GOVERNANCE OF PPP PROJECTS THROUGH CONTRACT PROVISIONS Michael J. Garvin 1

Transcription

1 ABSTRACT GOVERNANCE OF PPP PROJECTS THROUGH CONTRACT PROVISIONS Michael J. Garvin 1 Contracts serve as the vehicle for tangibly distributing benefits and risks in public-private partnership (PPP) arrangements, typically for the better part of 30 years or more. Ideally, the contractual framework is structured to balance the interests of the public and private sectors to promote reasonable outcomes for both parties. Yet, uncertainty with time and the limitations of contracts challenge the efficacy of PPP arrangements since unpredictable, incalculable events are inevitable. Thus, contracts cannot possibly account for every potential contingency. A body of literature, which has investigated the contractual frameworks associated with PPPs, provides the foundation for a longitudinal, transnational case study of highway PPP projects. The principal intent of the case study is to examine specific contract provisions to gauge their incentive structures, specification practices and flexibility to ultimately identify practices that address, as effectively as possible, the inevitability of uncertainty and the limitations of contracts. The basic research design is explained and some illustrative results are reported. KEYWORDS: Contracts, Infrastructure, Public-Private Partnerships, Risk, Uncertainty INTRODUCTION Infrastructure public-private partnerships (PPP) are an outgrowth of the contemporary movement in government to alter the public service delivery paradigm. Risks and contracts are central to these arrangements. In fact, the PPP paradigm is founded, rather significantly, upon the concept of risk allocation through contracts. The transfer of risk through the contractual framework often is the basis for the decision to deliver public services by a PPP arrangement and movement of assets off the public sector balance sheet. Given their long-term nature, however, uncertainty, defined here as that which is unknowable or incalculable ex ante, is bound to introduce problems in the contractual framework. This fact has not escaped the attention of the academic community. Many researchers, examining the United Kingdom s Private Finance Initiative (PFI) and others focused on PPPs more generally, have scrutinized the suitability of a contractual framework for the delivery of public services via a PPP arrangement. This body of work underpins the research proposed here. Through a longitudinal, transnational case study, this initiative intends to identify contractual practices that address, as effectively as possible, the inevitability of uncertainty and the limitations of contracts. BACKGROUND The Evolving Public Services Paradigm The increasing role of the private sector in the provision of public services has been a defining trend over roughly the last two decades worldwide. In industrialized countries, this issue has been primarily a matter of public policy while in developing regions one might argue that it has been generally a mode of constructive intervention to affect modernization. Regardless, opinion 1 Associate Professor, Myers-Lawson School of Construction, Virginia Tech, Blacksburg, VA USA; garvin@vt.edu 1

2 and empirical evidence suggest this trend shall continue. For instance, a survey of governments in the late 1990 s indicated that the most effective form of government in 2010 would have the private sector providing traditional public services (Economist Intelligence Unit 1999), and countries like Spain expect to fund roughly 40% of future transportation investments from nonbudgetary sources (Garvin et. al. 2009). Grout and Stevens (2003) define a public service as any service provided for a large number of citizens where the potential for market failure justifies government involvement through production, finance or regulation. As they contend: A major problem with any framework to deliver public services is that those delivering the service can have far better information than the government agency. This disparity allows them to pursue goals that may not fully coincide with society s objectives. The old fashioned model of simply creating a publicsector agency that would be expected selflessly to pursue the required objective is no longer accepted without question. It is essential to design activities to elicit correct information from agents responsible for delivery, and to put in place structures so that the incentives facing the agents coincide with society s objectives. Of course, this is far from easy (pp ). Their perspective explains, in part, the trend toward private participation while offering insights regarding the challenges of public service provision in general. 2 In the simplest terms, the organization responsible for public service provision, whether public or private, cannot suffer from the classic agency problem. The Efficacy of PPPs Infrastructure public-private partnerships (PPP) have emerged amidst this broader contemporary move to involve the private sector more deeply in public service provision. A distinguishing characteristic of a PPP as opposed to other forms of private participation such as privatization is the bundling of service provision into a single long-term contract between the public and private sectors (Grout and Stevens 2003; Bennett and Iossa 2006; Engel et. al. 2008). Hence, Grout and Stevens would likely characterize the public role in PPPs as one of regulator via contract. Further, the length and terms of these contracts effectively, if not legally, grant the private party ownership, or control rights. 3 This characterization of PPPs has prompted significant theoretical attention to their efficacy as a strategy for the delivery of public services. 4 Grout (1997) was one of the first to assess and comment upon the United Kingdom s Private Finance Initiative (PFI), which in essence is the country s PPP program. His assessment took issue with how the Value for Money analysis of that time priced risks passed to the private sector and, more importantly, discussed the suitability of the PFI structure. The justification for the PFI must be that the model where assets are owned by the private sector even when the public sector is the client is superior in some way to others (p. 63). He argued that if contracts could always be written to replicate any reward scheme, then it would not matter if design, construction, service provision, and ownership were supplied by a single entity or supplied by different entities with public 2 Certainly, the general desire to move assets of the public sector s balance sheet has and continues to be a driving factor in private provision as well. 3 Here, the important distinction is the degree of control that the private party has over the asset and its related services as opposed to asset ownership in the strictest sense. 4 PPPs have received substantial attention in the literature. See Garvin (2009) for a broad categorization. 2

