The impact of the global financial crisis on public private partnerships: a UK perspective
This chapter examines how Public Private Partnerships (PPPs) have been affected by the global financial crisis (GFC). After briefly discussing PPPs, particularly with reference to risk, the chapter outlines their contribution to the development of worldwide public infrastructure and highlights some initiatives designed to assist projects following the withdrawal of credit. It then analyses the effect the GFC has had on United Kingdom (UK) PPPs by investigating approximately 630 projects to assess whether difficulties in obtaining finance brought about by the crisis has led to a delay in PPPs reaching financial close. The findings confirm that PPPs currently in procurement are finding it more difficult to achieve financial close than pre-GFC projects and that therefore there are fewer PPPs now underway. The chapter concludes by looking at the future of PPPs in the UK following the change of government in 2010.
INTRODUCTION
Her Majesty’s Treasury (HMT) (2000) claims that PPPs bring the public and private sectors together in long term partnership for mutual benefit, asserting that the PPP label covers a wide range of different types of partnership including (p. 8):
the introduction of private sector ownership into state-owned businesses, with sales of either a majority or a minority stake;
arrangements where the public sector contracts to purchase services on a long-term basis so as to take advantage of private sector management skills incentivised by having private finance at risk. This includes concessions and franchises, where a private sector partner takes on the responsibility for providing a public service, including maintaining, enhancing or constructing the necessary infrastructure; and
selling government services into wider markets and other partnership arrangements where private sector expertise and finance are used to exploit the commercial potential of government assets.
In many cases, PPPs use a private company to design, build, finance and operate (DBFO) a new development such as a hospital or school over a contract period of 20-30 years. Throughout this period, payments are recouped from the public sector, which is ultimately responsible for the delivery of these services. When first introduced in the UK by the Conservative Government in 1992, the PPP initiative was met with relative scepticism and, although Labour opposed the scheme in opposition, it embraced PPP when it came to power in 1997. Since then the volume of PPPs, not only in the UK but throughout the world (Grimsey and Lewis, 2004), has increased significantly. Despite widespread criticism, the use of PPPs showed no sign of abating. However, the GFC that began in 2007 has significantly reduced the availability of private finance, and has therefore had a detrimental impact upon PPPs.
There are a number of well-documented reasons behind the GFC. One of the principal factors was the sharp rise in the number of subprime mortgages sold in the United States of America (USA) (Krinsman, 2007; Brunnermeier, 2009) and the subsequent ramifications in other countries, including the UK (Hall, 2008a). Uncertainty over the magnitude of the crisis has meant that banks have been reluctant to lend, with even the largest companies finding it difficult to obtain finance due to default and insolvency fears (Hall, 2008a). In the UK, the desire to safeguard existing PPPs and future infrastructure projects led at the time to government initiatives and increased borrowing from the European Investment Bank (EIB). Since, then, concerns over sovereign debt levels have also emerged in many countries within the European Union (EU) (for example Greece, Italy, Spain and Portugal) as well as elsewhere (for example, Japan and the USA). These concerns have exacerbated financial nervousness. Whilst Burger et al. (2009) provide some general statistics for how the GFC has affected PPPs in a number of countries; this chapter examines approximately 630 PPPs in the UK, a key user of this procurement method, to assess the impact of the GFC on the programme.
In terms of the format of the chapter, the next section reviews the literature associated with PPPs and risk. The effect of the GFC on PPP funding is then considered, together with several of the initiatives designed to safeguard existing and future projects. The methodology is then outlined before the research findings are presented and conclusions drawn.
public private partnerships
The introduction of PPPs was in response to concerns about the need to provide public infrastructure despite the high level of public debt, which grew rapidly during the macroeconomic dislocation of the 1970s and 1980s. As a consequence, pressure mounted to change the standard model of public procurement. In essence, a PPP is a contract between government and a consortium of private companies (referred to as a Special Purpose Vehicle (SPV)), under which the latter is required to DBFO an asset in return for payment over a number of years for both the cost of construction and the operation of the related service. Such payments may be based on either direct user charges (for example, toll roads), or a unitary payment from a public authority or a combination of both (Grimsey and Lewis, 2004).
