This paper reviews economic developments in the United Kingdom during 1991–96. The paper examines two sets of possible reasons why the United Kingdom's savings and investment rates are lower than elsewhere. The first consists of factors leading to lower optimal rates of savings and investment: these include demographics, economic structure and technology, a liberalized financial environment, and so on. The second consists of distortions leading to lower-than-optimal savings and investment rates. The paper also presents new estimates for the potential growth rate, the output gap, and the natural rate of unemployment.


This paper reviews economic developments in the United Kingdom during 1991–96. The paper examines two sets of possible reasons why the United Kingdom's savings and investment rates are lower than elsewhere. The first consists of factors leading to lower optimal rates of savings and investment: these include demographics, economic structure and technology, a liberalized financial environment, and so on. The second consists of distortions leading to lower-than-optimal savings and investment rates. The paper also presents new estimates for the potential growth rate, the output gap, and the natural rate of unemployment.


The starting point is a clear presumption that the PFI approach will generally be better than a traditional procurement; the better management inherent in a PFI project will give better value for money. As a matter of Government policy, public bodies should always consider private finance options unless it is clear that transfer of project control and risk to the private sector is not feasible, intimately, the vast majority of projects across the public sector will be suitable for the PFI. In the short term the PFI should not be allowed to hinder investment activity.45

There are substantial dangers that recourse to private finance will be used as a means of undertaking hidden public borrowing and expenditure. Without a significant transfer of risk to the private sector, schemes for private finance look like an attempt to circumvent budgetary controls on public expenditure, whether by creative accounting around definitions or by retiming the scoring of expenditure.46

A. Introduction

118. The private finance initiative (PFI) was launched in November 1992 with the view to mobilizing private sector finance and expertise for services traditionally provided by the state. Under the initiative, the government sheds its role as a direct provider of “public” services, concentrating on authorizing and regulating their provision by private firms. Rather than procuring assets to operate, the state contracts for services. The PFI is one element of policies—ranging from privatization to contracting out—designed to increase the involvement of the private sector in the delivery of services traditionally provided by the state.

119. The desired benefits of private sector involvement are budgetary savings through efficiency gains derived mainly from the appropriate sharing of risk between the private and public sectors. In a PFI project, private firms are delegated the design, financing and operation of a project, rather than just its construction. The private partner is thus responsible for, and financially exposed to, the project’s operation over its economic life. This exposure—a key element of the risk transfer emblematic of PFI projects—is intended to focus attention on the full-life costs of the services to be provided, including both construction and maintenance costs. The wide latitude—relative to orthodox procurement procedures—afforded PFI suppliers on how best to design, construct and operate an asset to provide a desired quality and level of service is expected to spur innovative cost reduction.

120. Not surprisingly, this novel approach to public procurement, with potentially far-reaching implications for both the public and private sectors, has generated considerable controversy. Skeptics47 have suggested that the PFI could be used as an accounting sleight of hand designed to push investment spending off the government books and artificially reduce the PSBR. These critics have cautioned that such a stratagem merely creates deferred liabilities that, if left unmonitored, could spiral out of control. Other commentators48 have been concerned to determine whether PFI expenditure was additional to, or a substitute for, traditionally procured and financed public investment, a concern that seems misplaced in a country with open and well-functioning financial markets.49 Another point of controversy has revolved around the setting of investment priorities: in a time of fiscal stringency, it is argued, projects might be selected and implemented on the basis of private sector willingness to participate rather than a disinterested assessment of public need.50 The ability of PFI projects to offer value for money has also been a bone of contention, with some commentators51 suggesting that the measurement of quantifiable PFI efficiency gains was likely to be elusive, while the higher cost of capital52 faced by the private sector was rather more tangible and a positive function of risk. Other commentators53 have wondered why the techniques used to foster efficiency in PFI projects could not be transferred to traditionally procured and financed investments, thereby improving efficiency without incurring higher financing costs. The issue of risk transfer from the public to the private sector—a key feature of PFI projects—has also raised concerns, with some54 arguing that such transfer is illusory as the state will ultimately be exposed to project failure. Others55 more favorably predisposed to the idea of the PFI, have expressed concern about the slow pace of its implementation, highlighting cumbersome tendering procedures and contract negotiations. These otherwise supportive commentators worry that the authorities’ ambitions for the PFI might be disappointed as a result of administrative bottlenecks and inexperience with a novel technique.

