How do megaprojects influence institutional change?

Abstract

A growing body of research emphasizes the importance of institutions for megaproject performance. Extant studies focus on understanding the characteristics of favourable institutional environments and the governance structures that projects can adopt to navigate both favourable and unfavourable environments. This paper contributes to this research by foregrounding the link between megaprojects and changes in the institutional environments within which they are embedded. Building on a case study of the Thames Tideway Tunnel—the UK’s largest water and sewerage infrastructure project since the privatization of the industry in 1989—the paper presents two distinct ways that projects can interact with formal institutions: institutional compliance and institutional change. A more holistic understanding of the latter is essential because megaprojects can reconfigure production and consumption patterns to the benefit of some and the detriment of others. By contributing to changes in the institutional environment, megaprojects can further entrench these reconfigured patterns into long-enduring formal rules and practices.

I. Introduction

Megaprojects are large-scale, multi-year ventures costing US$1 billion or more and likely impacting society in transformative ways (Flyvbjerg, 2014). Given their cost, size, and complexity, megaprojects are subject to multiple institutional constraints, including legal, administrative, and regulative frameworks (Miller and Lessard, 2001; Levitt and Scott, 2017). Along with informal norms, these formal institutions define the so-called ‘rules of the game’ that organize all forms of interaction (North, 1990; Ostrom, 1990).

A growing body of project-related research emphasizes the importance of institutions for megaproject performance. Conceptual differences notwithstanding, a commonality across studies that foreground the institutional embeddedness of projects is that they examine how higher-level systems affect and are influenced by micro factors (Cattani et al., 2011; Jooste et al., 2011). As such, this work advances earlier portrayals of projects as ‘lonely islands’—without linkages to their surroundings (Engwall, 2003). Indeed, these linkages are apparent in the necessary combination of policy work and legislation that megaprojects require, the budgetary bargaining between elected officials and departments, negotiations with local government, agencies, pressure groups, and communities, and procurement of national and international firms (Levitt and Scott, 2017).

Extant project-related research highlights the importance of institutions that facilitate legitimacy, trust, and capacity (Casady et al., 2020). For example, in the case of public–private partnerships (PPPs), project failure is sometimes attributed to the absence of a conducive institutional environment. More specifically, project performance is affected by a lack of public-sector capacity and political will, coupled with perceived legitimacy and trust issues between the public and private sector (Mahalingam, 2010; Mahalingam et al., 2011; Jooste and Scott, 2012; Soecipto and Verhoest, 2018). Therefore, governments are tasked with creating enabling environments by developing formal relationship standards and procedures, creating regulatory and legal frameworks, and ensuring that legitimate policy decisions are appropriately formulated (Hult and Walcott, 1990; Brinkerhoff and Brinkerhoff, 2011; Matos-Castaño et al., 2014). In turn, effective projects fit rather than challenge established norms and values.

There is less written about how institutions change over time, particularly how megaprojects contribute to institutional change (Morris and Geraldi, 2011; Matos-Castaño et al., 2014). Instead, megaproject-related institutions are typically taken as given, and changes in the institutional environment are attributed to discrete government action. There is, therefore, scope to examine how projects impact institutions with greater attention to the link between micro-level project dynamics and macro-level changes (Miller and Lessard, 2001; Morris and Geraldi, 2011). Such an approach can contribute to our understanding of how much megaprojects contribute to creating new rules, beliefs, and practices.

Against this backdrop, the primary question that this paper addresses is: how do megaprojects influence institutional change? A case-based research approach is followed to address this question. Such an approach allows for an in-depth examination of the phenomena, tracing processes in their natural contexts and ‘describing and explaining the temporal sequence of events’, which supports theory development (Van de Ven and Huber, 1990, p. 213; Flyvbjerg, 2006). Case studies are especially beneficial in areas with little previous research (Eisenhardt, 1989; Yin, 1994). This is the case with megaproject development; revelatory case studies are unusual in this context because such projects are undertaken infrequently and typically give only limited research access.

The paper presents a single case study of London’s Thames Tideway Tunnel (TTT). The TTT is a 25 km sewer under the tideway of the Thames River that is estimated to cost £4.2 billion (2016 prices). Built to minimize sewage spills in the Thames, the TTT project is the largest water and sewerage infrastructure project in the UK since the industry was privatized in 1989. Developing the TTT project called for a novel approach comprising a coherent financing model, project structure, and regulatory framework to cope with the size, cost, and risks associated with the tunnel (NAO, 2014).

The initial source of data collection was semi-structured interviews with industry participants involved in the Thames Tideway Tunnel. However, the case is primarily based on secondary data used to triangulate interview findings and develop a convergent pool of evidence. Secondary data sources included archival documents and other official publications internal and external to project affiliated organizations, including the Water Services Regulatory Authority (Ofwat), Thames Water, Thames Tideway, the Department for Environment, Food and Rural Affairs (Defra), National Audit Office (NAO), Agilia Infrastructure Partners, and The Infrastructure Forum (TIF).