The original objective of PPPs in the UK was to enable new infrastructure to be provided outside of the public sector borrowing requirement; however when Labour came into power in 1997 the emphasis shifted towards the achievement of value for money (VFM). This change in focus was in response to an amendment of Financial Reporting Standard (FRS) 5 which stated that the purchaser (the public sector) was required to demonstrate that the involvement of the private sector offered VFM when compared with alternative ways of providing the services (Accounting Standards Board (ASB), 1998). This VFM calculus was achieved by valuing the transfer of risk from the public to the private sector, with VFM being assessed through a comparison of the Public Sector Comparator (PSC) – the hypothetical cost of undertaking a project under conventional procurement – with the cost of procuring via PPP. VFM was deemed to be achieved when the price established under the PSC exceeded the price offered by the most competitive private bidder.
Since its conception, PPP has been heavily criticised. Some critics argue that the transfer of risk to the private sector is inappropriate, overvalued or does not take place at all. This criticism is on the basis that PPPs are rarely terminated, often due to potentially high litigation and counter claims by contractors (Hencke, 2003). Furthermore, as essential public services must continue to be delivered even if the contractor fails, this risk cannot be transferred. Edwards et al. (2004), for example, raised concerns over the level of risks actually being transferred and questioned whether those who are not best able to manage the risks are bearing them nonetheless, with the public sector being left with risks which are not easily quantified. Broadbent et al. (2008) also found that with regard to 17 PPP health projects certain items could be made invisible, whilst others that were either deemed more significant or possibly easier to monitor, were given unprecedented attention. Moreover, Pollock and Price (2008, p. 176) suggested that ‘the government’s central justification for PPP in terms of risk transfer remains largely unevaluated’ due to a lack of oversight in this area by government and reported that out of 622 PPP contracts signed up to October 2007, ‘only 10 financial inquiries into central government operational PFIs had been undertaken by the NAO [National Audit Office] by 2006, and of these only three examined the relationship between risk transfer and risk premiums’ (p. 177).
The risks posed by the GFC and today’s continuing market turbulence on PPPs stem from the interaction of threats and vulnerabilities. Burger et al. (2009, p. 10) identify the follows threats (i.e. the likelihood of a negative event occurring in the future):
The risk of an increase in interest rates leading to increasing costs, liquidity problems and project feasibility considerations for private partners and the possible postponement of projects by the government or it having to inject cash to support the SPV;
The risk of credit being unavailable leading to the termination of existing projects, existing projects failing to reach financial close and capital injections from government;
The risk of a decline in stock market prices leading to banks having reduced capital, which affects their ability to lend and causes reduced investment in new and existing PPPs;
The risk of exchange rate depreciation making new investments that rely on external borrowing less attractive, with private partners being tempted to export their services thus reducing the pool of domestic bidders; and
The risk that there is a slump in domestic demand leading to liquidity problems for private partners and lower domestic revenue for governments, meaning lower investment for new and existing PPPs.
PPP vulnerabilities (i.e. the preparedness of the partners to either prevent a threat or cope with its impact) to market turbulence can be project specific or extend more widely. The former includes overly optimistic revenue projections (for example, with respect to toll roads) while the latter may be related to the institutional framework. The institutional context is fundamental to managing PPPs to secure their benefits whilst containing the risks, which can be can be classified in a number of ways.
The ASB (1998) identified six main risks: demand; residual value; design; performance/availability; potential changes in relevant costs; and obsolescence. Moreover, a distinction may be made between commercial, macroeconomic and political risk. Macroeconomic risks entail aggregate demand risk, interest rate risk, liquidity risk and exchange rate risk. The materialisation of macroeconomic risk can, in turn, cause other risks. For instance, interest rate or demand risk can cause credit risk. Risk may also be categorised as exogenous and endogenous, with the latter being those risks that can be actively managed by changing behaviour. The risk management philosophy underpinning VFM has long asserted that risk should be allocated to the party best suited to carry, or manage, that risk. In principle, this should incentivise each party to act in a manner that manages the risk allocated to them and therefore improves the overall efficiency of the PPP.