121. The object of this paper is to describe the main features of the PFI and consider its merits. The next section places the PFI in the context of previous initiatives in the United Kingdom to increase the role of the private sector in the delivery of services to the public and highlights the similarities of the PFI with project finance techniques adopted elsewhere. Section C puts past and near-term prospective PFI spending in perspective. Section D outlines the structure and characteristics of PFI projects. The central issue of risk transfer and key objective of achieving value for money in PFI projects are considered in Sections E and F, respectively. Section G reviews the changes in bidding and procurement practices that the PFI entails. Accounting issues associated with the introduction of the PFI, including the proposal by the United Kingdom authorities to adopt an accruals-based system of resource accounting, are considered in Section H. The possible impact on the private sector of growing recourse to the PFI is outlined in Section I. The paper concludes, in Section J, with a positive assessment of the PFI that nevertheless highlights the need to monitor closely forward PFI commitments and overcome the administrative bottlenecks that have so far slowed its implementation.

B. Precursors and Analogues

122. The PFI is a continuation of attempts to involve private finance in public investment that were initiated in 1981. At that time, a committee set up by the Treasury under the chairmanship of Sir William Ryrie issued a report that recommended guidelines (the “Ryrie rules”) for such private financing. The Ryrie rules established two broad principles for the private financing of public investment. First, privately funded projects had to be cost effective under conditions of fair competition between the public and private sectors. In particular, the benefits of such projects needed to be sufficiently high, compared with a purely public alternative, to offset the higher cost of capital faced by the private sector. Second, privately funded projects were generally to be used to replace rather than supplement traditionally financed public investment. In 1989, the Ryrie rules were relaxed to permit privately funded projects to proceed without offsetting reductions in public investment.

123. The PFI further relaxed the Ryrie rules, mainly by shifting the presumption of efficiency to the private sector, notwithstanding that sector’s higher cost of capital. At the same time, initial PFI proposals56 suggested that one of its contributions would be the mobilization of additional resources for public investment, an objective that has subsequently been reinterpreted.

124. The PFI is somewhat analogous to project financing techniques used elsewhere. There are many examples—in Australia, New Zealand and the United States, among other developed countries—of infrastructure projects designed, financed, built, and operated by private sector developers. Private financing has also recently begun to play an increasingly prominent role in infrastructure projects in developing countries, especially in the energy, telecommunications, transport, water, and sanitation sectors. Although the difficulty of mobilizing sufficient domestic resources for large infrastructural needs is typically the main impetus for recourse to private financing for infrastructure projects in developing countries, such financing has also been promoted as a means of improving efficiency. Most such arrangements assign the cash flow and delegate the management of the project to the private (usually foreign) partner for a certain period, with ownership and control subsequently reverting to the government. In many developing countries, however, project risk is not limited to the uncertainty surrounding the ability of the project to generate sufficient income to compensate the investor. The regulatory framework—exchange controls, restrictions on profit repatriation, the possibility of nationalization, and changes in the rules governing the operation and pricing power of the project—is also a critical source of risk. As a result of the prominence of regulatory risk, the private financing of infrastructure in developing countries is often accompanied by government guarantees, a thorny issue involving subtle judgements about the suitability of indemnities and their pricing and monitoring that does not arise in the case of the PFI.

C. Past and Prospective PFI Spending

125. Eventually it is hoped that the scope of the PFI will expand to encompass virtually all public sector projects. However, the pace of approvals so far has been slow, owing to administrative problems (with project specification and the bidding process) arising from the novelty of the initiative and the complexity of structuring the desired transfer of risk. As a consequence, only £5 billion in projects have been agreed, and over half of this total is accounted for by one project, the fast rail link to the Channel tunnel. PFI approvals are projected to increase: departments expect to have concluded PFI contracts involving capital spending of £14 billion by 1998/99. This figure could be surpassed as the private finance panel—a body established in 1993 to promote the PFI and serve as a repository of best PFI practice—has identified over 1,000 projects worth £25 billion that could eventually come to fruition.