By foregrounding the link between megaprojects and changes in the institutional environment, the case demonstrates the dynamics of the latter. It also shows how multiple actors within the project environment influence institutional change beyond the national government. This conceptualization contributes to existing research by highlighting two distinct ways projects can interact with formal institutions: institutional compliance and institutional change. Institutional compliance entails adherence to the existing ‘rules of the game’, including that a megaproject is developed within legal bounds. Institutional change moves beyond a static conception of the ‘rules of the game’ and reflects the process by which megaprojects and institutions can develop in tandem.

The paper proceeds as follows. Section II provides an overview of the institutional setting within which the case study of the Thames Tideway Tunnel, presented in section III, is embedded. Section IV sets out the main findings from the case before the policy implications are discussed in section V.

II. Institutional setting

In the period leading up to and after the Second World War, the water industry in England and Wales was highly fragmented. It included over 1,000 local bodies involved in water supply and around 1,400 responsible for sewage disposal. A severe drought in 1959 and flooding events in 1960 spurred the government to coordinate water resource planning and consolidate local authorities (Ofwat, 2006). The Water Resources Act 1963 was subsequently passed.

Given that successive governments had chronically underinvested in water services, problems persisted. Water quality remained poor, and beaches and rivers were severely affected by sewage. Future demand forecasts further showed that these issues would not abate in the absence of radical change. As a result, the government restructured the water industry in the late 1960s and early 1970s (Ofwat, 2006). The Water Act 1973 established ten new regional water authorities to manage water resources and the supply of water and sewerage services.

The 1970s and early 1980s were a period of high inflation in the UK, and the government responded with tight fiscal controls (Nelson and Nikolov, 2003). This directly impacted regional water authorities because they were required to meet their investment needs by borrowing from the central government. Regional water authorities also relied on revenue from services provided to customers for capital investment, and the government was unwilling to increase user charges. Therefore, budget constraints, already high levels of inherited debt, and a lack of access to debt finance meant that water authorities were unable to meet the industry’s capital maintenance and investment requirements (Ofwat, 2006).

In response to growing public dissatisfaction with the regional water authorities, heightening environmental awareness and stringent legislation, the government introduced the Water Act 1983. Underlying the legislation was the assumption that efficiently run water utilities providing prescribed service standards at least cost would best serve customers (Young, 2001; Ofwat, 2006; Helm, 2020). The Water Act 1983 significantly reduced the role of local government in decision-making and paved the way for privatization by allowing companies to act more commercially with the scope of accessing private markets (Macrory, 1990).

Despite legislation, the central government continued to exercise control over the authorities’ borrowing, and regional water authorities struggled to borrow from private markets (Ofwat, 2006). Little improvement in environmental performance and a lack of investment meant that many pollution incidents continued to occur. As a result, the UK was in breach of several European Commission (EC) Directives. In the mid-1980s, the EC decided to start prosecution proceedings against the UK government for non-compliance. These proceedings underscored the need for further significant capital investment and pollution control (Ofwat, 2006).

The government was unwilling to provide additional public finance to meet the demand for capital investment by the water authorities (Ofwat, 2006). However, there was also a prevailing policy that favoured privatization to secure efficiency; British Telecom and British Gas had been privatized in 1984 and 1986, respectively (Holder, 1998; Young, 2001). As such, privatising the industry was deemed a viable way forward.

The proposals for the privatization of the water industry differed in three fundamental respects from those of the gas and telecoms industries. First, privatization would result in ten Water Authorities rather than one (as in gas and telecoms). Second, water and sewerage industries would also have responsibilities beyond service provision to protect the environment. Third, natural monopoly conditions were more prevalent in the water and sewerage industry because it consisted of local and regional monopolies with no national distribution network (Littlechild, 1988; Ofwat, 2006).

Nevertheless, many of the regulator provisions were copied from the 1984 Telecommunications Act across the four utilities and subsequent legislation (Thatcher, 1998). With the passing of the Water Act 1989, the government privatized the water industry and introduced a new regulatory and legislative framework. Assets and personnel of the ten water authorities were transferred into limited companies, which were floated on the London Stock Exchange to raise capital. Newly formed water companies received a one-off injection of public capital, a write-off of significant government debt, and the provision of capital tax allowances. In addition, to ensure that the interests of customers and the environment were secured, the government separated the roles of regulation and provision of water and sewerage services.

Post-privatization water companies have had greater access to capital markets. Nevertheless, the extent to which this access and companies’ initially ungeared balance sheets have resulted in more investment is mixed. The prevalence of financial engineering across the industry has sometimes benefited shareholders and executives at customers’ expense (Helm, 2020).

Today, all water companies in England hold an appointment as a water undertaker (or sewerage undertaker, as relevant) subject to licence conditions set by Ofwat. Three separate and independent bodies regulate the water industry—Ofwat, the Environment Agency, and the Drinking Water Inspectorate. Together, these bodies ensure that water companies comply with a range of water quality, environmental, and economic legislation in addition to adhering to their licence conditions.