3 ownership. This argument, however, presumes the existence of complete contracts to cover any and every contingency or eventuality. The fact of the matter is that design and construction contracts are often problematic and have information asymmetry problems. He suggested that the case for the PFI structure depends on writing contracts over the flow of services from infrastructure rather than over the design and construction process. Grout succinctly identified the strength of this argument: A single entity that designs, constructs and sells the services but is only compensated for successful service supply has little to no incentive to reduce quality even if the consequences are latent If this entity cannot push cost overruns to the client, then strong incentives exist to reduce costs without jeopardizing quality If technology choices are available, then the entity has ample incentive to design and construct an appropriate asset since revenues solely depend on the provision of quality services from the asset. The power of this line of reasoning, however, is diffused by several factors. The quality of the design and construction phase depends heavily on the nature of the service requirements and contract, which is often executed prior the start of design and construction; hence, the need to monitor the build phase will remain since limitations in the service contract are inevitable To limit expropriation, contracts must be long-term but this limits flexibility. Certainly, renegotiation can minimize this problem, but this introduces inefficiencies The service requirements need specification in advance, which are difficult at best to contemplate years hence Correct pricing of the service causes the usual problems of divergence between social and private costs to arise even when the customer is the public sector. For example, once prices have been set for road usage, a contractor may have incentives to encourage incremental road usage and to divert location of other construction... (p. 64). Grout concluded his assessment by suggesting that the PFI model is most sensible when service contracts are easy to write and build contracts are difficult. However, the PFI model is not out of the question when build contracts are easy to write and service contracts are less so, but the public sector will need to remain involved in the build process and the rationale for ownership or control rights to remain with the private sector are less clear. Hart (2003) continued the discussion regarding the boundaries between public service provision by either a public or private organization by drawing parallels between theories of the firm and privatization. In particular, he noted that the issues of vertical integration and privatization have more in common than not. Both are concerned with whether it is better to regulate a relationship via an arms-length contract or via a transfer of ownership (p. C70). Hart developed an economic model, which ignored the ownership question, and compared a bundled approach using a single contract with a private party for facility construction and service provision with a conventional approach using separate contracts with private parties for facility construction and service provision. 5 The intent was to evaluate the general issue that bundling encourages productive investments such as efficiency enhancements that raise the quality of 5 Private O&M contracts for infrastructure facilities are relatively common in the United Kingdom and other European countries. 3

