District Heating in the UK: a Technological Innovation Systems Analysis

David J. C. Hawkey

Institute of Governance, School of Social and Political Science, University of Edinburgh

Chisholm House, High School Yards, Edinburgh, EH1 1LZ

Tel: +44 (0)131 650 2841 Fax: +44 (0)131 650 6345 Email: dave.hawkey@ed.ac.uk

Revised paper submitted to Environmental Innovation and Societal Transitions

District Heating in the UK: a Technological Innovation Systems Analysis

Abstract District heating infrastructure could contribute to the UK’s energy policy goals of decarbonisation, renewable energy deployment, tackling fuel poverty and ensuring energy security. However, while a number of schemes have been developed over the last decade, deployment of the technology remains limited. This paper adopts a Technological Innovation Systems framework to analyse prospects for greater deployment in the UK, and to suggest how these prospects could be improved. While district heating networks are inherently local infrastructures, they are positioned in regulatory and market contexts organised at larger spatial scales, making geography an important factor and coordination across spatial scales an important policy area for accelerated deployment.

1. Introduction The UK has a long and chequered history of attempts to develop district heating (DH) systems – networks of insulated pipes which deliver heat via steam or hot water to serve the space and water heating demands of multiple buildings (Russell, 1993). UK Government and Devolved Administrations (particularly in Scotland) have indicated that accelerated roll out of the technology would contribute to achieving national energy policy goals (Department of Energy and Climate Change (DECC) 2012; Scottish Government, 2011). However, given a history of failed attempts to establish far-reaching DH programmes in the past, and the small share of DH in the space and water heating market (around 2% in comparison with Denmark’s 47% and Sweden’s 55%, Euroheat & Power, 2011), the extent to which DH will be deployed, particularly on the timescales established by 2020 carbon and renewable energy targets, is highly uncertain. This paper’s central

question, therefore is: “what are the principal challenges to significantly scaling up the deployment of DH in the UK?”

While the technical components of DH are relatively mature, having been developed over forty years of widespread use in Scandinavia (Dyrelund & Steffensen, 2004; Ericson, 2009; Rutherford, 2008), their deployment in different physical, social and institutional contexts presents new challenges requiring innovative organisational, contractual and commercial solutions. Two features of the UK context are important here. First, while DH is an inherently local infrastructure (limited to high density areas by financial, rather than physical, constraints, Roberts 2008), it is nonetheless situated in systems of regulation and government, resource flows and markets which operate at local, regional, national and international scales. Liberalisation and privatisation of the UK energy market have altered the scope for public authorities to direct development of energy systems towards social and environmental goals, and have consolidated existing assets under the control of a small number of companies whose international scope challenges development of locally-specific systems (c.f. Rutherford, 2008). Secondly, shifts in the role of local government, from service provision to enabling others to provide services (Bulkeley & Kern, 2006), accompanied by a proliferation of public and private service providers (Cook, 2009; Leach & Percy-Smith, 2001) has reduced the in-house capacities of local authorities to collect and interpret data, to plan and design technically and financially viable schemes. Both features contrast with the municipal energy companies which developed DH in Sweden and Denmark in the twentieth century (Dyrelund & Steffensen 2004; Summerton 1992). While the UK is arguably at an extreme end of these spectrums, given its early energy market liberalisation and history of centralised control over local authorities (Wilson & Game, 2002), these broad issues reflect the direction of travel in other European countries (Ericson, 2009; Monstadt, 2007; Rutherford, 2008). Addressing DH in the UK can therefore shed light on the processes by which contemporary municipal actors can orchestrate or influence local responses to sustainability challenges, and thereby contributes empirical material to a growing literature on the roles of geography in innovation processes (Geels, 2011; Hodson & Marvin, 2010; Truffer & Coenen, 2012).

The paper is organised as follows. Section 2.1 introduces the Technological Innovation Systems (TIS) analytical framework and source material used. The following three sections apply the analytical framework, section 3 describing the structure of the TIS, section 4 detailing the TIS’s functional pattern and section 5 appraising the functional pattern and identifying key policy issues. Section 6 discusses implications of the analysis for DH in the UK, and draws conclusions.

2. Research approach

2.1. Analytical Framework A Technological Innovation System (TIS) can be defined as “a network of agents interacting in a specific economic/industrial area under a particular institutional infrastructure or set of infrastructures and involved in the generation, diffusion and utilisation of technology,” (Carlsson & Stankiewicz, 1991, p.111). This definition provides a suitable framework for addressing the central question of this paper, for while DH is a well established technology in other contexts, new commercial and organisational forms for the development and operation of heat networks are central to the prospects for greater deployment of the technology in the UK.

Systems approaches to innovation focus on multiple agency and distributed learning in technological change (Winskel et al., 2006), and, drawing on Evolutionary Economics, emphasise the diversity creation, replication and selection (Malerba, 2002). A systems view broadens both the scope of enquiry and rationales for policy intervention than a narrow focus on market failure (Foxon, 2007). In addition to the value of systems-thinking, a TIS approach responds to the importance, as argued by Jacobsson & Bergek (2011), for technology-specific policy interventions in pursuit of sustainability. Jacobsson & Bergek’s argument is twofold: first, the closing window of opportunity to mitigate climate change (International Energy Agency (IEA), 2011) means that decarbonisation following sequential adoption of least-cost technologies is unlikely adequately address the challenge. Second, technology neutral policies ignore the unique and multi-dimensional growth processes of

different technologies which rarely (if ever) conform to the linear model of innovation such policies assume.