The regulatory framework aims to ensure that consumers receive high service standards at a fair price. Defra guides regulators by setting England’s overall water and sewerage policy framework. Defra’s mandate includes standard-setting, drafting legislation, and creating special permits (e.g. drought orders). In addition, given that until Brexit, European Union (EU) law took effect in English law, water and sewerage companies have had to comply with several different Acts of Parliament and European Directives. These cover economic regulation of the sector, water supply, sewerage services, drinking water quality, environmental standards, customer service, flood and drought protection and adaptation.¹ The Consumer Council for Water (CCW) represents customers within the water and sewerage sectors and Natural England is the government’s advisor on the natural environment (Ofwat, 2006).

As the economic regulator of the water and sewerage sectors, Ofwat regulates the charges consumers pay to the private companies responsible for planning and building monopoly water infrastructure in the UK. In 2013, the Water Industry (Specified Infrastructure Projects) (English Undertaken) Regulations 2013 (SIP) were passed, expanding Ofwat’s role to include the authority to ‘specify a project’ as an infrastructure project that water companies must put out to tender instead of delivering in house. As is outlined in further detail in the TTT case, the passing of SIP regulations has meant that Ofwat has the authority to provide so-called infrastructure providers, which are procured by existing water companies, with project licences to finance, construct, and operate large-scale infrastructure investments (Ofwat, 2015a).

According to the SIP Regulations, Ofwat (and Defra) can make such a specification if a project satisfies two conditions. First, the infrastructure project must be of a size and complexity that can threaten a water company or undertaker’s ability to provide services to its customers. Second, specifying the project would likely result in better value for money than if the project were not specified, considering charges to customers and any government financial assistance (Ofwat, 2015a). These regulations,

were introduced to help the delivery of necessary, large or complex water or sewerage infrastructure projects within England. They were designed to help contain, minimise and isolate the risks associated with the delivery of these projects from customers, undertakers and UK taxpayers, while also providing value for money and keeping customer bills as low as possible. (HL Deb., 2020)

The SIP regulations are based on the Thames Tideway Tunnel (TTT), the largest water and sewerage infrastructure project in the UK since the privatization of the industry in 1989. It is, therefore, the first and, to date, the only project under this regulatory framework. While it is yet to be seen, for some, the delivery model used for the TTT and enabled by the SIP regulations has the potential to make an immense scale of capital—in the order of hundreds of billions of pounds—available for infrastructure investment across multiple sectors (TIF, 2020). Examining the development of the TTT demonstrates how the megaproject contributed to institutional change. That actors at multiple levels shaped this dynamic process challenges the conceptualization of institutional change in the context of megaprojects as the result of discrete government action.

III. Case study: the Thames Tideway Tunnel

The Thames Tideway Tunnel (TTT) is the largest water and sewerage infrastructure project in the UK since the privatization of the industry in 1989 (NAO, 2017). Due to be completed in 2027, the tunnel will be a 7.2-metre wide and 25-kilometre-long sewer under the tideway of the Thames River in London, addressing untreated wastewater entering the Thames River (Defra, 2015). The expected cost of the tunnel is £4.2 billion (2016 prices), which is being paid for by Thames Water’s 15 million wastewater customers through their bills. Additional customer charges are estimated to be at most £20–25 annually (2015 prices) for the average household (NAO, 2017).

A key driver in the decision to proceed with the TTT project was that the level of sewage discharge into the River Thames contravened the European Union’s Urban Waste Water Treatment Directive. The Directive aims to protect the environment from wastewater discharges. In 2010, the European Commission launched enforcement action against the UK for non-compliance due to the frequency of spills. In 2012, the Court of Justice of the European Union found that the UK was in breach of the EU Directive. This ruling meant that the UK was and continues to be at risk of infraction fines if the problem is not addressed. Fines include a lump-sum penalty and daily fines, estimated to total hundreds of millions of pounds each year until compliance is achieved (NAO, 2017).

Upon completion, the TTT project is expected to reduce spill events from 40–107, depending on the year (39.5 million cubic metres per year), to a maximum of four per year (2.35 million cubic metres per year) (Thames Water, 2006; NAO, 2017). Hence, there are several reasons for the government’s endorsement of the project: fewer sewage spills, improving the tideway’s environmental quality, and ensuring sufficient strategic sewer capacity to accommodate London’s growth for at least the next hundred years (NAO, 2017). The project was subsequently developed using a novel approach that required complementary institutional arrangements.

(i) Traditional economic regulatory framework

Thames Water is the UK’s largest water and wastewater service company responsible for public water supply and wastewater treatment in most of Greater London and other parts of England. Under England’s traditional economic regulatory framework, the company would have developed the TTT project in house (UK Government, 2007). Further, to raise private finance, Thames Water would rely on the industry-wide weighted average cost of capital (WACC) or return to investors, as set by Ofwat every 5 years. Ofwat sets the WACC so that the return on capital is no more than necessary for an efficiently run company to get the financing it needs from capital markets.