4 services as well as unproductive investments to cut costs and quality, albeit while adhering to the letter of the contract. Hart s model reinforced Grout s earlier assertions; In a case where specifying quality in the building process was relatively easy but doing the same for service provision was not, unbundling construction from service provision appeared more appropriate. In this instance, unbundling did not promote underinvestment in elements that make a facility more aesthetically pleasing and easier to operate whereas bundling tended to prompt overinvestment in initiatives that reduce total costs and overall quality. Conversely, in a case where specifying quality of service was rather easy (or more generally where good performance measures could be used to either reward or penalize) but specifying building quality was not, bundling was advantageous. In this instance, unbundling produced underinvestment in aesthetics and operational efficiency while bundling did not generate overinvestment in cost or quality reduction initiatives. Hart concluded with an interesting observation, One of the (modest) benefits of the current paper is that it may shift attention from what seem to be secondary financing issues to what seems to be the central issue: (relative) contracting costs (p. C75). Bennett and Iossa (2006) extended the work by Hart (2003) by: (a) studying two defining characteristics of PPP models: whether it is optimal to bundle the different stages of production and whether control rights should be given to the private partner and (b) examining the impacts of the residual value and the ownership of the facility once the contract expires. 6 Their analysis concluded: Bundling or unbundling does play a critical role in design and construction innovation. Innovations may be associated with a reduced cost in the operations or service provision phase, which they term a positive externality. For example, investment in technology can reduce labor requirements such as automated tolling which eliminates the need for toll collectors. Innovations may also be associated with increased costs, which they term a negative externality. For example, investments in technology can improve service quality but can increase operating and maintenance costs such as improved HVAC and lighting systems in a hospital that lead to better clinical service but are more costly to operate and maintain. The former innovation is likely associated with bundling since the positive externality is internalized whereas the latter innovation may be more likely in unbundling. With a positive externality, private control rights, or ownership, are not necessarily optimal. In some cases, the government should retain ownership. When private gains are possible in the absence of further incentives, private control rights are preferred. In general, the PPP model is more likely to be preferred (a) the more positive (or less negative) is the externality; (b) the stronger the effects that innovations in building and management have on the residual market value of facilities; and (c) the weaker the effect that innovations have on the benefit from provision of public services (p. 2159) Automatic transfer of facilities back to the public at the end of the contract period reduces ex ante investment incentives whereas the option to negotiate a mutually beneficial transfer should heighten investment incentives. Froud (2003) characterized PFI as public services management that embraces competition, efficiency and contractualism. She further argued that PFI uses the logic of 6 This work differentiates itself from Hart s since it accounts for the possibility of renegotiation, ownership at the end of the contract, and facility residual value. 4

5 contractualized risk transfer, which raises serious questions regarding risk identification and measurement. In addition, she probed whether a contract can serve the public s interest and suggested that current practices in effect ignore uncertainty, which has significant implications. Clearly, risk is the key feature of the PFI program (and one could argue more broadly of PPPs generally) since it legitimates the shift in public services management; risk is central to a value for money analysis and a determination, quite often, of whether infrastructure assets remain on the public sector s balance sheet. 7 Perhaps the most meaningful insight from Froud was her assertion that uncertainty is misunderstood, and therefore, neglected in the overall PFI program. She essentially distinguished risk and uncertainty as the calculable from the unknowable. The significance of this distinction to a program that relies upon contracts is the impact that current contractual decisions and specifications can have on an uncertain future. In other words, tomorrow is shaped in part by the decisions and actions of today. As a consequence of this flaw, Froud correctly pointed out that often the public sector retains more risks or pays a higher premium for risks than anticipated through the creation of new risks via the contractual framework. This occurs, in part, because the contract: (a) locks the public-sector into a service provision regime where changes require negotiation and cost and (b) cannot anticipate unknown or does not acknowledge particular risks so they, as a consequence of contractual rigidity, remain with the public sector. She stated: The implicit value from the security of the PFI contract to the private sector is not incorporated into the appraisal of risks, just as the constraint imposed upon the respective department of government through the ringfencing of PFI payments is not explicitly acknowledged in weighing up potential risks and costs (p. 581). Froud further advanced the limitations of contracts in general and contractualized risk transfer specifically to conclude that the state ought not to bind itself by this approach to the delivery of public services. Instead, government should retain the power and the flexibility to engage with positively and to help shape an uncertain future (p. 587). Grout and Stevens (2003) examined broadly the different arrangements for the delivery of public services, recognizing both the possibilities and the realities. A significant part of their assessment was the increasing role of the private sector. Ultimately, their conclusions reinforced the discussion previously regarding contracting, the crux is related to where quality is most effectively defined and measured. In addition, however, their analysis indicated that motives of those involved in delivery and the incentive structures in place are just as critical, if not more, as the choice of whether a public or private organization should hold the responsibility. Guash (2004) made an important contribution by examining renegotiation of concession agreements in Latin America. While the institutional environment in this region differs markedly from developed nations, the propensity for opportunistic or coercive behavior and the incompleteness of contracts are global issues. He defined renegotiation as a significant contractual change or amendment not envisioned or driven by stated contingencies in any of the following areas: tariffs, investment plans and levels, exclusivity rights, guarantees, lump-sum payments or annual fees, coverage targets, service standards, and concession periods (p. 12). His study of roughly 1,000 concessions across several infrastructure sectors found: (a) that the rate of renegotiation varied by sector, with water & sanitation and transportation having the highest rates by far, (b) concessionaires initiated renegotiation three times as often as the government while mutual initiation occurred least frequently, (c) far less renegotiation occurred 7 In Europe and elsewhere, the off-balance sheet test depends upon the level of risk transfer, which is different from the standard in the U.S. See pp of Yescombe (2007) and Buxbaum and Ortiz (2007) for further discussion. 5