This paper adopts TIS scheme of analysis presented by Bergek et al. (2008) which builds on a rich empirical and theoretical literature. Early TIS studies emphasised the structure of a TIS, in terms of the actors involved, networks among them and institutional context. While this remains an important component of analysis, shortcomings in an exclusively structural approach (for example, strong networks could either facilitate or stifle a TIS, Jacobsson & Bergek, 2011) have led to inclusion of specific activities which contribute to the overall TIS performance. This “functional TIS” approach abstracts a set of processes from innovation and other literatures, whose fulfilment contribute to a TIS’s overall function of achieving technological change (Bergek et al., 2008; M. Hekkert et al., 2007; Negro et al., 2007).

TIS has largely been applied to particular renewable energy generating technologies (Jacobsson & Bergek, 2011) whereas the current study applies it to an infrastructure technology. DH networks, with high upfront costs and long payback periods create powerful interdependencies between subscribers (who regard heat as an essential service) and network owner (whose investment recovery relies on long term stability of subscribers) (Summerton, 1992). This close coupling between subscribers and network developer blurs the boundaries between categories of actor, particularly when (as is often the case in the UK) major subscribers take on roles in development and financing DH. While innovation systems often focus on “enterprises, universities and government research institutes” (OECD, 1997, p.7) analysis here draws the boundary of the TIS more widely to include other government bodies (see section 3.2).

The role of space in technological change is recognised in the literature as an underdeveloped area (Truffer & Coenen, 2012). One aspect of spatial relations which has received some attention is the role of cities in low carbon transitions (Bulkeley et al., 2011). Historical analyses of the role of cities in sociotechnical change highlight differences in the spatial scale of technologies (Geels, 2011). However, as Hodson & Marvin (2010) note, organisations involved in contemporary sustainable sociotechnical change in cities (either as drivers or barriers) often operate at different scales, creating significant

coordination challenges. DH, as a locally-bounded technology is confronted by these challenges in the UK as other energy systems and (often) key subscriber organisations operate at supra-local scales.

Bergek at al.’s (2008) scheme involves six steps, which shape the structure of this paper. Steps 1 (defining the TIS) and 2 (identifying structural components) are dealt with in section 3. Step 3 (mapping the functional pattern of the TIS) is undertaken in section 4. Steps 4, 5 and 6 (assessing functionality, identifying inducement and blocking mechanisms and identification of key policy issues) are in section 5.

2.2. Research methods This analysis draws on evidence gathered between 2009 and 2012 as part of comparative research on DH development.1 This includes

• over 35 semi-structured interviews with UK, devolved and local government officers, DH practitioners and current or potential DH subscribers;

• documentary analysis (including UK, devolved and local government policy and project documents, case studies, media reports and DH guides);

• participation in and observation of 14 meetings and workshops directly addressing DH policy and/or project development at UK, Scottish and municipal scales.

This research material is analysed to provide a contemporary snapshot view of DH in the UK in order to identify ways key policy areas. As the approach taken here does not involve temporal analysis or comparison across cases, a qualitative rather than quantitative approach is adopted. A qualitative approach also affords greater flexibility in the kinds of material which contribute evidence of the performance or otherwise of TIS functions.

1 The “Heat and the City” project (www.heatandthecity.org.uk) began in 2010 and is funded by the UK Research Councils. See for more information.

3. Structure of the UK DH TIS

3.1. Defining the TIS in focus Defining (and delimiting) the TIS is crucial to determining the outcome of analysis. Following Bergek et al. (2008), the TIS is understood as an analytical construct, and delineated on the basis of the research question, guided by characteristics of contemporary DH practice.

Many configurations for the supply of heat via insulated pipes are possible. As this paper is motivated by (inter alia) the policy goal of decarbonising heat supply, I restrict analysis, in terms of artefacts and applications, to networks which have the potential to supply large quantities of low carbon heat, including by network expansion or interconnection with other heat networks. This growth potential is clearly greatest in urban centres, so small rural heat networks are excluded. I include heat generation (or capture) equipment as separation of heat generation from distribution (organisationally and financially) is currently rare in the UK. Common praxis in the UK is to establish heat networks around gas fired CHP, with the value of electricity generated supporting heat sales competitive with onsite gas boilers (Kelly & Pollitt, 2010), so CHP is also included. The flexibility of DH to take heat from different sources is important to its sustainability (Sustainable Development Commission (SDC), 2009) so other heat sources (including biomass, large heat pumps, heat recovery from large centralised power stations) are also included, albeit peripherally. Industrial heat demand is generally higher temperature than space and water heating demand, so networks supplying heat for industrial application (and industrial CHP) are excluded.

Territorially, this paper focuses on England, Wales and Scotland, reflecting the low levels of deployment of DH across urban centres, and commonalities in the market, regulatory and cultural context within the UK.2 This spatial focus does not preclude inclusion of actors and institutions based outside England, Wales and Scotland, if they are relevant to the deployment of DH in

2 Northern Ireland, whose energy markets are distinct from the rest of the UK (being integrated with the Republic of Ireland), is excluded from this analysis.

this spatial domain. Within this territory, activity at local level around specific projects is an important aspect of the TIS, as is activity at national level.