126. Actual PFI capital spending in the three years through 1995/96 is estimated to have reached only £1.2 billion, however. Looking forward over the three-year budget planning horizon, PFI annual capital spending is targeted to increase steadily from £1.9 billion in 1996/97 (compared with total investment of £22.4 billion) to £2.8 billion in 1998/99 (versus £22.0 billion). About half of the £7.3 billion in PFI capital spending expected during 1996/97–1998/99 is accounted for by the transport sector. Even taking into account anticipated PFI expenditure, however, total publicly sponsored investment is set to decline as a share of GDP during this period (Tabulation below).

Public Sector Capital Expenditure

(In percent of GDP)

article image
Source: H.M. Treasury.

D. Structure of PFI Projects

127. PFI schemes can in principle run the gamut of public procurement, from investment in “bricks and mortar”, rolling stock and other tangible assets to the provision of information management services. Thus far, however, most PFI projects have tended to involve lumpy, immovable investments with high up-front costs and planned initial overcapacity. Their benefits typically have had a large public good component, and their exposure to changes in regulation have often been higher than most purely commercial ventures. Although an ideal PFI project would target multiple uses, in some cases the assets so created have had only a sole use, heightening the risk of shortfalls in demand.

128. To accommodate these characteristics, the PFI envisages three types of project structure. First, financially free-standing projects—such as toll roads or bridges—are undertaken on the expectation that the income stream generated by the project itself will be sufficiently remunerative to induce private investors to bear the project’s full costs and risks. In such cases, the services provided are typically sold directly to private users, and the role of the state is limited to authorizing the project and supervising its operation. In particular, financially free-standing projects in a monopoly position are regulated to ensure that that position is not abused. The departmental sponsor of a such a PEL project seeks to ensure that a genuine transfer of risk is achieved with no explicit or implicit guarantees provided by the public sector. Since financially free-standing projects do not have recourse to public funding or guarantees, an appraisal by the sponsoring department of whether the scheme provides value for money is not needed. Rather, the cost effectiveness of the project is assured through competitive tender.

129. Second, projects may be financed through payments by the state for services sold directly to it. This structure is appropriate when the correct balance of risk sharing is muddied by the ability of the public sector to control (or greatly influence) the demand for the service in question. Examples include privately financed and operated prisons. Under PF1 guidelines, projects involving services sold directly to the public sector must be shown to provide value for money and in some cases be compared against a hypothetical public sector comparator. However, such a comparison is necessary only if there is a realistic expectation that a public sector alternative would be available in a similar time frame.57 Unlike a financially freestanding PFI project, all risk need not be transferred to the private supplier. Rather, the design of PFI projects that involve the sale of services directly to the state seeks to strike an optimal balance between shifting risk to the private sector and higher cost.

130. Finally, the PFI encompasses joint ventures under overall control of the private partner that involve private and public capital and social benefits not easily captured in financial returns. To qualify as a PFI project, effective control of the joint venture must lie with the private partner. The public sector contribution to such joint ventures is typically a function of the social benefits not captured in the project’s financial returns and the level of public funding needed to ensure that the project is sufficiently remunerative to the private partner. Examples of joint ventures include the Channel tunnel rail link, and urban regeneration schemes. In the case of joint ventures, only the public sector financial contribution need demonstrate value for money.

E. Allocation of Risk

131. Risk transfer is the essence of the PFI. Without an appropriate apportionment of risk, efficiency gains are unlikely to be realized. The underlying principles governing risk transfer in PFI projects are that such transfer should create appropriate incentives and be optimal rather than for its own sake, the latter necessitating that risk be borne by the party best able to manage it. Gauging what constitutes optimal risk transfer is difficult, however. While private firms are likely to be better able to bear certain risks than the public sector, the transfer of unfamiliar risk to the private sector is unlikely to result in value for money.

132. PFI projects face a number of risks, including: the risk of cost or schedule overruns in design and construction; the risk of failing to meet standards for operating levels, quality, or efficiency; the risk of a shortfall in demand from anticipated levels; the risk that the residual value of the asset will fall short of expectations; the risk of obsolescence; the risk of changes in regulation; and the risk of failing to raise affordable financing.