To best allocate risk, two questions need to be answered (Organisation for Economic Co-operation and Development (OECD), 2008): first, which party is best able to prevent an adverse occurrence from occurring, and thereby ensure that the actual outcome conforms as closely as possible to the expected outcome; and second, in the case where no party can prevent an adverse occurrence (an exogenous risk), which party is best able to manage its outcome. Different parties carry different types and amounts of risk, and not all are affected in the same way. This may alter the attractiveness of PPPs for the parties most affected and reduce their interest in participating in PPPs unless they are compensated. As such they may not want to: enter into new PPPs; refinance debt in existing PPPs; or continue operating under an existing agreement. Risk can be managed in several ways including through (OECD, 2008): risk avoidance – the risky activity is not undertaken as, for example, when a public body forgoes an investment; risk prevention – action is taken to reduce vulnerabilities, for example, when a PPP consortium borrows in domestic currency to avoid exchange rate risk; and risk transfer – risk is transferred to another party through a contractual arrangement, such as minimum traffic guarantees.
The notion of ‘risk transfer’ plays an important role in justifying PPPs. Firstly, it is a key element in Eurostat’s definition of whether the debt is treated as being on or off the government’s balance sheet. Secondly, it is used, especially in the UK, to justify the use of PPPs which do not demonstrate that they are better value than the public sector option. This occurs when the aforementioned PSC is compared with a PPP bid and the latter is made less expensive by factoring in risk. However, transferring risk is not free. While it is possible to create contracts that transfer the risk of construction delays to the contractor, such contracts cost about 25 per cent more than conventional contracts (Hall, 2008b). Nor is risk transfer necessarily the best policy option, and it needs to be subjected to a cost-benefit analysis. For example, a theoretical analysis of risks and PPPs concluded that it is most efficient for demand risk to remain with governments (Engel et al., 2011). The International Monetary Fund (2004, p. 14) warns that governments may ‘overprice risk and overcompensate the private sector for taking it on, which would raise the cost of PPPs relative to direct public investment’. It is argued that this may have occurred in the UK as no attempt appears to have been made to monitor if risk transfer happens in reality, or how much benefit it really brings (Pollock and Price, 2008).
THE IMPACT OF GFC ON PPPs
The value of PPPs in Europe (excluding the UK) rose sharply during the period 2004 to 2006 to approximately €18 billion per annum (European PPP Expertise Centre (EPEC), 2010). The total value of PPPs signed by the end of 2006 was €31.6 billion, of which €23.6 billion were signed between 2004 and 2006. Moreover, at the start of 2007, projects valued at €67.6 billion were in procurement (Hall, 2008a). In the UK, the annual PPP programme increased from nine projects valued at £667 million in 1995 to 65 projects valued at £7.6 billion in 2002 (HMT, 2003). In addition, it was estimated that a further 200 projects with a total value of £26 billion would be closed between 2005 and 2010 (HMT, 2006). However, the value of PPP transactions reaching financial close fell sharply across Europe in 2008 and 2009 (from a high of approximately €30 billion in 2007) and, whilst returning to the 2004-6 levels in 2010, it remains well below the record years of 2005-7 (EPEC, 2010). In terms of the number of transactions, the UK remains by far the most active market across the European Union, with 44 PPP deals reaching financial close in 2010 (EPEC, 2010) and 20 during the first half of 2011 (with a total value of approximately €1.8 billion) (EPEC, 2011a). In value terms, Spain was the largest PPP country in 2010 (with 13 deals totalling approximately €4.4 billion) (EPEC, 2010), with France being the largest during the first half of 2011 (with eight deals totalling approximately €8 billion). Regardless of the improvement in 2010 and the first half of 2011 reflected in the figures above, the numbers and value of PPP transactions remain considerably less than those observed prior to 2008. To put the extent of this reduction in context, while the value of all European PPP deals for 2007 was approximately €30 billion, this figure had fallen to €18.3 billion in 2010 (recovering from just over €15 billion in 2009); this represents a decline of 39 per cent. (See Figure 1 for an illustration of the financial details of European PPPs between 2003 and 2011a.) Moreover, few large deals closed in the UK in 2010 and the first half of 2011 (EPEC, 2010 and 2011a).