3.2. Identifying structural components of the TIS Actors, networks and institutions comprise the structural components of a TIS (Bergek et al., 2008). In European countries with widespread DH coverage, local authorities (LAs) played crucial roles in development and operation of heat networks (Grohnheit & Gram Mortensen, 2003). The UK follows this pattern: where DH is or has been developed, LAs are key actors (Kelly & Pollitt, 2010), taking often multiple roles including:

• Gathering data and initiating investigation into potential schemes;

• Establishing an area wide, strategic, long term vision for heat into which particular initiatives can be embedded;

• Drawing on established relationships to broker new relationships between local heat users and DH contractors;

• Taking heat for the significant and varied heat demands of the LA estate;

• Integrating DH with local priorities including local emissions reduction, local regeneration, tackling fuel poverty, and waste management (via Energy from Waste facilities);

• Using planning policy to co-locate heat sources and heat demand, requiring or encouraging new development to connect to a heat network, and advising network developers on planning issues;

• Statutorily, LAs have powers to install and maintain heat networks;

• Contributing low cost finance to project costs;

• Taking a role in system governance (i.e. participating in ongoing decision making about system operation and policies, relationships with local stakeholders and strategic issue, often through representation on the board of the company operating the DH system) helps mitigate subscribers’ perceptions of risk of monopoly exploitation.

At a local level, large heat subscribers are also included in the TIS as they are in many cases involved in project development, system financing and

governance of DH systems (through formal representation in the decision making procedures of the company operating the system, Hawkey et al.). Project-level interviewees report that formation of local networks around a nascent project is often challenging, in part due to the absence of established relationships among local stakeholders creating communication and trust challenges. These challenges can be exacerbated when local buildings are owned and/or managed by organisations operating above a local level.

The scale of DH activity in the UK is described by some industry practitioners as a “cottage industry,” and this is reflected in the small number of small scale specialist contractors, who are a mix of UK companies and subsidiaries of European companies, and straddle local and national scales. There is some activity among the UK’s “Big Six” utility companies but this too is at a small scale, particularly in relation to their other activities.

A number of intermediary organisations also contribute to the TIS (including government agencies established to promote efficiency and carbon reduction in energy end-use, sustainable energy deployment, and research into building-design). A range of DH consultancy services are offered by DH-specialist, engineering, finance and legal consultancies.

Several networks undertake different activities. The recently (2011) established District Heating Vanguards Network attempts to foster interaction across LAs, and to date has operated as a loose affiliation with biannual issue-specific workshops.3 Prior to their abolition, some English Regional Development Agencies sought to develop regional networks on various sustainable energy technologies, including DH. Two well-established industry associations represent the interests of commercial DH practitioners, particularly through lobbying UK Government.

The institutional infrastructure within which agents operate to diffuse and utilise DH is unsupportive in several ways. Formal institutions are weak as DH is not specifically regulated (falling instead under general consumer protection regulation) in contrast with the extensive regime of regulatory

3 The author has contributed to planning and running these workshops through the Heat and the City project.

control and protection which governs electricity and gas supply. The lack of regulation creates risk that future regulation will disrupt the long-term business model of a DH system (the time inconsistency problem, Helm, 2010), and compounds challenges created by cognitive institutions, as the unfamiliarity of DH often leads to heightened perception of risk among potential subscribers, particularly concerning service reliability. DH challenges routine organisational behaviours regarding energy supply in the UK, particularly where a heat network developer requires long term contracts with subscribers.

4. Functional pattern of the UK DH TIS The seven functions proposed by Bergek et al. (2008) are here presented in the following order to aid exposition: (1) entrepreneurial experimentation, (2) knowledge development and diffusion, (3) market formation, (4) resource mobilisation, (5) influencing the direction of search, (6) legitimation and (7) development of positive externalities. An advantage of these functions being abstracted from a broad literature is that they are flexible and can be tailored to the specific case under consideration.

4.1. Entrepreneurial experimentation Entrepreneurial experiments reduce uncertainty by testing the space of possible new configurations of finance, business models and organisational forms. At a project level, fashioning a viable system from local physical and organisational arrangements requires significant leadership from project developer and/or local authority. A range of permutations of investment, ownership and governance roles across local stakeholders and non-local contractors is possible (London Energy Partnership (LEP), 2007), and have to be tailored to both the cash-flow profile of a project and the preferences and capacities of project-level stakeholders (Hawkey et al., forthcoming).

Within local organisations involved in project development, particularly local authorities, project success often depends on the presence of a committed champion able to entrepreneurially bring together organisational-internal and -external components into productive alignment, often going well beyond the requirements of their post. This can include linking environmental, asset

management, fuel poverty and planning functions along with lifecycle costing (an accounting approach which challenges a traditional local government revenue/capital split) with non-local networks of experience and expertise (Hawkey et al., forthcoming).

The high degree of organisation-internal effort and cross-stakeholder coordination required to develop and implement projects slows, and sometimes stifles, activity. Some aspects of project development are necessarily bespoke to the particularities of the place and organisations involved, but others (e.g. compliance with state aid and procurement rules) are less so, and some of the effort involved in configuring elements at a local level replicates efforts undertaken elsewhere (Lovell et al. 2011).

Relatedly, weak coordination across urban areas stifles entrepreneurial experimentation. For example, various approaches to local planning policy could support DH, including specific agreements with developers and area-wide policies requiring connection. However, as DH sits outside many domestic and commercial developers’ established models, LA officers report that planning policies for DH create perceived risks of developer flight (to near by administrative areas). In London this issue has been addressed by entrepreneurialism above the local level, as the Greater London Authority (GLA) has established London-wide DH planning policies (GLA 2009).

4.2. Knowledge development and diffusion As a mature technology, basic research (for example, into new insulation materials), while important for the incremental improvement of DH, is less relevant here than knowledge of local circumstances (particularly energy demand) which underpins project development, and knowledge of project development, commercial arrangements, organisational forms and legal issues.