133. In principle, the private sector PFI partner is expected to assume essentially all risk associated with meeting the cost and timing targets for the design, construction, and operation of a project. It is by unambiguously placing the responsibility for performance in these areas with the private sector, that PFI projects are expected to generate efficiency gains. Tying payments under PFI projects to the delivery of a clearly specified level of service ensures that the impact of construction delays and cost overruns are borne by the private sector provider, not the public sector as is often the case in the traditional approach to public procurement.

134. The risk of a shortfall in demand from projected levels is apportioned differently depending on the nature of the PFI project. In the case of a financially free-standing project, the private sector supplier bears all risk of a shortfall in demand, although in the case of a monopoly supplier, the level of demand risk will depend on the terms struck for pricing and the duration of the concession. In the case of services sold directly to the state, the scope for shifting demand risk to the private sector supplier is reduced, and typically becomes a matter of negotiation that attempts to strike a balance between risk transfer and cost. In some cases, the government has attempted to identify this balance by seeking bids both with and without minimum usage guarantees.

135. PFI guidelines specify that regulatory risk, except certain changes in regulation that are specific to the service, should be borne by the private sector. Regulatory changes—such as prison regulations—that have a specific and significant impact on a PFI project would be typically be borne by the public sector, while changes to more widely applicable tax or environmental regulations would not.

F. Value for Money

136. PFI projects are expected to be more efficient than their orthodox procurement counterparts for a number of reasons. First, as discussed in the previous section, the PFI seeks to improve the analysis and allocation of the risk of creating and managing fixed assets. In PFI projects, risks are to be allocated to those best able to bear them as a means of increasing efficiency and improving the incentives to perform Second, PFI guidelines require contracts to specify detailed standards for services and tie payments to observing those standards, thus explicitly exposing the supplier to the risk of not delivering the desired level of service efficiently. Third, PFI projects are oriented to ensuring efficiency over the entire life of the project, reversing the incentive under the traditional approach to procurement to focus on limiting the initial capital outlay with little regard to the higher recurrent costs for maintenance and operation that might result. Fourth, PFI projects afford greater latitude to bidders for PFI contracts to determine how best to design, construct and operate the fixed assets concerned, increasing the scope for innovation and the prospect of minimizing full-life costs. Fifth, PFI suppliers are encouraged to design assets that are amenable to providing services to third parties and likely to retain significant residual value at the end of the PFI contract. Under the PFI, for example, an asset’s excess capacity could be sold to a third party. Finally, by facilitating the imposition of user charges, PFI projects should contribute to efficient asset use by better matching supply and demand.

137. There are a number of features—most of which revolve around the degree of flexibility available in the design and operation of the asset—thought to make a project particularly amenable to the PFI. For example, a project that serves a number of potential end-users, is output driven, offers scope for innovative design, spans a long horizon, involves a large element of operation and maintenance, and entails mainly commercial risks is likely to be well-suited to the PFI. In contrast, projects less amenable to the PFI are those that are designed for a sole use, are asset driven, impose inflexible design specifications, involve only limited operating and maintenance costs, and are politically sensitive or entail risks unfamiliar to the private sector. Similarly, projects whose need for government support is limited to the provision of licensing and operating concessions are more likely to be suitable for the PFI than those requiring government guarantees.

138. The cost effectiveness of PFI projects is to be assured through competitive bidding, and in some limited circumstances by comparison with a public sector alternative. Public sector comparisons are required only if: (i) public money is involved; (ii) the public sector is the main source of the project’s revenue; and (iii) the project might have proceeded as a public sector endeavor in a similar time frame. Moreover, since public sector comparators do not have to face the discipline of a market test and are consequently likely to be optimistic, PFI guidelines urge that their use not be allowed to delay the growth of PFI projects. Responsibility for assessing the merits of individual PFI projects falls to individual spending departments, subject to normal expenditure controls and ex post audits.

139. The Treasury (1996) has tentatively estimated the cost saving from the average PFI project to be 21 percent, representing the weighted average of a 20 percent reduction in operational costs (accounting for 70 percent of total) and a 24 percent reduction in capital costs. The operational cost saving estimate reflects the experience with the government’s program of contracting out, another program of private sector involvement in public procurement. The capital cost saving estimate reflects the average cost overrun experienced in past publicly funded construction projects, a cost that, under the PFI, would be borne by the private sector.58 In the Treasury’s estimation, even if the average PFI project faced a much higher cost of capital (10 percent, say) than the public sector alternative, the resultant increase in total project cost (3 percent) would be more than offset by the PFI’s efficiency gains.

