspectives upon key decision-making and on inter-organisational relations.

The Glasgow stock transfer policy issue is a good illustration of what Exworth & Powell (2004, building on Evans & Davies, 1999) call the “need to address the ‘spatial domains’ of implementation” (p. 278). Looking at two of the ‘policy windows’ they discuss (the other being the ‘horizontal dimension at the centre’), it can be seen, first in the ‘vertical dimension’, weaknesses in policy design and a reluctance and naivety on the part of central government as regards policy oversight. In the local, horizontal dimension, the issue is not simply (as they discuss it) that of whether local partners can agree a common purpose or have it imposed on them, but also whether central government, either directly or through its agent (in this case Communities Scotland) is itself willing to become a player in the local policy network and engage in network management (Klijn & Teisman, 1997) in order to shape outcomes. This may be important for two reasons: first, the spatial arena of Glasgow is nationally significant in scale and political terms (see Kintrea, 2005), so the outcome is especially important for social housing in Scotland; and, second, the local housing network, which is a highly politicised domain, is both the subject of policy and the means for achieving policy, a rather unusual context (in Hull & Hupe's terms).

But as has already been stated, the Glasgow housing stock transfer policy as a whole has endemic policy conflicts: between housing improvements as an urgent priority and community empowerment as a goal; and between community empowerment and housing system efficiency and effectiveness. Policy design and delivery has to get cleverer at taking into account such trade-offs and tensions, rather than continuing to operate as if policy targets can be measured and achieved as if they were mutually exclusive phenomena (Mulgan & Lee, 2001). The Glasgow case provides further evidence of ‘competing institutional logics’ in social housing, which Mullins (2006) described as between local accountability on the one hand, and scale and efficiency on the other. Here, similar tensions have existed between community empowerment (incorporating but going beyond accountability) through LHOs, and economy and effectiveness of delivery through the GHA. But a third institutional logic could be added that will also come into play and which has a distinctly spatial focus, namely the production of an effective housing system and housing governance structure for the city as a whole.

Acknowledgements

This research was conducted as part of the GoWell Research and Learning Programme. GoWell is a collaborative partnership between the Glasgow Centre for Population Health, the University of Glasgow and the MRC Social and Public Health Sciences Unit and is sponsored by Glasgow Housing Association, the Scottish Government, NHS Health Scotland and NHS Greater Glasgow & Clyde. Thanks also to Professor Hal Pawson for discussions with him about Glasgow's stock transfer.

Notes

¹ This is not necessarily to accept the label of ‘failure’ applied to the stock transfer policy in Glasgow.

² The period of study is 1999 to mid-2007. There have been some recent policy developments aimed at furthering community empowerment, but at the time of writing, there had still not been any second stage transfers.

³ However, there is a community consultation process underway in relation to eight large regeneration sites.