Heat demand data, drawn together in a “heat map” of an area is critical to DH project- and strategic-planning. Local DH developers, however, find access to consumption data held by utility companies is restricted by commercial confidentiality and data protection. National and devolved government have intervened to develop heat mapping approaches. For England, DECC have produced a heat map covering the whole country

(DECC, 2012), while in Scotland a standard methodology for LAs to develop heat maps has been created (Scottish Government, 2011).

Other forms of knowledge concern effective business models, legal structures and implications, finance models, risk and contractual forms. Some LAs have a degree of in-house expertise which can handle these issues, but frequently external sources of expert advice are required. Independent advice is available, but as the intellectual property of consultants it is often expensive. Some DH companies offer lower cost advice, but the inherent information asymmetry means LAs run the risk that rather than a project being optimised in terms of the LA’s goals, it is instead designed to fit the strengths or established model of the company (Homes and Communities Agency (HCA) 2011).

LA project teams with experience developing DH systems have much knowledge of project development, commercial arrangements, legal restrictions and organisational forms which other LAs value. However, the limited development of networks mean this knowledge has to date remained fragmented. While some LA project teams express interest in using their experience to offer consultancy services or to develop projects in other areas, only one has done so (Woking Borough Council, via its arms-length company, operates a heat network in Milton Keynes).

4.3. Market formation Market formation concerns the trajectory of market scale (from “nursing”, via “bridging” to “mass” markets), and the factors driving formation (Bergek et al. 2008). The low levels of deployment indicate DH is currently at a “nursing market” stage.

Two techno-economic characteristics of DH systems impose important constraints on market formation at a local level. First, larger networks accommodate more balanced load profiles through diversity, support heat storage and lower heat generation costs by connecting different heat sources, they have stronger financial viability than smaller networks (IEA, 2005). However, the complexity of coordinating a large number of stakeholders precludes development of large networks in a single phase. DH systems are therefore often scheduled with an initial “catalyst” phase based on a small

number of large heat users with subsequent expansion envisaged but uncertain. This first phase is likely to have poorer financial viability than an expanded system. The second characteristic is the long timescale required for investment in DH to payback and achieve a target rate of return – timescales are project specific, but business models of 25 years or longer are not uncommon.

These characteristics make public sector buildings key subscribers in the initial phases of market development. Their scale can anchor first phase networks. Insolvency risks for public sector bodies is perceived to be low, meaning long term heat supply contracts can be established. However, project developers interviewed find coordinating different public sector organisations, with individual budgets, different schedules for decision making and heating refurbishment, and organisation-internal carbon and energy management plans, is challenging, particularly given the dependence on local decision makers embedded in supra-local organisations.

While public sector organisations’ commitments to reducing greenhouse gas emissions, and energy requirements in regulations for new buildings have potential to establish market niches for low carbon heat solutions, DH competes with other technologies. The spatial interdependencies of DH mean that this form of competition is particularly acute, as on-site solutions can prevent key anchor loads connecting to an area wide network, undermining its viability. Onsite technologies are currently supported by Feed in Tariffs (including a Renewable Heat Incentive), available regardless of the spatial relationship between an installation and other sites of energy demand.

The market for decentralised CHP electricity in the UK is limited. Accessing the well established national electricity market is challenging for a number of reasons. Small generators struggle to connect to distribution networks on acceptable terms, and complain of a lack of price transparency and responsiveness from network operators (Ofgem, 2011). Once connected, they are rarely able to participate directly in electricity markets, lacking the technical capacity, financial scale and balancing portfolio of generating plant necessary to handle the risks of participation in wholesale markets. Consequently electricity sale on the public system is often via consolidators who take on balancing and other market risks but in return offer low tariffs

(Toke & Fragaki 2008). Greater value may be captured by direct supply of electricity (at retail rather than wholesale price), though the UK’s licensing regime restricts the power that small suppliers can deliver. Until recently private wire networks could ensure CHP operators a long-term market (often supplying electricity along with heat), but a European Court of Justice ruling (the “Citiworks” case) requires third party suppliers have access to these networks, undermining the long-term stability of the revenue they generate.

4.4. Resource mobilisation Development and implementation of DH systems requires various resources to be mobilised, including engineering, legal and commercial expertise, development funding and capital funding. LAs are often dependent on external sources of expertise (as described above) and the costs of project development can be high (around 10% of project capital costs). Development funding is available, for example from the European Investment Bank (EIB), though mobilising this resource is both a lengthy procedure and generates financial risk: the EIB threatens to claw back its technical assistance if recipients do not leverage in project investment 20–25 times the amount granted.

Finance interviewees report that, while the low rates of return and long timescales of DH business models can be acceptable to institutional investors (such as sovereign wealth funds and pension funds), the minimum scale these investors consider is larger than the capital required for individual projects, particularly in the first phase. While in the long run these deep pools of capital could be used to refinance a portfolio of projects, they are not available for initial development.

Schemes drawing on other sources of commercial finance (including ESCo and utility company sources) are found to require higher rates of return. In part practitioners attribute this to unfamiliarity of funders with investment in heat, the lack of a delocalised, standardised business model, and the sunk-investment character of heat networks. The immobility of insulated pipes once installed means their value is the value of the heat loads connected to it (rather than a re-use or recovery value), so heat off-take and bad debt risks are crucial for funders. Reducing these risks is a key challenge in mobilising

financial resources, but the willingness of local authorities to take on risks associated with other (public or commercial) organisations’ heat off-take agreements, either by underwriting commercial finance or contributing public finance, is low. The resulting low levels of “bankability” of heat demand for projects that reach beyond the local authority estate is therefore a key challenge to the mobilisation of finance.