G. Bidding and Procurement

140. Competitive bidding is at once a key element of ensuring the PFI’s ability to provide value for money and a main stumbling block to its implementation. High bidding costs, the inexperience of both the public and private sectors with a novel approach to bidding, and the inherent complications of assessing bids for PFI projects relative to traditionally procured and financed projects have contributed to delays in the implementation of the PFI. There is also a tradeoff between the desire that private finance options be explored for all projects, and the need to be selective to ensure that only those projects most likely to be implemented successfully in a reasonable time frame are put out to tender.

141. An appropriately structured PFI invitation for bids focuses on the desired outputs, in the expectation that leaving design and implementation to the bidders would result in efficiency gains. However, a call for tenders that is light on operational details can complicate the evaluation of bids, arguably increasing the subjectivity of such evaluations, especially when assessing competing designs offering varying tradeoffs between cost and risk transfer. Orthodox public procurement, in contrast, which provides detailed specifications on how to provide the service is more likely to result in similarly structured bids whose main distinguishing feature is price. In addition, PFI bids are likely to be more complicated than their orthodox counterparts as they extend over the entire life of the project rather than merely over the construction phase. These factors can create uncertainty among bidders and boost bidding costs, eroding the ability of PFI projects to provide value for money, and, in the worst case, reducing the willingness of firms to tender bids.

142. To address these issues, recent Treasury guidelines59 for PFI projects emphasize the need for better vetting of departmental PFI proposals to ensure that those offered for tender are likely to have the characteristics of a successful PFI project. The Treasury has also recently taken steps to reduce bidding costs by developing PFI expertise in departments, improving the vetting of projects before the bidding process, and standardizing PFI contract clauses.

H. Accounting Issues

143. The PFI raises two main accounting issues. First, PFI projects typically entail explicit long-term forward commitments, in contrast to the implicit commitments associated with the procurement of an asset through traditional means. In the case of the PFI, the stream of forward payments are contractual obligations tied to the provision of a desired service. The explicit forward commitments of PFI projects have highlighted the need for improved accounting and budgeting controls to avoid the accretion of unsustainable levels of obligations. However, largely because of the novelty of the initiative and the relatively low level of PFI spending commitments to date, a system for centrally monitoring the forward commitments arising from PFI projects is not yet in place, a shortcoming that the Treasury plans soon to correct.60

144. Second, compared with their traditionally procured and financed counterparts, PFI projects tend to substitute recurrent spending over a number of years for capital spending concentrated in relatively fewer accounting periods. This characteristic profile of PFI-related expenditure has led some to charge that the PFI is merely an accounting sleight of hand designed to ease pressure on the measured budgetary aggregates.

145. In order to blunt the incentive to use the PFI as a means to circumvent a cash-based system of expenditure control, the accounting for PFI projects follows principles established for financial leases, which pose a similar problem of how to time the recognition of spending commitments.61 In both cases, the accounting treatment attempts to reflect the economic substance of the transaction, rather than merely the cash flow generated. Thus, a PFI project (or lease) that is judged to be merely a financing mechanism with the private sector (or lessor) bearing little or none of the risk of the transaction would be fully capitalized and count against the control total of the year of acquisition.62 State support to PFI projects substantially financed by receipts from third parties would not normally be capitalized, however. In the case of services sold directly to the public sector by PFI projects, the payments for services would be capitalized only if a substantial element of risk resides with the public sector. In light of the emphasis on risk transfer in PFI projects, it is unlikely that any such project would be capitalized. Indeed, PFI guidelines specify that any PFI project likely to entail capitalization should be renegotiated to expose the private financier to greater risk.

146. Although the PFI admittedly complicates the fiscal accounts, it also has the potential to improve the budgeting process. The replacement of lumpy investment expenditure with a smooth stream of payments for services could improve investment planning, as the latter is less likely to be subject to extraneous considerations than the former. Moreover, under the PFI, forward spending commitments to sustain a desired level of services are made explicit over long time horizons.