Given the large scale of capital expenditure that the TTT project would require, Thames Water deemed it too costly to undertake alone. The utility’s already high gearing, the ratio of the company’s debt-to-equity, coupled with the risk and uncertainty related to the project, could significantly impact Thames Water’s credit rating, cost of debt, and cost of equity (Moody’s Investor Service, 2012). Ultimately, this would also threaten Thames Water’s ability to provide services for its customers (NAO, 2014). The option of government support as a risk management strategy was explored. However, limiting it to a single project in Thames Water’s portfolio would be challenging (NAO, 2014; UK Government, 2014). As such, Thames Water ruled out even limited financial involvement in the project.

Defra considered using public finance to construct the TTT project and then sell it after commissioning. However, this was also deemed problematic as it ‘could require legislation to take the project outside the existing regulatory framework’ (NAO, 2014). Therefore, in a first for the privatized English and Welsh water sector, The Thames Tideway Tunnel is being delivered by a specially created company. The project also benefits from a contingent support package provided by the government in its construction phase (NAO, 2017). Delivering the project in this way saw the introduction of a novel structure for the TTT project and complementary institutional mechanisms to attract private finance at a reasonable cost.

(ii) A parallel regulatory regime

Developing the TTT project called for a novel approach comprising a coherent financing model, project structure, and regulatory framework to cope with the size, cost, and risks associated with the tunnel (NAO, 2014). The SIP regulations were crucial in this case because they enabled the Secretary of State or Ofwat to specify certain large and complex infrastructure projects to be delivered by infrastructure providers or ‘special purpose vehicles’. Per the regulations, Thames Water, as the undertaker, could carry out prescribed and listed preparatory works but would then be required to tender the specified project through competitive procurement (UK Government, 2014).

Ofwat could issue a project licence to an infrastructure provider within SIP regulations. This licence would allow the provider to own the TTT project, coordinate its financing and construction, and ultimately operate it. Ofwat would also regulate the infrastructure provider in the project’s construction and operational periods per its duties under the Water Industry Act 1991. The TTT project predates the SIP regulations, and the model forms the basis of their formation.

While some aspects of the TTT project and its regulatory framework are common with standard price control regimes, others are unique. First, the TTT project is a single project with the risks of the construction programme ring-fenced within the designated and licenced infrastructure provider’s company. Second, the regulatory asset base (RAB) is applied to a stand-alone greenfield asset rather than to an established utility as is typically the case. Third, a bid-related weighted average cost of capital (bid WACC) would apply to the infrastructure provider and be fixed for the duration of the construction period (instead of being set every 5 years as in the traditional regulatory framework). To that end, investors will have certainty over some cash flow elements up to and including 31 March 2030. Fourth, the commercial framework used for the TTT project has significant risk-mitigating elements, including a government support package (CEPA, 2015). These unique features are discussed further below.

A single project with the risks of the construction programme ring-fenced within an Infrastructure Provider company

In compliance with SIP regulations and consultation with the government and Ofwat, Thames Water undertook preparatory work for the TTT project. This work included preparing the reference design and planning application for the project, conducting a land acquisition programme (as appropriate), and conducting certain enabling works. It also included initiating the procurement of construction contracts (which would ultimately be awarded by the designated and licensed provider), putting a corporate architecture in place, and ensuring that the project had sufficient delivery expertise and a management team (UK Government, 2014). Of the total £4.2 billion (2016 prices) cost of the project, Thames Water’s preparatory work represented one-third or £1.4 billion (2016 prices) of the investment excluding financing costs. The infrastructure provider would deliver the remaining two-thirds or £2.8 billion (2016 prices).

In compliance with the Utilities Contracts Regulations 2006, Thames Water put the TTT project out to tender by running a competitive procurement for an infrastructure provider. Separate from Thames Water, the selected provider would be responsible for delivering the project, including its financing, subject to the conditions of its Ofwat-issued project licence (NAO, 2017).

In July 2015, Thames Water announced Bazalgette Tunnel Limited as the preferred bidder. The company is owned by a consortium of pension funds and long-term investors represented by Allianz, Amber Infrastructure (representing International Public Partnerships and Swiss Life), Dalmore Capital, and DIF Capital Partners. Following public consultation, in August 2015, Ofwat awarded Bazalgette Tunnel Limited a project licence. Then, as the official TTT project infrastructure provider, Bazalgette Tunnel Limited signed the project documentation and three main construction contracts with three consortia to construct the tunnel (UK Government, 2015). The structure of the project is illustrated in Figure 1.

Institutional change was necessary to accommodate the novel financing model, project structure, and regulatory framework in the water sector required to advance the TTT project. The project’s estimated cost of £4.2 billion (2016 prices) represented almost half of Thames Water’s regulated asset base. Given the project’s large size and considerable risk, it could impact the company’s investment rating and threaten its ability to deliver services to customers (NAO, 2014). The introduction of the SIP regulations made it possible for the utility to put the project out to tender.