FIGURE 1 European PPP Market 2003-2011 by € Billion
Source: EPEC, 2011b
PPPs are normally funded by 90 per cent debt finance and 10 per cent equity finance. Equity is higher risk as it will be lost first if the project company fails. Therefore, such shareholder loans are seen as junior to the external debt, known as senior debt, which is repaid first (NAO, 2010). Between 1995 and 2002 the use of both indexed linked and wrapped bonds in the financing of PPPs grew (see Kirk and Wall (2002) for a fuller explanation). However, following the 2007 housing market decline, the monoline industry, which guaranteed bond repayment if an issuer defaulted, collapsed resulting in the closure of the wrapped bond market (BBC, 2009). Consequently, the only viable source of finance for infrastructure projects was banks; however, the demise of Lehman Brothers in September 2008, widely accepted as the tipping point for the GFC, meant that the global interbank lending market dried up as banks stopped trusting each other. At the height of the crisis, banks were unable to fund themselves at the wholesale money market reference rates and there were suggestions that those rates had become unrepresentative. This constraint on liquidity meant:
less debt available for any given project and the need for a consortium of banks for all but the smallest of projects;
a higher price of debt, making it harder for privately financed deals to beat the PSC;
a shorter term for debt leading to refinancing risk and hedging issues; and
greater conditionality relating to the debt during the procurement phase.
A global review by PricewaterhouseCoopers (2008) reported that interest rates for lending to infrastructure projects had risen between 1.5 and 2 per cent above the lowest rates obtainable by governments, causing difficulties for both existing and future PPPs. Indeed, the NAO (2010) found that loan margins (i.e. above the interbank rate) for UK PPP projects had increased to around 2.5 per cent on average, with some complex projects facing margins of 3 per cent. EPEC (2010) reports similar commercial debt pricing. With respect to existing PPPs, loan repayments become more difficult, refinancing problematical due to the reluctance of banks to provide funding and, for concession-type PPPs such as toll roads, forecasted earnings are unlikely to be achieved due to a slump in domestic demand (Hall, 2009). Consequently, as reported above, the flow of new PPPs has slowed down.
Standard and Poor’s (2008) warned that some Spanish public authorities may have their credit rating revised downwards unless expenditure on employment and services is reduced because of the inflexible burden of PPP debt coupled with declining tax revenues due to the GFC. Ironically, this will increase the cost of debt and further reduce uncommitted income. In other countries there is also evidence that PPPs are being cancelled because of the GFC. For example, in Ireland, six social housing PPPs have been cancelled, a planned prison PPP was postponed indefinitely (Hall, 2009) and a metro PPP has been deferred (Department of Transport, Tourism and Sport, 2011). In Australia, despite its extensive use of the PPP model, there has also been a renewed questioning of overinflated traffic forecasts (Ferguson, 2009; KPMG, 2009a) following the failure of a number of projects.
The GFC has also had a significant impact on PPPs in the UK with only 34 deals being signed in 2008 (Hall, 2009), which was approximately half that of 2007 and the lowest level of activity for over a decade (Kapoor, 2008). Although, as noted above, EPEC reports that this has recovered slightly to 44 deals in 2010 (albeit the value of such deals is much lower). The NAO (2010, p. 9) highlighted that as well as charging higher margins, banks are adopting a more cautious approach to lending following the credit crisis and are as a result: lowering the proportion of debt in projects; requiring the private sector to inject equity earlier; and placing more onerous conditions on when the private investors can withdraw cash from the project. Hence, UK public service programmes that currently rely on PPPs may suffer. Moreover, the GFC led to the collapse of land and property values which contributed to the failure of businesses, declines in consumer wealth, substantial financial commitments incurred by governments and a significant decline in economic activity. This impacted upon PPP deals which involved disposal of land as part of the financing, and may have contributed to the termination of the Defence Training Review programme (see Case Studies) (Defence Policy and Business, 2010). Further issues include increased government guarantees, greater state involvement in some UK banks and direct HMT lending (see below) which makes the achievement of VFM more difficult. This increased public sector risk coincides with substantial strain being put on public finances, arguably due to the measures taken by the previous Labour Government to protect the fragile economic recovery, support growth and job creation and provide reassurance to capital markets (Ostry et al., 2010). Consequently, the Coalition Government has reversed a number of policies implemented by the previous Labour Government, including the Building Schools for the Future (BSF) programme (see Case Studies) (Richardson, 2010). Given these difficulties, the UK and other countries have sought to introduce measures to assist PPPs struggling to reach financial close. Some of these are now outlined.