147. The proposed adoption of an accruals-based system of resource accounting (see box) would tend to level the playing field between PFI projects and traditionally procured and financed assets and reduce the incentive to pursue a PFI project as a means of spreading expenditures over a number of accounting periods, an improvement over the imperfectly monitored implicit commitments of assets procured and accounted for in the traditional way.

I. Implications for Private Sector Suppliers

148. Increased recourse to the PFI as a technique for providing services to the public will likely engender changes in the private sector. The bundling together of the design, construction, operation and financing components of a project is likely to encourage the increased use of consortia. If PFI projects reach a sufficiently large share of the market, construction companies will have a strong incentive to become owner-operators. The longer horizon of PFI projects and the expectation that the private supplier will amortize capital costs through recurrent project revenue rather than direct payments during the construction phase of a project suggest the need for a higher share of equity financing, rather than the current heavy reliance on relatively short-term debt in the case of traditional projects. PFI suppliers—construction companies and general contractors—may not have sufficiently deep balance sheets to take on the long-term financial commitment likely to be needed in PFI projects, suggesting that, to be successful, the PFI will need to spawn new project financing mechanisms. The securitization of cash flow streams from PFI projects could be one such mechanism that would facilitate both the mobilization of financing and the unbundling and dispersion of risk. Similarly, a bond market oriented toward refinancing bank debt could be developed to ease funding and spread risk. The availability of equity or venture capital financing for PFI projects could be enhanced through the development of specialist funds for qualified investors. Such vehicles have been launched to mobilize capital mainly from institutional investors for infrastructure development in (mainly Asian) developing countries.

Resource Accounting

The United Kingdom government has decided to supplement the current cash-based system of public sector accounts with a system of accrual accounting and budgeting. This system of budgeting would focus on resources consumed to achieve desired outputs rather than cash outlays. Under the proposed system, the current cash control total would be supplemented by one that recorded transactions on an accruals basis. Resource accounts are to be introduced in most departments by April 1997 and in all departments by April 1998; such accounts will first be published and presented to Parliament in 1999–2000.

Resource accounts would thus explicitly recognize that fixed assets have a useful life beyond the current accounting period, are consumed over time, and need eventually to be replaced. In addition to the depreciation charge, departments would face a cost of capital reflecting the opportunity cost of assets at their disposal. In this way, capital charges would reflect both the cost of holding an asset as well as the procurement of new assets. By facilitating a comparison of the capital charges of owned assets with the recurrent costs of a PFI project, the adoption of resource accounting would establish a level playing field for these two approaches to the provision of public services. Current liabilities would be accounted for as incurred, rather than only when payments are actually made. As part of the greater emphasis on matching resources used with outputs to permit an assessment of the efficiency with which services are provided, departments would also produce information on their outputs.

Although not issues in the case of the United Kingdom, the application of accrual accounting for current transactions facilitates the tracking of arrears, provides an early indicator of impending insolvency, and prevents transactions whose sole motivation is the smoothing of the cash accounts.

J. Conclusion

149. The PFI is a welcome innovation that has the potential to increase the efficiency with which services are provided to the public. If some of the claims for the PFI made by its proponents appear overstated, its detractors seem at times to contrast the PFI’s possible pitfalls with an idealized vision of traditionally financed and procured public investment—a system that has shown itself to be far from ideal in practice. The PFI never had the prospect of creating additional investment, except in the sense of achieving more with a given level of spending. Neither can it be dismissed as merely a technique of creative accounting, although that possibility remains in a cash-based system, necessitating careful monitoring of forward PFI commitments. The PFI does raise several interesting issues about how best to manage the increased involvement of the private sector in activities that have been in the past the domain of the public.

150. One such issue is the concern that the PFI might distort public investment decisions. In this view, the combination of tight limits on expenditure and the availability of “off balance sheet” PFI financing might unduly influence the shape of publicly sponsored capital spending. At present, however, the PFI does not appear to be influencing investment decisions; on the contrary, the pace of PFI spending is generally seen to be slow. Even when the PFI has come to full fruition, it seems likely that it will remain merely a mechanism to provide services deemed appropriate by purchasing departments, not a determinant of which services are provided.