Zhivov (2018, p. 8) describes the resulting approach as allowing for

measures that are capable of providing incentives for the private sector to finance and deliver this large new infrastructure project efficiently and at a reasonable cost while discouraging the moral hazards that can come with relieving private parties of risk.

The novel use of the regulatory asset base (RAB) model was at the heart of this approach, as discussed further below.

Applying the regulatory asset base (RAB) model to a single greenfield asset

The RAB model is the foundation of price-control regulation in the UK. It is used to regulate private-sector monopolies’ infrastructure assets such as water, gas, and electricity networks and aviation. The RAB is an estimate of the total value of the capital assets employed by the utility in performing its regulated functions. Said differently, it is the book value of the net assets in a regulated business, as recognized by the regulator and therefore eligible for a return (Newbery et al., 2019). Under a RAB approach, the regulator sets prices to allow for a utility’s operating expenses, an allowance for the depreciation of capital assets, and a return on the total value of the assets employed.

Until the TTT project, the RAB model was only applied to established utilities. Therefore, the TTT project is the first example of a RAB model for a single greenfield asset. Multiple characteristics of the TTT model went beyond the generic RAB model. For example, because there is no pre-existing asset base with a greenfield development, expenditure incurred in connection with the project is added to the RAB as construction progresses. Further, unlike in the case of established water utilities, project expenditure for the TTT is not subject to any form of regulatory efficiency assessment. Instead, there is a pre-agreed overall project cost and risk envelope, which the regulator approves. The infrastructure provider is then incentivized to keep costs low through a pain/gain share mechanism, i.e. the TTT provider gets to keep 30 per cent of any outperformance relative to the target cost but is accountable for 40 per cent of any overspend (NAO, 2017). The RAB model also provides a structure for risk-sharing between the market (including investors, contractors, and the developer), consumers, and the taxpayer (Newbery et al., 2019).

The RAB model relies on the availability of a customer base that can afford to make a long-term, stable contribution towards the funding of the infrastructural asset. In terms of the TTT project, Thames Water’s customers are funding the project through additional charges (NAO, 2017). These charges align with the revenue estimation methodology set out by Ofwat and captured in the project’s licence. For investors, this represents a long-term stable yield. Investors also benefit from the ongoing scrutiny of an expert regulator over the entire project life-cycle. Therefore, the project was an attractive investment opportunity and, per the NAO, the project’s delivery model ‘help[ed] secure the lowest cost of capital and keep costs down for customers’ (NAO, 2014).

A market-related rate of return

Under the traditional regulatory framework, Ofwat sets a WACC for the water industry for a 5-year period. The regulator does this by reviewing market evidence and exercising judgement to determine the efficient rate of return to investors for particular infrastructure. The WACC must allow investors an appropriate rate for the risks they face while delivering value for money for customers. Ofwat also has to secure that an efficiently run company can finance its functions. The regulator does this by ensuring that each company can earn a post-tax return at least equal to the cost of capital and a requirement to secure that the revenues, profits, and cash flows allow each company to raise finance on reasonable terms in the capital markets.

In contrast to the above approach, Thames Water carried out a financing competition for the TTT project, which required bidders to submit their required cost of capital. This WACC would then be used to calculate the allowable returns for the infrastructure provider during the project’s construction period. In essence, the bidder prepared to accept the lowest WACC would be best placed to win the tender.

The tender for the Thames Tideway Tunnel resulted in a bid-related WACC, which was considerably lower than the equivalent WACC for an established utility at the time. Bazalgette Tunnel Limited’s winning bid offered a WACC of 2.497 per cent, subject to the Project Licence terms, whereas the water industry WACC set by Ofwat was 3.7 per cent. The difference suggested that investors were willing to accept a relatively lower return.² Given that Thames Water customers will ultimately repay the project financing through an additional bill charge on behalf of Bazalgette Tunnel Limited, the lower WACC also means that they pay relatively less. Ofwat estimated that customers would pay at most £20–£25 more on their water bills, which is 70 per cent under initial estimates of £70–£80 (Ofwat, 2015b).

An additional project-specific adaptation to the regulatory framework was a modified incentive regulation framework developed by Ofwat. The modification was such that the 2.497 per cent bid WACC would apply as the regulated rate for the duration of the project’s construction period (Ofwat, 2015b). Beyond this period, and during project operation, Ofwat will set an allowed cost of capital, similar to the approach it applies to other water and sewerage company licensees (Ofwat, 2015b; CEPA, 2015). As such, even though the construction period will last for more than 10 years, it will be treated as a single regulatory period, and the WACC will be fixed until 31 March 2030 (exceeding the standard 5-year regulatory period). Following construction completion, the regulator will set the WACC at periodic intervals. For the TTT project, this will fit into the standard 5-year asset management periods used to regulate the sector by Ofwat.