6 in non-competitive awards, which was explained as a consequence of bilateral negotiation ex ante that resulted in more favorable terms for the private party, (d) private sponsors were far more likely to initiate renegotiation under a price-cap regulatory regime, (e) renegotiation occurred more frequently when the regulatory framework was embedded in the contract as opposed to in a decree or law, (f) the typical private party argument for renegotiation was an imbalance in the financial equilibrium, and (g) the public arguments for renegotiation included altered priorities, political issues, and contractual non-compliance. Guash argued that renegotiation can be a positive instrument when it addresses the inherently incomplete nature of concession contracts (p. 19). If renegotiation only occurs when called for by the contractual contingencies specified or by major unexpected events, then a competitive award should identify the most effective concessionaire since the likelihood of opportunistic renegotiation is minimized thus eliminating not the most efficient provider, but the one most skilled at renegotiations (p. 19). Guash concluded by offering various recommendations for contract design and regulation. Notably, he suggested that contracts should clearly define the treatment of assets, outcome indicators, and procedures and guidelines to adjust and review tariffs; contracts should contain clear triggers for renegotiation actions; claims for renegotiation should be transparently examined and any subsequent changes must be plainly explained to the public; and an appropriate regulatory framework and agency should be in place prior to concession award, with sufficient autonomy and implementation capacity to ensure high-quality enforcement and to deter political opportunism (p. 21). Grimsey and Lewis (2007) presented a thorough review of PPPs examining their potential benefits as well as limitations. Interestingly, one of the first issues they tackled is related to incomplete contracts. They recognized that long-term arrangements potentially limit government flexibility; however, they contended that a conventional delivery (i.e. unbundled separate contracts) is executed in an environment largely removed from the economic signals to which private entities are exposed... and the principals involved are often insulated from the consequences of their actions and decisions (p. 172). Certainly, infrastructure decisions have consequences across generations, but as Stone (2006) indicated politicians and government officials making such decisions bear little personal responsibility for their consequences and so have little incentive to change their behavior. Conversely, a well structured PPP arrangement can introduce clear lines of accountability, transparency of outcomes and performance, clarity as to the roles and responsibilities of the contracting parties, an assessment of project risks, competition for the delivery of services, and the motivation to succeed (p. 172). Grimsey and Lewis cited three specific drivers of productive efficiency in PPP arrangements: (a) ownership rights, (b) bundling and (c) risk transfer. All of these elements have already been discussed. They also identified several limitations of PPPs. The most significant include: (a) incomplete risk transfer, (b) inflexibility and (c) access. The first two were discussed previously, but the access problem introduces complex challenges. In essence, ad-hoc implementation of PPP contracts without proper attention to their relationship with the broader network of infrastructure can create perverse incentives and unintended consequences. Finally, Vining and Boardman (2008) provided eight rules of government concerning the implementation and administration of PPPs. Their rules were derived principally from transaction cost economics and a positive theory perspective of these arrangements. In particular, their positive theory approach recognized that the objective functions of the two partners are divergent, so goal conflict is not surprising. Further, the public sector s objective function has limited correlation with enhancing social welfare. The public sector, or 6