151. Assessments of the PFI based on its ability to mobilize additional funding suffer two main flaws. First, in the absence of a counterfactual—how much would “public” investment spending be in the absence of the PFI?—attempts to measure additionality are meaningless. Second, and perhaps more telling, the discussion of additionality hinges on an artificial—from the perspective of national savings and investment balances—distinction between private and public finance for publicly sponsored capital spending. Whether an appropriate level of public investment is financed by increasing public borrowing or encouraging the private sector to provide financing in another form should not affect real interest rates, even if the measured PSBR is lower. As suggested earlier, the only possible element of additionality in the PFI is increased efficiency that allows more and higher quality services to be provided for a given level of investment.

152. However, because of measurement problems, the greater efficiency of PFI projects over publicly financed alternatives must be taken largely on faith. PFI guidelines seem in effect to instruct departments to have faith and presume that such gains will be realized. In contrast, the private sector’s higher cost of capital is inconveniently tangible, easily measured, and a positive function of risk. Nevertheless, it does not require a tremendous leap of faith (or even strict adherence to Treasury’s estimates of efficiency gains outlined in Section F) to conclude that the PFI’s more explicit attention to and better allocation of risk, increased scope for innovation, and focus on minimizing full-life project costs should result in real cost savings versus the traditionally procured and financed alternative. Assessments of completed PFI projects should aim at measuring, to the extent possible, the efficiency gains arising from private sector involvement. Suggestions by some PFI detractors that the lessons in cost effectiveness it provides could be assimilated by the public sector seem to ignore the role of incentives in spurring efficiency.

153. The increased efficiency of the PFI hinges critically on risk transfer that creates appropriate incentives. Without the risk of loss to concentrate the mind, the hoped-for efficiency gains of private sector involvement are less likely to be realized. PFI guidelines appropriately emphasize that to reduce costs, the various risks of PFI projects must be borne by those best able to shoulder and control them.

154. Although PFI guidelines are clear in directing that the apportionment of project risk should strike an optimal balance between risk transfer and cost, grey areas are likely to arise in practice. For example, PFI projects that accord the supplier monopoly power could facilitate the shifting of cost overruns to users and dampen the incentive for efficiency gains. In addition, the formal transfer of risk might not be credible and the contingent budgetary liability large. It is unlikely, for example, that the government would fail to intervene to maintain a critical service. These possible deficiencies can be exaggerated, however, and need to be compared with the imponderable developments to which even traditional procurement is prone. It is hard to see why a monopoly created under a PFI project could not be regulated as well as any other. Similarly, if a PFI provider fails to perform in the midst of a long-term contract, the public sector will face unanticipated costs, but it is not clear that these costs would necessarily be greater than the costs of a similar failure under a traditional procurement.

155. Proper monitoring of the recurrent costs to the budget of PFI projects is critical. The impending establishment of a system for monitoring forward PFI commitments should effectively rebut the criticism that the PFI is merely a buy-now-pay-later sleight of hand. The proposed introduction of an accruals-based system of budgeting would level the playing field between traditional and PFI projects while also contributing to the more efficient use of public assets.

156. Although the pace of PFI project identification and implementation has been slow, these delays have not yet had a material impact on overall investment levels, owing to the small amounts of PFI spending budgeted so far. Given the authorities’ ambitions for the PFI, delays in implementation could eventually adversely affect investment spending with little chance for public spending being able to fill the gap. The main administrative sources of delay include inadequate project specification and screening before tendering for bids. Contract terms designed to apportion risk between the private and public sectors have proven difficult to negotiate and complicated to structure. In addition, bid evaluation for PFI projects is more complicated than traditional procurement given the extra dimension, in the case of the PFI, created by the need to assess varying tradeoffs between cost and risk transfer. These administrative bottlenecks should gradually ease as familiarity with the PFI increases, and as a result of recently revised procurement guidelines and measures to improve the institutional support to PFI projects.


  • Confederation of British Industry, 1996, Private Skills in Public Service - Tuning the PFF (London: CBI).

  • Efford, Don, 1996, “The Case for Accrual Recording in the IMF’s Government Finance Statistics System”, IMF Working Paper WP/96/73 (Washington: International Monetary Fund, July).

    • Search Google Scholar
    • Export Citation
  • Heald, David, 1996, “The Private Finance Initiative: Value for Money and Public Expenditure Controlin House of Commons Treasury Committee, The Private Finance Initiative, Sixth Report (London: HMSO).