Intermittent grant programmes in the UK have supported development and capital costs. These include grant programmes targeting DH (Community Energy Programme, 2002-2005; Low Carbon Infrastructure Fund, 2009-2010) and an area-based energy company obligation (Community Energy Saving Programme, 2009-2012). While these have been important stimulants to activity, and helped leverage in additional financial resources, their short timescales relative to project development made them most suitable to projects that were already being pursued, and created spikes in demand for consultancy and contractor services, pushing prices and lead times up. Additional finance, mobilised via building standards (which allow investment in off-site solutions to offset on-site emissions) and future energy company obligations could be significant, though uncertainties in the rules for these programmes and a history of unexpected changes (for example, ‘zero carbon’ building standards were redefined in 2011, Zero Carbon Hub, 2011) contribute uncertainty.

4.5. Influencing the direction of search The “search” referred to in Bergek et al.’s (2008) scheme of analysis refers to factors which draw organisations into the TIS. While DH has potential to contribute to three key energy policy goals, decarbonisation, affordability and security (through flexibility and diversity), this does not inevitably draw actors into the TIS. The translation of these pressures into a case for entering the DH TIS is in part dependent on the social acceptance of DH and its perceived alignment with these goals (i.e. legitimation, function 6 below) in a context of competing heat supply options.

Estimates of economic potential contribute to beliefs in growth potential, but because the economics of heat networks are often marginal, small changes in modelling assumptions can lead to large differences in such estimates. For

example, modelling work commissioned by DECC in 2009 produced estimates ranging from zero to 14% of UK heat demand (Pöyry Energy, 2009). Data access and aggregation also have significant impacts on estimates of economic potential. The heat demand database constructed for England’s national heat map suggests up to half of England’s heat demand is in areas which could support financially viable heat networks (DECC 2011b).

Factor and product prices influence organisations’ decisions to enter a TIS (Bergek et al. 2008), and these do not currently provide strong incentives for entry. The small scale of DH industry in the UK means equipment (particularly insulated pipes) is imported, and other factor costs are high relative to other European countries. In part this reflects high risk premiums attached to civil engineering works (trench digging, filling and resurfacing) due to a lack of experience among civils contractors and to subsurface utility congestion (Pöyry Energy 2009). Concerns about volatility in fuel and electricity prices have also acted as a disincentive to entry (IEA, 2008), and while recent price movements have improved the viability of DH/CHP (Kelly & Pollitt, 2010), the possibility of a “golden age of gas” (IEA, 2012), which could depress competing heat prices, contributes uncertainty. Long term visibility of energy prices in the UK is generally poor reflecting low levels of market liquidity (CHPA, 2011), challenging development of robust long-term DH/CHP business models.

Statutory climate change duties, a new carbon tax and planning guidance are variously cited by LAs within the TIS as reasons for entry. However, interpretations of and responses to these pressures vary. For example, planning guidance from central and devolved authorities encourages LAs to adopt decentralised-energy policies, but gives no indication of where in a hierarchy of priorities this should be placed (Williams, 2010).

4.6. Legitimation Social acceptance and compliance with relevant institutions constitute legitimacy. At a project level, DH developers report investing considerable effort in establishing legitimacy among subscribers and local decision makers for DH as an unfamiliar and unregulated energy supply. Local pilot projects are often important in establishing legitimacy among LA officers and elected

representatives. Perceptions of risk among potential subscribers leads to often protracted negotiation between system developers and subscribers around price, service level guarantees, and redress procedures. Strategies employed to build legitimacy at a local level include connection of well regarded businesses (such as national supermarket stores), open book accounting, demonstration of previous experience and performance (particularly in the UK), and local authority involvement in system governance.

At national levels, legitimacy is weakened by concerns that DH will lock in fossil fuels (gas CHP), and by difficulties incorporating it (and other network technologies) into computational models of energy system evolution and future scenarios (DECC 2012). DH advocates point to successful heat source diversification in Scandinavian DH networks, and examples of DH contributing balancing services to electricity systems (Rotheray 2011; Toke & Fragaki 2008), though the scale of DH activity in the UK means similar examples from the UK are limited.

Industry representation shows a degree of fragmentation, with two associations separately advocating the technology to policy makers, and the larger association (the Combined Heat and Power Association) also advocating other technologies (particularly industrial and micro CHP). In 2009 the UK Government proposed a Heat Markets Forum with industry, government and consumer representatives, to consider regulation and other ways of building confidence (DECC 2009), but this is yet to be established.

4.7. Development of positive externalities “Positive externalities” referred to in Bergek at al.’s (2008) scheme refer to system-wide resources which may emerge as the result of activity within the TIS, and which often enhance the performance of other functions.

Industry representatives report that intermittent nature of grant funding (section 4.4) has muted the development of positive externalities, as without stable and long-term support, firms have not been incentivised to invest in skills or supply chain development. However, some organisations (including the Energy Saving Trust, the Homes and Communities Agency and English Regional Development Agencies, RDAs) did develop capacity over the last decade to support LA officers developing projects, though budget pressures,

and the abolition of the RDAs has since led to these activities being scaled back or abandoned. London’s Decentralised Energy Programme Delivery Unit (DEPDU) represents a notable exception, having secured European funding to provide London boroughs and other project sponsors with technical, financial and commercial assistance.

5. Appraisal of the DH TIS

5.1. Assessment of TIS functionality and process goals Bergek et al.’s (2008) step four aims to identify what changes to the TIS would represent improved functionality, as this varies across different TISs at different stages of development. Here this is approached by considering the stage of development of the TIS and relationships between functions.