By reducing uncertainty and providing predictability, the regulatory adaptations helped investors minimize any regulatory risk they might have faced during construction if subjected to a variable rate of return (NAO, 2014). As detailed in the sections that follow and summarized in Table 1, contributions from actors across multiple levels transformed the TTT project from a project that was potentially high risk for Thames Water to a relatively low-risk, government-backed, and inflation-protected investment for TTT project investors.

Multilevel-multiparty institutional change

The institutional change enabling the TTT project resulted from tiered and calibrated action. At the operational level, Thames Water, with oversight from Ofwat, developed the detailed plans and costings for the TTT project before commencing the competitive processes for the infrastructure provider and construction contractors. When the construction competition commenced, the designs for the tunnel were so detailed that Thames Water produced a cost estimate with an 80 per cent probability of actual costs being lower than the estimate (i.e. a P80 estimate) (Ofwat, 2015b). Providing detailed design and costing was deemed crucial for reducing risk and uncertainty and, therefore, reducing the need for construction bidders to have large contingency budgets. This approach differs from the ‘design and build’ contracts regularly used for PPPs, where bidders are required to develop detailed designs for the project (Yescombe, 2014; Zhivov, 2018).

Incentives were put in place to share construction risks between investors and contractors, in the first instance; between investors and customers, in the second instance; and with the government as the final resort. In terms of the last, the government committed to a government support package that would protect the infrastructure provider from responsibility for difficult-to-quantify, high-impact, low-probability risks, and uncertainties (Ofwat, 2014; NAO, 2014). Among other things, the support package makes provision for support if cost overruns exceed 30 per cent or if the infrastructure provider is unable to access capital markets due to economic or political events. Following Bazalgette being named the preferred bidder for the project, the European Commission assessed the support package for compliance and deemed it compatible with the European Union’s internal market (UK Government, 2015).

At the regulatory level, Ofwat provided various ‘regulatory shields’ to allow for a market-tested rate of return while minimizing uncertainty for the infrastructure provider. In addition to fixing the WACC until 31 March 2030, Ofwat’s adaptations ensured that the infrastructure provider would have sufficient funds to meet its debt servicing costs. To that end, the regulator will allow investors to access some returns during the construction period and make adjustments to allow for the cost of debt. As described by Ofwat, these regulatory changes entailed ‘adopting and adapting a number of elements of the current regulatory framework that are familiar to investors seeking utility-like risk profiles’ (Ofwat, 2014).

Against this backdrop, the process of institutional change was not the result of discrete government action. Instead, it was a dynamic process shaped by actors at multiple levels within the TTT project ecosystem. As detailed in the discussion section, the TTT project has contributed to shaping institutional change in the water industry that will potentially have significant long-term implications. Such changes include an amendment to the SIP regulations passed by Defra in 2014, giving Ofwat the authority to include conditions in an infrastructure provider’s project licence. In part, Ofwat did this to enable infrastructure providers to raise any licence concerns with the Competition and Markets Authority, like other water companies. Alternatively, concerns would have to go through an appeal with the High Court.

In 2020, a potentially more significant amendment was passed, which allowed Defra to remove the ‘sunset clause’ from the SIP regulations. The clause made a provision for Parliament to review and decide on the merits of the regulations after a period of 5 years, which will no longer be required. With the amendment, the SIP regulations will continue in force as part of the regulatory framework of the water industry with direct implications for the TTT project and qualifying projects in the water industry at large. Without the amendment, the 2013 SIP regulations, developed in tandem with the TTT project, would have expired on 27 June 2020.

IV. Discussion

(i) Institutional compliance and institutional change

This paper contributes to the growing body of research that foregrounds the importance of institutions for megaproject performance (Cattani et al., 2011; Jooste et al., 2011; Levitt and Scott, 2017; Gil et al., 2019; Casady et al., 2020). Contrary to the dominant framing, the TTT case demonstrates that the process of institutional change is not always external to the megaproject development process. To that end, the case highlights two distinct ways projects can interact with formal institutions: institutional compliance and institutional change. Compliance entails adherence to the existing ‘rules of the game’, such as ensuring that megaprojects are developed within legal bounds. In the case of the Thames Tideway Tunnel, this was evident given that a key driver for the project was to comply with the EU’s Wastewater Treatment Directive. Other examples include the procurement of the infrastructure provider following the Utilities Contracts Regulations, and developing the government support package in compliance with European Commission guidelines.

Creating a parallel regulatory framework to accommodate the TTT project demonstrates how projects contribute to institutional change. Crucially, change, in this case, was spurred by Thames Water’s inability to take on the scheme, given that the risk and uncertainty associated with the project could significantly impact its credit rating, cost of debt, cost of equity, and ultimately, its ability to provide services for its customers. Hence, factors contributing to change included a challenge to the existing system and the urgency to get the project under way given the ruling by the Court of Justice of the European Union.

In the case of the TTT project, change was driven by Ofwat, Thames Water, and Defra, rather than being the result of discrete government action. Together these organizations aimed to create a coherent financing model, project structure, and regulatory framework to cope with TTT project’s size, costs, and risks. The resulting dynamic process of change aligns with the notion that actors constantly enact and (re)produce institutions through their various activities (Scott, 2001).