    • Search Google Scholar
    • Export Citation
  • House of Commons Treasury Committee, 1996, The Private Finance Initiative, Sixth Report (London: HMSO).

  • The Economist “Cooking the Books”, October 28, 1995, pp. 6566.

  • Institute for Fiscal Studies, 1995, Options for 1996: The Green Budget, Commentary No. 50 (London: Institute for Fiscal Studies).

  • Organization for Economic Cooperation and Development, 1993, “Accounting for What? The Value of Accrual Accounting to the Public Sector”, Occasional Paper on Public Management, Vol. 1, No. 16 (Paris: OECD).

    • Search Google Scholar
    • Export Citation
  • Oxford Economic Research Associates Ltd., 1996, Infrastructure in the U.K.: Public Projects and Private Money (Oxford: OXERA Press).

  • H. M. Treasury, 1992, Autumn Statement 1992 (London: HMSO).

  • H. M. Treasury, 1995, “Better Accounting for the Taxpayer’s Money: Resource Accounting and Budgeting in Government” (London: HMSO).

    • Search Google Scholar
    • Export Citation
  • H. M. Treasury, 1996, “The Private Finance Initiative: The Government’s Response to the Sixth Report from the [Treasury] Committee in Session 1995–96” (London: HMSO, June).

    • Search Google Scholar
    • Export Citation
  • H. M. Treasury and Private Finance Panel, 1995, Private Opportunity, Public Benefit: Progressing the Private Finance Initiative, (London: HM Treasury).

    • Search Google Scholar
    • Export Citation
  • H. M. Treasury and Private Finance Panel, 1996, Private Finance Initiative: Guidelines for Smoothing the Procurement Process [April].

  • Wolf, Martin, 1996, “A Clever Spending Wheeze”, Financial Times, July 16, p. 18.


Prepared by David J. Ordoobadi.


See, for example, footnote 3 above and numerous press reports, including the Economist (1995) and Wolf (1996).


As discussed further in Section B, this debate was sparked by early claims for the PFI.


In light of its implications for accountability for and control over spending decisions, this issue is explored in some detail in House of Commons Treasury Committee (1996).


Oxford Economic Research Associates Ltd. (1996) and House of Commons Treasury Committee (1996) provide examples of the higher financing costs of privately financed infrastructure projects.


The House of Commons Treasury Committee (1996, p xiv) states, for example: “There is no obvious reason why the benefits of better design could not be obtained under traditional procurement.”


Confederation for British Industry (1996), Oxford Economic Research Associates Ltd. (1996) and private sector PFI participants in testimony to House of Commons Treasury Committee (1996) review obstacles to PFI implementation in some detail.


The 1992 Autumn Statement (H. M. Treasury (1992, p. 27)), for example, indicated that: “…privately financed spending will be additional to public provision. The spending which is financed by Government will come out of departmental spending allocations. This is consistent with the principle that the control total should cover the spending which the public sector undertakes, or which it controls.”


The limitation on the need to assess a PFI project against a public comparator is intended to ensure that, in contrast to the Ryrie rules, a purely hypothetical public sector alternative with no chance of implementation does not derail the PFI investment.


While it is true that cost overruns will be borne by the private supplier in the case of PFI projects, this transfer of risk and the changed incentives that it engenders make it unlikely that PFI bids will systematically underestimate project costs, suggesting that the capital savings on PFI projects should be considerably less than past cost overruns in traditionally procured investments.


H.M. Treasury (1996) indicates that this monitoring will provide data on both private sector capital spending under the PFI and associated public sector forward commitments, with the view to ensuring that the latter is consistent with sound budgeting overall and, on the departmental level, does not compromise a department’s ability to undertake other priority spending, including capital spending not suited for the PFI.


The distinction between financing and operating leases hinges on which party—lessor or lessee—effectively bears the risk and reward of ownership of the leased asset. The terms of a financing (operating) lease place effective ownership with lessee (lessor).


Fully expending an asset likely to be productive over a number of years is a major shortcoming of cash accounting. A switch to accrual accounting would address this deficiency and result in a more level playing field between PFI projects and the traditional approach to procurement.