UK Initiatives
Four main approaches being were trialled in the absence of traditional financial approaches in the UK; they were mini perm structures, HMT lending, the non-profit distributing model and the prudential borrowing framework. Each of these is now briefly explained.
Mini-perm structures
Broadly speaking, a mini-perm is a short-term financing tool, usually payable in three to five years and typically used to pay off income-producing construction or commercial properties. The term ‘perm’ is short for ‘permanent’, alluding to permanent financing, albeit for a short period of time as indicated by the word ‘mini’. Mini-perm financing might be used by a developer until a project has been completed and can therefore start producing income and establish an operating history. In other words, this type of financing is used prior to being able to access long-term financing or permanent financing solutions. The interest payable on a mini-perm will usually be higher that longer-term financing options, often with a balloon payment at the end of the term in anticipation that the loan can then be easily refinanced due to the fact that the asset now has an operating history on which to successfully obtain less-expensive permanent financing. They can be split into two distinct types – hard and soft (KPMG, 2009b). The former has a relatively short maturity, typically five to seven years, at which point the bulk of the loan remains outstanding. Arguments for hard mini-perms are that they force refinancing, which would be at prevailing market prices, and they allow the lenders to price on a short term basis. In contrast, soft mini-perms have a longer maturity, for example 26 years of a 28 year contract. Nevertheless, two features encourage early refinancing. Firstly, incremental step ups of 25-50 basis points at certain dates result in the cost of borrowing being more expensive if the loan is not refinanced. Secondly, a cash sweep at a certain date is used to repay the outstanding debt rather than distribute rewards to shareholders. In 2009, it was reported that two large PPP projects and one small one had been financed using a mini-perm; however, it was felt that these projects had caused affordability issues for the public sector and increased the private sector’s risk exposure (KMPG, 2009b). Therefore it is doubtful that such structures will prove to be much of a solution, particularly with their emphasis on refinancing which is less likely due to the GFC. An example of a large project using a soft mini-perm structure would be London’s Riverside waste-to-energy £570 million PPP, which reached financial close in July 2008.
HMT lending
In 2009, the Labour Government announced its intention to lend to PPPs which were unable to raise sufficient finance (HMT, 2009). The aim was not to replace banks or capital markets, but to provide additional funding, with the private sector and EIB continuing to supply the majority of finance. To qualify for HMT lending, there must have been a failure to secure finance following a competitive process and any funding offered must have been unrepresentative of market terms. These loans would still bear interest and be repaid over the life of the project. However, HMT hoped that if favourable market conditions returned the loans could be sold prior to maturity at a profit. These loans were to be issued by HMT’s own finance unit known as The Infrastructure Finance Unit (TIFU), which was established with the aim of supporting PPP schemes in procurement, thereby safeguarding £13 billion of public investment (HMT, 2009). These included the Greater Manchester waste project (see Case Studies), the M25 widening, Merseyside Waste, Building Schools for the Future (see Case Studies) and a number of hospital projects. However, TIFU only provided lending for the Greater Manchester waste project (see Table 5) and was subsequently placed under the umbrella of Infrastructure UK, which has a much wider role in reducing the cost of infrastructure projects (NAO, 2010). As with mini-perms TIFU loans were seen as a short-term solution until the project could obtain more conventional finance.
Non-Profit Distributing model (NPD)
Although similar to PPP, the main difference is that the NPD provides economic or social infrastructure financed 100 per cent by debt (90 per cent senior and 10 per cent junior). This differs from PPP deals, which normally consist of 90 per cent debt and 10 per cent equity. Under an NPD, SPV shareholders receive a capped return on their capital, with any surpluses remaining at the end of the contract being passed to a designated charity as opposed to being paid out as dividends. Subsequently the dividend opportunity is removed, which is considered to flatten out overall risks when compared to equity-based PPPs or public procurement. NPDs are therefore still attractive to banks, but not as popular with investors or bidders as they do not obtain the same returns (Hellowell and Pollock, 2009). This model was piloted in Scotland in the Argyll and Bute Council’s schools’ project, which reached financial close in September 2005. Since then, two more schools’ projects (in Falkirk and Aberdeen) reached financial close in May and December 2007 respectively. Moreover, the National Health Service in Tayside has used this model for a PPP and the Borders rail link project, which was announced in 2008, will also use NPD.