It is worth noting that not all institutions are subject to change, nor is the process of change costless or frictionless. Indeed, when considering public finance as an option for the TTT project, the government deemed it problematic because ‘it could require legislation to take the project outside the existing regulatory framework’ (NAO, 2014, p. 25). Moreover, in the post-implementation review of the SIP regulations, Ofwat found that ‘the initial delivery of the regulatory framework was significantly more complex and time-consuming than anticipated’ (House of Lords Debates, 2020).

(ii) Megaprojects as demonstrators: a mechanism for institutional change

The comprehensive modification of the formal institutional framework of regulation in Britain in the 1980s and 1990s shows that institutional reform is possible. Privatization and the establishment of new regulatory regimes for the gas, electricity, water, and telecoms utilities modified the government’s capacities and role in policy-making. The result was a major realignment of the boundaries between the state and private sectors of the UK and a break with the past across multiple industries (Thatcher, 1998; Young, 2001). Typically, however, institutional change is confined to crises, or it is slow and takes place incrementally (Thatcher, 1998). Change is especially prolonged when institutions do not already exist, or existing institutions must be readapted in response to rapid technological and economic change (Ostrom, 2005).

The development of the SIP Regulations in tandem with the TTT project is arguably an example of incremental institutional change. The project served as a test case for developing a coherent financing model, project structure, and regulatory framework. As can be expected, given the novelty of the approach used to develop the TTT project, it has undergone multiple reviews and attracted much interest. As relates to institutional change, this aligns with the notion that institutional change is an ongoing process, and institutions have to be underpinned by stakeholders with incentives and influence to see change enacted (North, 1990; North et al., 2009).

To that end, in 2017, the NAO reviewed the TTT project and stated that the project would be successful if the tunnel delivered Defra’s strategic objectives while risks and costs were well-managed. However, the NAO also found that Thames Water’s analysis of alternative tunnel options was not sufficiently detailed; its costing of alternatives was not independently scrutinized, and combinations of alternative technologies were not appraised. For the NAO, correcting for inaccurate predictions could have resulted in a smaller, lower-cost tunnel. Defra, in turn, responded that based on Thames Water’s previous estimates, a tunnel of smaller diameter would not have cost significantly less and would have carried a greater risk of non-compliance and fines from the European Commission. Defra also put forward that the chosen size would offer greater certainty that the tunnel will be ‘future-proof’ (NAO, 2017).

Likewise, as was required, Defra undertook a 5-year review of the SIP regulations in 2020. Ahead of the review, Defra carried out consultations and the general finding was that the regulation effectively protected the capital investment of the undertakers while reducing risks and costs. Crucially, of the eight actors consulted, the five respondents included Ofwat, Thames Water, Bazalgette Tunnel Ltd, Bazalgette Tunnel investors, and the Consumer Council for Water (CCW). Except for the CCW and Ofwat, all the other actors had a financial interest in the project. Arguably, they would therefore have an incentive to review the legislation positively and support its continued use. On the contrary, the CCW observed that customer handling on the project was insufficiently financed and, therefore, ineffectively managed.

Another review was carried out in 2020 by TIF, which brings together key infrastructure players. The Forum likened the potential scale of capital resources made available by the TTT model to a potential infrastructure investment revolution (akin to both privatization and the introduction of PPPs). However, the report contended that the suitable implementation of the model would require that several weaknesses be addressed. For example, it is challenging to market test the cost of capital given the lack of suitable comparators for the cost of equity. In addition, the model can potentially promote a bias towards capital investment over maintenance, given the prospect of returns. Finally, private companies may opt to diverge from the industry-wide WACC and take on higher levels of debt relative to equity than is deemed sustainable. Although ‘efficient in the short term, [such an approach can] prove unsustainable and inefficient in the longer term on the ability of the business to respond to changing trading conditions and risks, as well as providing windfall gains to equity investors’ (TIF, 2020, p. 14).

The issues raised in the NAO, Defra, and TIF reviews highlight the importance of independent regulatory bodies and independent scrutiny more generally. With increasing financialization, public interest can be subject to ‘the increasing role of financial motives, financial markets, financial actors and financial institutions in the operation of domestic and international economies’ (Epstein, 2005; see also Young, 2001; Loftus and March, 2019). The Consumer Council for Water reporting that customer handling on the TTT project was ineffectively managed because it was insufficiently financed suggests that distributional considerations are not a standard feature of the commercial world.

Suppose we accept that megaproject-related institutional change is a dynamic and potentially system-wide process. In that case, in addition to understanding the content of regulation, it will become increasingly important to understand the interests and incentives of those influencing the ‘rules of the game’.

(iii) Wider implications

In new institutional economics, ‘actors’ (individuals or organizations) can be distinguished from the institutional ‘rules of the game’ that constrain their actions. In this context, Schlager and Ostrom (1992, p. 250) define rules as ‘generally agreed-upon and enforced prescriptions that require, forbid, or permit specific actions for more than a single individual’. Similarly, the ‘rules’ developed in tandem with the TTT project have had a far-reaching impact beyond the project itself.