7 government, objective function is to minimize the sum of current government expenditures, public liabilities (or debt), and political costs. The private sector wishes to maximize profits through time and at a risk-adjusted rate. Three of their eight rules are particularly relevant to the current discussion: (a) separate the agencies that analyze, evaluate, contract/administer, and oversee PPPs (a rule that echoes one of Guash s earlier suggestions); (b) be wary of projects that exhibit high asset-specificity, are complex or involve high uncertainty, and where in-house contract management effectiveness is low; and (c) prohibit the private-sector contractor from selling the contract too early. The first rule may introduce inefficiencies but it will preclude incentivizing throughput or deals. The second is highly correlated with prior discussions, although asset-specificity is rather inherent in infrastructure assets. The final rule helps maintain the synergies of bundling discussed earlier since a contract sold early in the service provision stage is basically equivalent to an unbundled arrangement. Implications for PPP Arrangements Certainly, the literature just reviewed is not exhaustive. However, it is representative of the longitudinal research that has examined the general efficacy of PPPs. So what should we make of this work? Foremost, PPPs are challenging procurement and contractual exercises. While this may be stating the obvious, whether the decision to initiate a PPP and the structure of a PPP are correct, in nearly any context, is far from obvious. Risks, uncertainty, incomplete contracts, information asymmetry, opportunism and perverse incentives complicate matters significantly. This literature, however, has several common and interrelated themes that deserve attention. Incentives: proper internal organizational and external service provision incentives are more important generally than which sector holds the control rights or ownership, except perhaps in a case where significant residual value may exist at the end of a term under a private service provision scenario. Alternatively, the political environment of infrastructure can decouple decision-makers from latent consequences, which is essentially an issue of moral hazard. Bundling: coupling design and construction with service provision: (a) is preferable when service provision specification is easier than construction specification, (b) tends to promote innovations that reduce operating & maintenance costs as opposed to innovations that heighten service quality and increase operating & maintenance costs, and (c) is likely to induce innovations when significant, claimable residual value may exist at the end of the contract term. Contracts: the incompleteness of contracts will introduce inefficiencies and costs, particularly in long-term arrangements where project uncertainty is substantial; further, ex ante contractual conditions will influence many facets of an arrangement since such conditions transform the future and affect participant behavior. Regardless, long-term contracts may work as effectively as any other public service delivery arrangement if structured, monitored and enforced cognizant of opportunism and uncertainty. Those who have followed the evolution of the PPP market will recognize the impact of this literature upon contemporary PPP practices. The emphasis placed upon service outcomes and performance measures has increased substantially over the last decade as well as the attention given to accommodating and streamlining the contract change process. Indeed, these notions are worthy of consideration when structuring PPP contracts, and they have implications 7

8 for the push to standardize contracts. While standard provisions can serve both sectors well by reducing the transaction costs associated with closing an agreement, standard conditions may need to vary across infrastructure sectors and supplementary conditions that fit the context of each project may also be necessary. PRELIMINARY RESEARCH DESIGN The background literature has become the basis for a longitudinal and transnational case study of PPP contractual frameworks in highway projects. This initiative ignores the philosophical and policy-oriented question of whether PPPs are appropriate for public service delivery; instead, it focuses upon examining the contractual structures of PPPs assuming that governments will continue to adopt this approach. Table 1 illustrates the project contracts that are serving as the cases for analysis. Project Pair Location Services Date Term (years) Pocahontas Parkway I-495 HOT Lanes Virginia, USA Virginia, USA D+B+F+O D+B+F+O Variable 80 A1(M) M25 United Kingdom United Kingdom D+B+F+O D+B+F+O CityLink EastLink Victoria, Australia Victoria, Australia D+B+F+O D+B+F+O Table 1. Research Case Studies The primary intent of the research is to examine each project s contractual framework through content analysis to evaluate specific provisions and schedules to assess incentive structures, specification practices and flexibility. In addition, assessment of the overall project environment will determine if unique project conditions had any correlation with the contract provisions adopted. Table 2 illustrates the project conditions and provisions of interest. Organizational learning is also gauged by examining projects let by the same public sector agency in different timeframes. The goal is to identify those provisions that effectively address the limitations of contracts and promote the outcomes desired from PPPs. Project Conditions Project scope Public agency characteristics Main sponsors characteristics Funding mechanism o Tolls o Government payment Financial structure Procurement process Contract Provisions Capital delivery management Competing facilities or proximate work Funding mechanism management Changes or contract modification Termination or suspension Service provision performance measurement Handback Table 2. Project Conditions and Contract Provisions One may view the project conditions as the independent variables and the contract provisions as the dependent variables of the investigation. For instance, the characteristics of a project s public agency and main sponsors along with the procurement approach implemented are likely to influence the terms and the regulatory framework adopted in a contract. An inexperienced and vertically integrated PPP division (even if complemented with external advisors) paired with an experienced, multinational sponsor team in a non-competitive bilateral negotiation is likely to produce a contractual framework that differs from one where the public 8