Prudential Borrowing Framework (PBF)
Although it could be argued that the PBF was not initiated in response to banks’ reluctance to lend, it does provide an alternative to PPP. Indeed, as it excludes the private sector from all aspects of the project apart from construction it is closer to traditional procurement. Under the Local Government Act 2003, local authorities were given greater freedom over their capital expenditures; therefore whilst most of their revenue still comes from central government, it now has less say over how this money is spent. Therefore local authorities are no longer forced down the PPP route by central government. However, whilst Hood et al. (2007) believe that the PBF has benefits, they feel it is not as robust as PPP regarding the treatment and allocation of risk. One UK local authority spent £11 million on a programme of highways structural work via the PBF, which they calculated would not only deliver a better long-term solution but would deliver savings of £1 million from the highways maintenance budget and reduce future liabilities by a further £1.9m. Moreover the resurfacing work would increase the operational life of the road and reduce the number of insurance claims and litigation from potholes created by adverse weather conditions. Had they gone down the PPP route they would have had to deal with the substantial running costs of such projects, complex contractual arrangements and extended contractual periods, all of which contribute to a heightened risk profile (Hood et al., 2007).
Other initiatives
PPPs in France have never been equivalent to PPPs elsewhere from a legal perspective, but recent financial turmoil has prompted financial reforms, there too. In order to alleviate the problems with the financial markets, several measures have been introduced (Hall, 2009). These include: a government guarantee for all PPP bank loans; tax allowances; allowing the government to advance to a bank the majority of the loan required by the private partner (thus enabling the bank to pass on lower interest rates obtainable by government); and allowing PPPs to be signed on the basis of ‘adjustable financing’ without finalising a deal with banks so that it can proceed on the basis of government advances while waiting for improved conditions in the financial market. PPPs have been widely promoted in developing countries for many years by the World Bank and other donors and development banks. However, the International Finance Corporation (IFC), the World Bank’s private sector arm, believes that the GFC will make it even harder to finance PPPs. It estimates that projects totalling $110 billion may be delayed or cancelled, and that $70 billion of existing PPPs are at risk because of increased financing costs (IFC, 2008). Therefore, the IFC has created a global equity fund and a loan financing trust to support PPPs.
It can be seen therefore that the financial threats highlighted by Burger et al. (2009) have impacted on the PPP initiative. Interest rates have increased, credit has become less available, there has been a slump in domestic demand and increased borrowing from the EIB could expose the UK public sector to exchange rate risk.
METHODOLOGY
The data used for this research was obtained from two primary sources: HMT (2011) statistics for both signed projects (698) and those still in procurement (61) at 16 March 2011 and Partnerships UK’s (PUK) (2011) project database. The HMT signed projects’ list is revised on a six-monthly basis to reflect the updates HMT receives from departments at Budget and Pre-budget Review. The list of projects in procurement is also updated regularly. The PUK database holds details of 920 projects that have all achieved financial close. In compiling this database, PUK liaise with HMT, government departments, the Welsh Assembly and the Northern Ireland (NI) and Scottish Executives. The main reason for the difference between the HMT signed list of 698 projects and the PUK database of 920 is that the latter also contains non-PPP projects. Furthermore, with respect to the HMT list of both signed projects and those in procurement, the same amount of information is not provided for each project. Accordingly, for the purposes of this research, 570 (82 per cent) signed projects and 57 (93 per cent) of those in procurement were deemed useable.
In order to ascertain whether the GFC has led to delays in projects being closed, the length of time between the appearance of the project in the Official Journal of the EU (OJEU) and financial close was measured. The EU Public Procurement Directives require all public sector bodies to publish details of tenders and contract opportunities in the OJEU and financial close is deemed to be when both the bidder and the purchaser have reached agreement on: all the contractual documents; all relevant technical issues; and all matters affecting the unitary charge. The only remaining issue is for the bidder to fix the interest rate on the debt taken out to finance the project. Comparisons were then made between the length of time from the OJEU notice to financial close for projects signed since the beginning of the PPP scheme in 1992 and those in procurement at 16 March 2011. It was expected that these latter projects would have been delayed due to the reluctance of banks to lend.
Three case studies are also presented to demons