Although interest in DH in the UK can be traced back to the eighteenth century (Babus’Haq & Probert 1996), the TIS is currently in a formative phase with low levels of deployment and limited institutional support (cognitive and formal). At this phase, market formation and entrepreneurial activities by LAs are both necessary to establish new networks and so strengthening of these functions are central to expanding the deployment of DH. Expanding the market beyond initial networks would represent a subsequent phase of TIS development, though legal and physical configurations of “starter” networks could be future-proofed to facilitate expansion. Market formation is facilitated by heat loads able to mitigate first phase disadvantages, (e.g. the public estate), but local coordination and legitimacy building remain key challenges. Entrepreneurial experimentation by LAs is a central aspect of market formation by creation of new networks, but is stifled by difficulties mobilising development and project finance, access to knowledge resources, limited LA in-house capacities and fragmentation across LAs (reducing the development of positive externalities). There is scope for high UK construction costs to fall through increased activity (Pöyry Energy 2009), contributing to the guidance of the search, but industry interviewees suggest this would be improved by long term visibility of DH deployment in contrast with the intermittent activity grant funding has produced in the past.

5.2. Key inducement/blocking mechanisms and policy issues Inducement and blocking mechanisms are TIS-internal and –external factors which prevent or promote TIS functioning, and Bergek et al.’s fifth and sixth steps are identification of such mechanisms and specification of consequent policy issues (Bergek et al. 2008).

The absence of regulation blocks TIS functioning, both by creating additional challenges to project developers in recruiting subscribers (i.e. weakening market formation) and creating perceived risks among financiers that future regulation will undermine current business cases, leading to challenges mobilising resources. Regulation of pricing, service levels and redress procedures for DH could take various forms, from minimum standards to overarching specifications of DH business models (such as the Danish requirement that DH companies operate non-profit models, Dyrelund & Steffensen 2004).

Limited coordination across local and national levels of activity blocks functioning of the TIS in several ways. Market formation activities are generally undertaken at a project level, though many organisations which own buildings suitable for the first phases of network development operate a larger spatial scales. Local market formation would be facilitated if these organisations, particularly in the public sector, adopted clear organisation-wide approaches to DH which local building managers could follow.

A second national/local coordination issue concerns sensitivity to the spatial arrangement of heat demand. At a local level this is crucial for identifying project opportunities, but national level policies are formulated which are neutral with regard to this dimension. In particular, market interventions subsidise building-scale technologies with which DH competes regardless of other heat demand in the vicinity. Adding a spatial component to national heat-related policies could mitigate this blocking issue, and provide clearer guidance of LAs towards DH by establishing a common approach to appraising the suitability of different heat technologies to different areas. The regulatory framework for DH in Norway provides an interesting model, where applications to the national regulator for a concession to operate a DH

network are appraised against a range of social and technical factors (NVE 2009).

First phase disadvantages are an endogenous blocking factor, particularly stifling mobilisation of financial resources. Project developers suggest that scheme viability can often be improved by either injecting capital (improving the cash-flow profile) or underwriting heat off-take and bad debt risks (reducing required returns), but that the capacity and willingness of LAs to take on these issues is limited. Central or devolved governments, could play a role here, for example by pooling risk across different initiatives via an underwriting mechanism. The spatial dimension is relevant here, as uncertainties in the trajectory of UK energy systems means the “optimum” mix of technologies and infrastructures cannot be determined, making maintenance of diversity an important strategy to avoid lock-in to underperforming approaches (DECC 2011b). However, the value of diversity implies aggregation across systems, and is therefore less visible to local actors than from a regional or national perspective.

The limited capacity of many LAs to engage with DH is a further key block to local activity. Given that UK LAs have not directly engaged in energy provision since the nationalisations of energy sectors in the 1940s, industry representatives report it is often unclear which, if any, officer in an LA has a remit which includes exploring DH opportunities. Investment in LA officers in dedicated posts within urban authorities would both increase local capacity, but also contribute to forming a local authority network which could facilitate coordination and joint learning across urban areas.

The limited access to electricity revenues for small generators could be tackled in a number of ways, including regulatory changes for distribution network operators, modifications to the electricity supply licensing regime, facilitating aggregation of small generators into a “virtual” generator, and establishing a clear role for CHP generators in forthcoming capacity mechanisms (DECC 2011a).

6. Discussion and conclusions Two features of DH in the UK set this case apart from other innovation system analyses. First, as a mature technology, basic R&D activities are less relevant to DH than many other technologies. Nonetheless, the role of multiple-agency and distributed learning mechanisms (Winskel et al., 2006) are still important aspects of DH innovation, particularly in commercial and organisational arrangements. Second, due to the multiple roles of local government (as broker, entrepreneur, planner and subscriber) and the close coupling between large subscribers and system governance, these actors were included in the TIS, stretching the range beyond traditional “enterprises, universities and government research institutes” (OECD, 1997). In spite of these differences, the functional analysis proposed by Bergek et al. (2008) was found to be sufficiently flexible to apply to this case and provided a comprehensive view of system-internal dynamics.

Interaction with other sustainable heating technologies was considered within the TIS scheme in terms of blocking mechanisms and policy recommendations. While section 5 suggests that interaction with other sustainable heat technologies could be managed by supra-local planning (at least, in the design of relevant market interventions), the dynamics of activity within other sustainable heat TISs (such as heat pumps and biogas injection) was not considered (for example, whether such planning would be compatible with system functionality for those technologies). Such an analysis would require expansion of the TIS framework to consider multiple interacting TISs, along lines suggested by Markard & Truffer (2008).