9 agency is experienced and segregated (to include an autonomous regulatory/auditing unit) and a competitive procurement process is utilized. Further, a highway project funded by tolls will present different risks and managerial issues for a contract than a project funded by periodic government service payments. In either case, for example, the competing facilities or proximate work clauses are significant but their terms should differ. Finally, the financial structure can bring alternative players with differing norms and expectations. A project financed substantially with debt obtained from a syndicate of lending institutions introduces actors with practices and motives dissimilar from debt acquired from investors in project bonds. These respective financial structures will influence certain contract provisions. The contract provisions identified were selected not only for their potential correlation with project conditions but also due to their inherent relation to the defining characteristic of most PPPs: the bundling of services in a long-term contract. For example, the literature reviewed has suggested that the bundling of services is most sensible where specifying service provisions is more easily done than specifying building provisions. Further, bundling tends not to promote investments in service quality but rather investments that produce cost savings. Hence, the structure of the capital delivery management clauses (building specifications) and the service provision performance measurement clauses (service specifications) will provide an indication of the nature of the respective specification practices. In addition, the character of the performance measurement specifications may indicate whether or not the public agency is attempting to incentivize service quality investments that might not otherwise occur. Table 3 explains the briefly rationale for the contract provisions selected. Contract Provisions Rationale for Examination Capital delivery management When bundled with service provision, specification structure should theoretically encourage cost reduction during service provision Strict specification of capital delivery may indicate inadequate service provision specifications or lack of public sector experience with PPP model Auditing burden could present information asymmetry issues or dispute potential Competing facilities or proximate work Flexibility or lack thereof afforded to public sector could incentivize certain private or public sector behavior Over contract term, triggers/milestones for review/modification may be needed Funding mechanism management Specification structure logically introduces incentives Over contract term, structure can be amenable to changes in expectations or triggers/milestones for review/modification may be needed Auditing burden could present information asymmetry issues or dispute potential 9

10 Contract Provisions Rationale for Examination Changes or contract modification Governs the general capacity of the contract to adapt as uncertainty resolves Termination or suspension Specification structure may incentivize opportunistic behavior by the public sector or hold-up behavior by the private sector, with the former being more probable Service provision performance measurement Advantages of bundling depend on effectiveness of specification and focus on outcomes Specification structure logically introduces incentives Over contract term, metric structure can be amenable to changes in expectations or triggers/milestones for review/modification may be needed Auditing burden could present information asymmetry issues or dispute potential Handback Specification structure logically introduces incentives Significant, claimable residual value (or perhaps remuneration) may incentivize innovative investments or practices Auditing burden could present information asymmetry issues or dispute potential Table 3. Rationale for Selection of Specific Contract Provisions Not surprisingly, analysis of the contract provisions will focus upon incentives, specification structure and flexibility. The emphasis on each analytic likely depends on the provision. For instance, the focus relative to competing facilities provisions is primarily one of flexibility. The extremes are absolute protection afforded to the private partner in a regional area from any public sector improvements and absolute freedom afforded to the public sector to initiate improvements in a regional area. Discrete conditions between these extremes are certainly plausible. These provisions shall be analyzed to categorize them along this continuum. Similarly, the focus relative to service provision performance clauses is principally one of specification structure. Certainly, these clauses also establish important incentives and have flexibility characteristics. Specifications may be generally categorized as prescriptive or performance-based; further, a significant distinction is whether they are customer or asset focused. As the literature suggests, the former are more suitable for bundled arrangements. Further, service provision clauses that relate to macro transportation objectives like trip reliability and safety are likely more sensitive to changes in expectations over the longterm than clauses that relate to asset condition. As the research progresses, the possibility certainly exists to code project conditions and contract provisions into binary or discrete conditions, which might then allow analysis through configurational comparative methods (Rihoux and Ragin 2009) 10