In the water sector, for example, the success of the TTT procurement in delivering a significant value for money investment for wastewater customers encouraged Ofwat to move forward with Direct Procurement for Customers (DPC). Ofwat has drawn on the TTT project as an example of the potential benefits of directly procuring infrastructure providers and how raising project financing through a competitive process can lead to a lower cost of capital, driving cost savings for customers. To that end, the regulator has introduced the DPC approach, which requires water companies to put out large infrastructure projects to tender when contracting a third party for delivery, is likely to offer better value for customers. While the TTT project is the first and only to have been carried out using this approach to date, albeit with some modifications, Ofwat has identified four large or complex infrastructure projects currently in development that may benefit from being specified per the SIP regulations over the next 10 years (Ofwat, 2017).

In the energy sector, Ofgem, the energy regulator, is considering a Special Purpose Vehicle model as an option to increase competition in onshore electricity transmission assets. In exploring this model, the energy regulator is monitoring the developments in Ofwat’s rollout of the DPC for potential learning. In terms of similarities between the two cases, Ofgem states that ‘both models involve the incumbent infrastructure operator tendering out the construction, operations, and financing to a third-party provider, and the regulatory arrangements are established through the geographical incumbent’s licence, rather than awarding a new licence’ (Ofgem, 2018, p. 9).

In nuclear power generation, on 26 October 2021, the UK government published a Nuclear Energy (Financing) Bill, which makes provision for the implementation of a regulated asset base model for nuclear energy generation projects. The legislation paves the way for applying a model similar to that used on the TTT project on a £20 billion nuclear power plant at Sizewell in Suffolk. The Department for Business Energy and Industrial Strategy (BEIS), the sponsoring department, has concluded that by providing regulated returns to investors, a RAB model can potentially attract new investors to new nuclear projects and reduce their cost of finance, thereby reducing consumer bills and maximizing value for money for consumers and taxpayers. BEIS has also put forward that ‘a RAB approach could present a sustainable and value for money model for funding new nuclear projects’ (BEIS, 2019, p. 9).

The RAB model differs from the contract for difference (CfD) approach previously used to finance nuclear power generation. The critical difference between the two approaches is that with the CfD, consumers do not start paying until the power station is up and running, and construction risk sits with the developer. Contrastingly, with the RAB model, consumers share in the costs from the start, reducing the amount of interest owed on loans. The RAB model also enables some level of risk-sharing between investors and consumers while maintaining the incentive on the private sector to minimize the risk of cost and schedule overruns.

Beyond the above examples, the TTT model has potentially broader applicability across energy, transport, telecoms, environment, and accommodation. Its wider significance is that different aspects of the model can be used for investment programmes, the development of highly repeatable lower-risk assets, and higher-risk ‘one-off assets’ (TIF, 2020). As the model is refined, a range of constraints may limit the eventual number of viable projects, including that new applications require changes in primary legislation. As in the case of nuclear generation, for some projects, contributing to the creation of said legislation may well be part of the project development process.

V. Policy implications

Megaprojects can boost economic potential in the short and long term. In the short term, a significant increase in capital expenditure on labour-intensive infrastructure projects, for example, can stimulate the economy by creating jobs and supporting growth. In the medium to long term, improved access to related services such as energy, transport, water, and digital connectivity can boost productivity. Crucially, given their cost, size, and complexity, megaprojects can also reconfigure economic activity to the benefit of some and the detriment of others. By contributing to changes in the institutional environment, megaprojects can further entrench reconfigured consumption and production patterns into long-enduring formal rules and practices.

The privatization of the water industry gave water utilities long-term access to capital markets, with the view that they would fund much-needed investment programmes. Similarly, the broader application of the TTT model has the potential to make a significant scale of capital—in the order of hundreds of billions of pounds—available for infrastructure investment across multiple sectors (TIF, 2020). While the benefits of such interventions are potentially immense, the need to carefully balance the interests of consumers and investors cannot be taken for granted. Policy-makers, therefore, have a role to play in setting clear public and societal expectations for the regulatory system.

There is almost no credible empirical research to definitively address questions of water industry performance since privatization (Helm, 2020). Similarly, there is no comprehensive evidence base of the profits made and losses incurred by contractors, financiers, and subcontractors over the 30 years of the Private Finance Initiative (PFI) in the UK (HM Treasury and IPA, 2019; TIF, 2020). Consequently, it is challenging to validate the performance of the £90 billion PFI programme spanning several hundred projects. Fostering a culture of disclosure, developing a repository of project-related information, and tracking projects over time can provide a deeper understanding of the consequences of new institutional arrangements over and above those intended.

I am grateful to Simon Quinn and an anonymous reviewer for helpful comments on an earlier draft of this article and to Amar Qureshi and Diane Coyle. All and any errors remain mine.

Footnotes

References