The spatially bounded nature of both DH systems and the jurisdictions of key actors within the TIS makes space a key theme of this analysis. Regulation at a national level is a key policy area, as are other national-scale activities which go beyond setting a context for local activity to include more direct forms of participation (for example, pooling risk across schemes by underwriting, and inclusion of DH in supra-local subscriber organisations’ approaches to energy). Resolving these issues needn’t be interpreted as cities “receiving” national transitions (c.f. Hodson & Marvin, 2010), but could represent a division of labour between the local level and actors operating at a broader scale, reflecting both the strengths and weaknesses of each and their different

perspectives on the value of DH. It is likely that development of other sustainable sociotechnologies in contemporary urban contexts would require a similar division of labour when (a) part of the value of adopting an innovation is increased resilience through diversity, more visible above the local level, and (b) local coordination around an unfamiliar system is required, especially when local stakeholders are embedded within organisations whose spatial scope extends beyond the spatial domain of particular initiatives.

While DH is a mature technology, widely used in other European countries, its deployment in the UK in pursuit of climate change, long term affordability and energy security goals remains challenging: functional analysis revealed a range of reverse salients to accelerated deployment. Successfully tackling these will involve both capacity building at local level and collaborative action across local and national/devolved levels of government. Supra-local government action is not limited to establishing a suitable regulatory environment, but includes more active forms of support and coordination across localities.

7. References Babus’Haq, R.F. & Probert, S.D., 1996. Combined heat-and-power

implementation in the UK: Past, present and prospective developments. Applied Energy, 53(1-2), pp.47–76.

Bergek, A. et al., 2008. Analyzing the functional dynamics of technological innovation systems: a scheme of analysis. Research Policy, 37(3), pp.407–429.

Bulkeley, H., Castán Broto, V., Hodson, M. & Marvin, S. eds. (2011) Cities and low carbon transitions. Abingdon, Oxon, Routledge.

Bulkeley, H. & Kern, K., 2006. Local Government and the Governing of Climate Change in Germany and the UK. Urban Studies, 43(12), pp.2237 - 2259.

Carlsson, B. & Stankiewicz, R., 1991. On the nature, function and composition of technological systems. Journal of Evolutionary Economics, 1(2), pp.93–118.

CHPA, 2011. Consultation on Reform of the Electricity Market: A response from the Combined Heat and Power Association, London: Combined Heat and Power Association.

Cook, I.R., 2009. Private sector involvement in urban governance: The case of Business Improvement Districts and Town Centre Management partnerships in England. Geoforum, 40(5), pp.930-940.

DECC, 2009. Heat and Energy Saving Strategy Consultation, http://hes.decc.gov.uk.

DECC, 2011a. Electricity Market Reform White Paper, http://www.decc.gov.uk/en/content/cms/legislation/white_papers/emr_wp_2011/emr_wp_2011.aspx.

DECC, 2011b. The carbon plan: Delivering our low carbon future, http://www.decc.gov.uk/en/content/cms/tackling/carbon_plan/carbon_plan.aspx.

DECC, 2012. The future of heating: A strategic framework for low carbon heat, http://www.decc.gov.uk/en/content/cms/meeting_energy/heat_strategy/heat_strategy.aspx.

Dyrelund, A. & Steffensen, H., 2004. Best practice in Danish district heating, Denmark: Rambøll. http://www.energymap.dk/Cache/65/65ae68be-e645-4a66-b5e4-2a4f4b757f02.pdf.

Ericson, K., 2009. Introduction and development of the Swedish district heating systems: Critical factors and lessons learned, http://www.res-h-policy.eu/downloads/Swedish_district_heating_case-study_(D5)_final.pdf.

Euroheat & Power, 2011. District Heating and Cooling: Country by Country 2011 Survey, Belgium: Euroheat & Power. http://www.euroheat.org/Publications-8.aspx?PID=46&Action=1&NewsId=12.

Foxon, T.J., 2007. The Rationale for Policy Interventions from an Innovation Systems Perspective. In J. Murphy, ed. Governing Technology for Sustainability. London: Earthscan, pp. 129–147.

Geels, F.W., 2011. The role of cities in technological transitions: Analytical clarifications and historical examples. In H. Bulkeley et al., eds. Cities and low carbon transitions. Abingdon: Routledge.

GLA, 2009. Powering ahead: delivering low carbon energy for London, Greater London Authority. http://www.london.gov.uk/sites/default/files/powering-ahead141009.pdf.

Grohnheit, P.E. & Gram Mortensen, B.O., 2003. Competition in the market for space heating. District heating as the infrastructure for competition among fuels and technologies. Energy Policy, 31(9), pp.817-826.

Hawkey, D.J.C., Webb, J. & Winskel, M., forthcoming. Organisation and governance of urban energy systems: district heating and cooling in the UK. Journal of Cleaner Production.

HCA, 2011. District Heating Good Practice Guide: Learning from the Low Carbon Infrastructure Fund, http://www.homesandcommunities.co.uk/district-heating-good-practice-learning-low-carbon-infrastructure-fund.

Hekkert, M. et al., 2007. Functions of innovation systems: A new approach for analysing technological change. Technological Forecasting and Social Change, 74(4), pp.413-432.

Helm, D., 2010. Infrastructure and infrastructure finance: The role of the government and the private sector in the current world, European Investment Bank, Economic and Financial Studies. http://ideas.repec.org/p/ris/eibpap/2010_005.html.