11 I-495 HOT LANES A brief overview of the I-495 High Occupancy Toll (HOT) Lanes in Virginia and some illustrative results follow. Overview of Project The project will build fourteen miles of new HOT lanes (two in each direction) on I-495 between the Springfield Interchange and just north of the Dulles Toll Road in Northern Virginia. These HOT lanes will allow the Beltway to offer HOV-3 connections with I-95/395, I-66 and the Dulles Toll Road for the first time. In addition by providing new travel choices, this project will also improve the Beltway s 45-year-old infrastructure, replacing more than 50 aging bridges and overpasses, upgrading 10 interchanges and introducing new bike and pedestrian access. When completed, buses, carpools and vanpools with three or more people, and motorcycles can ride in the new lanes for free. Vehicles carrying less than three people can either travel for free in the regular lanes, or pay a toll to ride in the HOT lanes. Tolls for the HOT lanes will change according to traffic conditions, which will regulate demand for the lanes and keep them congestion free. The project will be electronically tolled using transponder technology. In 2002, the Virginia Department of Transportation (VDOT or the Department) received an unsolicited PPT conceptual proposal from Fluor Daniel to develop, finance, design and construct High Occupancy Toll (HOT) Lanes on the Capital Beltway. Although VDOT advertised for competing proposals, none were received. In the spring of 2003, VDOT submitted a grant application to the Federal Highway Administration to study HOT lanes and other "value pricing" applications in Northern Virginia. In September 2007, the Capital Beltway Express LLC (the Concessionaire), formed by Fluor and Transurban, and VDOT reached an agreement in principle for the design, construction, operations and maintenance of the Capital Beltway HOT Lanes. This comprehensive agreement was finalized on December 20, Under this agreement, the Department will own and oversee the HOT lanes and the Concessionaire will construct and operate them. The total length of the concession is 80 years - 5 years of construction and 75 years of operation. A total of $1.93 billion in capital is necessary for development; financing of the project consists of: (a) $587 million senior debt in PABs; (b) $587 million TIFIA loan; (c) $350 equity (Transurban 90% and Fluor 10%); and (d) $409 VDOT funds. The internal rate of return (IRR) is projected to be 13% when the road begins operation, and the concession includes a revenuesharing agreement with VDOT, under which it will receive a portion of the gross revenue once certain levels of return are met. VDOT s entitlement starts at 5% when IRR is over 12.98%, rising to 15% when IRR is over 14.5%, and 40% when the IRR exceeds 16%. Transurban will also receive 1% of the net asset value of the concession as a base management fee. Illustrative Results Project Conditions Ultimately, the project was the result of a bilateral negotiation between VDOT and a multinational sponsor team. The Innovative Project Delivery (IPD) division, which is a vertically integrated planning, procurement and contract management unit for Virginia s Public- Private Transportation Act (PPTA) projects, was principally responsible for handling the procurement and negotiation. A preliminary assessment indicates that IPD had low to modest experience with PPP transactions, but it did engage a significant team of advising consultants to supplement its expertise. The sponsor team had modest PPP sponsorship experience, significant 11

12 domestic design and construction experience and little managed lanes operations experience (but modest toll road operations experience). The financial package was a mix of state funds, a federal loan instrument, tax-exempt bonds, and sponsor equity. The proportion of private equity to debt (excluding the TIFIA loan) is roughly 40% to 60%. Handback Provisions Table 4 lists the handback provisions in the Amended and Restated Comprehensive Agreement between VDOT and Capital Beltway Express. Contract Clauses Upon the end of the Term, the Concessionaire shall hand-back the HOT Lanes Project to the Department, at no charge to the Department, with asset condition having a remaining life of the greater of: (i) five years; or (ii) life within its normal lifecycle. In addition, if requested by the Department, the Concessionaire will dismantle the HOT Lanes toll system as required to convert the HOT Lanes back to GP Lanes; provided that the Department shall notify the Concessionaire at least one year prior to the end of the Term if the HOT Lanes are to be converted back to GP Lanes. Any such dismantling of the HOT Lanes toll system shall be at Concessionaire s sole cost and expense. Beginning 20 years prior to the expiration of the Term and every five years thereafter, the Concessionaire, the Department, and the Independent Engineer will jointly conduct inspections of the HOT Lanes Project, for the purposes of jointly (i) determining and verifying the condition of all HOT Lanes Project assets and their residual lives, and (ii) determining, revising and updating the Life Cycle Maintenance Plan. Beginning five years prior to the expiration of the Term, the Concessionaire, the Department and the Independent Engineer will jointly conduct annual inspections of the HOT lanes to ensure that the Handback Requirements will be met. The Concessionaire shall diligently perform and complete all work contained in the Life Cycle Maintenance Plan prior to reversion of the HOT Lanes Project back to the Department, based on the required adjustments and changes to the Life Cycle Maintenance Plan resulting from the inspections and analysis. The Concessionaire shall complete all such work prior to the end of the Term. Starting five years prior to the expiration of the Term, the Concessionaire shall post a ten-year irrevocable stand-by Letter of Credit or a Performance Bond to the Department for a period of five years after expiration of the Term in an amount equal to 50% of the nominal lifecycle cost expended in the previous five years of the Term pursuant to the most recent Life Cycle Maintenance Plan approved by the Department. This Letter of Credit or Performance Bond would be drawn upon by the Department only in the event that subsequent to termination or expiration of the Term, the HOT Lane assets are found to fail to address the Handback Requirements and in the amount required to address such failures up to the full amount of the Letter of Credit or Performance Bond. The Department will determine whether the HOT Lane assets meet the Handback Requirements based on routine inspections up to five years after termination or expiration of the Term ( Handback Period ). If the Concessionaire disagrees with the Department s determination of the condition of the HOT Lanes during the Handback Period, the Concessionaire may, at its own expense, retain an engineer to inspect the facility and review the findings of the Independent Engineer. Resolution of the issue will be subject to dispute resolution process. Table 4. Handback Provisions in I-495 HOT Lanes Contract The specification structure of the handback provisions follows a typical approach in PPP arrangements where residual lives for assets are specified. Certainly, the specification of required residual life creates incentives (as well as options) for the private sponsors to manage the assets to meet or exceed these requirements; however, the approach introduces two challenges. First, the auditing burden is substantial it shall begin 20 years prior to the end of the term and the determination of residual life is subject to discretionary judgment and dispute. 12

13 Second, the private sponsors have no claimable value at the end of the term, but rather quite the opposite. The sponsors may be required to dismantle the toll system without remuneration, and they must post security letter of credit or a bond that is callable after the expiration of the contract term. Discussion While these illustrative results are in no way conclusive, a few observations are possible. Prior research suggests that non-competitive, bilateral negotiations of the type observed here may produce terms that favor the private sponsors. Future investigation and comparison to provisions with alternative project conditions may provide evidence of this sort. The impact of the mixed financing, particularly the introduction of the TIFIA loan, could be reflected particularly in the funding mechanism management and the termination or suspension clauses such as step-in rights afforded the debt holders. The handback provisions are compliance oriented and somewhat punitive in nature. The sponsors are incentivized to avoid financial losses as opposed to sustaining or retaining financial value created. Theory suggests this will not naturally entice innovative investments or practices. Their nature also heightens the significance and effectiveness of the auditing process to minimize information asymmetry and avoid conflict. FUTURE RESEARCH Clearly, this research is in its preliminary stages; however, the case study proposed holds promise for identifying features of contractual frameworks that may address, at least partially, the limitations of contracts in PPP arrangements. Past research has suggested that incentives, specifications and flexibility may dictate the capacity of a PPP contract to deliver as expected. Subsequent analysis of the contractual frameworks employed in both domestic and international projects will characterize and compare the governance practices in place. Practices will be assessed for generalization by examining their propensity to encourage proper incentives and reasonable flexibility. ACKNOWLEDGEMENTS The constructive comments provided by the paper s reviewers were quite beneficial and helped to focus attention upon the correlation between the background literature and the research design. REFERENCES Bennett, J. and Iossa, E. (2006). Building and managing facilities for public services. Journal of Public Economics, 90(2006), Buxbaum, J. and Ortiz, I. (2007). Protecting the public interest: the role of long-term concession agreements for providing transportation infrastructure. Research Paper 07-02, Keston Institute for Public Finance and Infrastructure Policy. Economist Intelligence Unit and Andersen Consulting. (1999). Vision 2010: forging tomorrow s public-private partnerships. New York. Engel, E., Fischer, R., and Galetovic, A. (2008). "The basic public finance of public-private partnerships." Working paper 13284, National Bureau of Economic Research. Froud, J. (2003). "The Private Finance Initiative: risk, uncertainty, and the state." Accting, Organizations and Society, 28(6),

14 Garvin, M. J. (2009). Enabling development of the transportation public-private partnership market in the United States. forthcoming in Journal of Construction Engineering & Management. Garvin, M. J. et. al. (2009). Public-private partnerships for highway infrastructure: capitalizing on international experience, Report No. FHWA-PL , Office of International Programs, Federal Highway Administration, Washington, D.C. Grimsey, D. and Lewis, M. (2007). Public private partnerships and public procurement. Agenda, 14(2), Grout, P. A. (1997). "The economics of the private finance initiative." Oxford Review of Economic Policy, 13(4), Grout, P. A., and Stevens, M. (2003). "The assessment: financing and managing public services." Oxford Rev. Econom. Policy, 19(2), Guasch, J. L. (2004). Granting and renegotiating infrastructure concessions: doing it right, World Bank, Washington, DC. Hart, O. (2003). "Incomplete contracts and public ownership: remarks, and an application to public-private partnerships." The Economic Journal, 113(486), C69-C76. Rihoux, B. and Ragin, C. C. eds. (2009). Configurational comparative methods: Qualitative comparative analysis (QCA) and related techniques, Sage: Thousand Oaks, CA. Stone, T. (2006). PFI is there a better way? KPMG Global Infrastructure and Projects Group, London. Vining, A. R. and Boardman, A. E. (2008). Public-private partnerships: eight rules for government, Public Works Management & Policy, 13 (2), Yescombe, E. R. (2007). Public-private partnerships: principles of policy and finance, Elsevier: Boston. 14