Hodson, M. & Marvin, S., 2010. Can cities shape socio-technical transitions and how would we know if they were? Research Policy, 39(4), pp.477-485.

IEA, 2005. A Comparison of distributed CHP/DH with large-scale CHP/DH, Paris: International Energy Agency

IEA, 2008. CHP/DHC country scorecard: United Kingdom, Paris: International Energy Agency. http://www.iea.org/g8/chp/docs/UK.pdf.

IEA, 2012. Golden rules for a golden age of gas, Paris: International Energy Agency. http://www.worldenergyoutlook.org/goldenrules/

IEA, 2011. World Energy Outlook, Paris: International Energy Agency. http://www.worldenergyoutlook.org/publications/weo-2011/.

Jacobsson, S. & Bergek, A., 2011. Innovation system analyses and sustainability transitions: Contributions and suggestions for research. Environmental Innovation and Societal Transitions, 1, pp.41-57.

Kelly, S. & Pollitt, M., 2010. An assessment of the present and future opportunities for combined heat and power with district heating (CHP-DH) in the United Kingdom. Energy Policy, 38(11), pp.6936-6945.

Leach, R. & Percy-Smith, J., 2001. Local governance in Britain, Hampshire: Palgrave.

LEP, 2007. Making ESCos Work: Guidance and Advice on Setting Up & Delivering an ESCo, London: London Energy Partnership. http://www.lep.org.uk/uploads/lep_making_escos_work.pdf.

Lovell, H. et al., 2011. Heat and the City workshop on Municipal Leadership and Organisation for District Energy: Workshop report, Edinburgh: Heat and the City. http://www.heatandthecity.org.uk/our_work/workshop_-_leadership_and_organisation_for_district_energy.

Malerba, F., 2002. Sectoral systems of innovation and production. Research Policy, 31(2), pp.247-264.

Markard, J. & Truffer, B., 2008. Technological innovation systems and the multi-level perspective: Towards an integrated framework. Research Policy, 37(4), pp.596-615.

Monstadt, J., 2007. Urban Governance and the Transition of Energy Systems: Institutional Change and Shifting Energy and Climate Policies in Berlin. International Journal of Urban and Regional Research, 31(2), pp.326-343.

Negro, S.O., Hekkert, M.P. & Smits, R.E., 2007. Explaining the failure of the Dutch innovation system for biomass digestion–A functional analysis. Energy Policy, 35(2), pp.925–938.

NVE, 2009. Veileder i utforming av konsesjonssøknad for fjernvarmeanlegg (Guidance on the design of a license application for district heating), Oslo: Norges vassdrags- og energidirektorat. http://nve.no/Global/Konsesjoner/Fjernvarme/Fjernvarmeveileder2009.pdf.

OECD, 1997. National Innovation Systems, Paris: OECD. http://www.oecd.org/dataoecd/35/56/2101733.pdf

Ofgem, 2011. High Level Summary of DG Forum Responses, London: Ofgem. http://www.ofgem.gov.uk/Networks/ElecDist/Policy/DistGen/Documents1/high%20level%20summary%20of%20DG%20Forum_published.pdf.

Pöyry Energy, 2009. The potential and costs of district heating networks: A report to the Department of Energy and Climate Change, Oxford: Pöyry Energy. http://www.decc.gov.uk/assets/decc/What%20we%20do/UK%20energy%20supply/Energy%20mix/Distributed%20Energy%20Heat/1467-potential-costs-district-heating-network.pdf.

Roberts, S., 2008. Infrastructure challenges for the built environment. Energy Policy, 36(12), pp.4563–4567.

Rotheray, T., 2011. Onsite generation and demand response – the role of the EMR. http://www.heatandthecity.org.uk/__data/assets/pdf_file/0010/71758/7_-_Tim_Rotheray_-_Onside_generation_and_demand_response_the_role_of_the_EMR.pdf.

Russell, S., 1993. Writing Energy History: Explaining the Neglect of CHP/DH in Britain. British Journal for the History of Science, 26(1), pp.33–54.

Rutherford, J., 2008. Unbundling Stockholm: The networks, planning and social welfare nexus beyond the unitary city. Geoforum, 39(6), pp.1871-1883.

Scottish Government, 2011. 2020 Roadmap for Renewable Energy in Scotland, http://www.scotland.gov.uk/Publications/2009/07/06095830/2020Routemap.

SDC, 2009. Renewable heat in Scotland, Sustainable Development Commission. http://www.sd-

commission.org.uk/publications/downloads/SDC%20Renewable%20Heat%20Report.pdf.

Summerton, J., 1992. District heating comes to town: The social shaping of an energy system, Linköping: Linköping University.

Toke, D. & Fragaki, A., 2008. Do liberalised electricity markets help or hinder CHP and district heating? The case of the UK. Energy Policy, 36(4), pp.1448–1456.

Truffer, B. & Coenen, L., 2012. Environmental Innovation and Sustainability Transitions in Regional Studies. Regional Studies, 46(1), pp.1-21.

Williams, J., 2010. The deployment of decentralised energy systems as part of the housing growth programme in the UK. Energy Policy, 38(12), pp.7604-7613.

Wilson, D. & Game, C., 2002. Local Government in the United Kingdom 3rd ed., Hampshire: Palgrave Macmillan.

Winskel, M. et al., 2006. Energy policy and institutional context: Marine energy innovation systems. Science and Public Policy, 33(5), pp.365-376.

Zero Carbon Hub, 2011. Allowable solutions for tomorrow’s new homes: towards a workable framework, http://www.zerocarbonhub.org/definition.aspx?page=9.