ARTICLE IN PRESS Availableonlineatwww.sciencedirect.com e ScienceDirect MANAGEMENT ELSEVIER International Journal of Project Management xxx(2006)xXX-XXX www.elsevier.com/locate/ijproma Risk allocation in the private provision of public infrastructure A. Ng. Martin Loosemore Faculty of the Built Entironment, University of New South Wales, Sydney, NSW 2052, Australie Received 7 March 2006: received in revised form 5 May 2006: accepted 20 June 2006 Abstract Communities benefit most from the private provision of public infrastructure when project risks are distributed appropriately between rivate and public sectors. This is not easy given the technical, legal, political and economic complexity of infrastructure projects and the range of constituencies involved. Too often, risks are under estimated and allocated to parties without the knowledge, resources and capabilities to manage them effectively. The result is increased costs, project delays and services which fail to deliver value-for-money to the community. This paper presents a case study of the controversial $920 million New Southern railway project in Sydney, Australia It analyses the rationale behind decisions about risk distributions between public and private sectors and their consequences. It also dem- onstrates the complexity and obscurity of risks facing such projects and the difficulties in distributing them appropriately. The paper concludes with a series of recommendations to better manage risks in such project C 2006 Elsevier Ltd and IPMA. All rights reserved. Keywords: Procurement; Private/public partnerships; Alliances; Risk management; Infrastructure; Community 1. Introduction However, all infrastructure projects share several common characteristic n most countries, the stock of public infrastructure rer resents an enormous asset, which effectively managed, 1. They are generally long lived and typically involve signif- plays a critically important role in attracting foreign invest- icant technical, legal, political and economic risks, long ment and supporting a nations social, cultural and eco- payback periods, high gearing and negative returns in nomic stability, productivity, development and early years. For example, the consortium which built Aus- prosperity. For example, in 1994, the US Public Accounts tralia's largest ever privately funded infrastructure pro- Committee predicted that continued reductions in US high- ject, the s1. 2 billion Melbourne City Link was given a way investment could, over 20 years, cause a 3.5% reduc 34 year concession period to operate the private toll road. ion in GDP, an 8% increase in inflation and a 2.2%/ 2. In fear of monopolies, public authorities sometimes increase in unemployment [1]. introduce excessive competition, regulation and control Typically, infrastructure projects can be divided into which can stifle innovation [a major reason for going two broad categories; economic infrastructure and social to the private sector]. For example, experience has infrastructure [2, 3]. Economic infrastructure projects shown that approval processes in such projects can be include bridges, drainage systems, sewage treatment plants, frustratingly lengthy and costly, government objectives telecommunications networks and road rail and air trans. can be unclear, tendering costs can be excessive, govern port facilities, etc. Social infrastructure includes education ment commitment can vary, abatements can be excessive prisons, health, tourism and recreational facilities, etc and usage rates of the final facility can be less than antic- ed[45] ing author.Tel:+61293856723:fax:+61293855613 E-mail address: m loosemore(unsw. edu. au(M. loosemore) tor operators, levels of community involvement and 0263-7863/$30.00 e 2006 Elsevier Ltd and IPMA. All rights reserved doi:l0.1016 l. ijproman.200606.005 Please cite this article as: A. Ng, Martin Loosemore, Risk allocation in the private provision of public infrastructure, International Journal of Project Management(2006), doi: 10. 1016/j. ijproman 2006.06.005
Risk allocation in the private provision of public infrastructure A. Ng, Martin Loosemore * Faculty of the Built Environment, University of New South Wales, Sydney, NSW 2052, Australia Received 7 March 2006; received in revised form 5 May 2006; accepted 20 June 2006 Abstract Communities benefit most from the private provision of public infrastructure when project risks are distributed appropriately between private and public sectors. This is not easy given the technical, legal, political and economic complexity of infrastructure projects and the range of constituencies involved. Too often, risks are under estimated and allocated to parties without the knowledge, resources and capabilities to manage them effectively. The result is increased costs, project delays and services which fail to deliver value-for-money to the community. This paper presents a case study of the controversial $920 million New Southern Railway project in Sydney, Australia. It analyses the rationale behind decisions about risk distributions between public and private sectors and their consequences. It also demonstrates the complexity and obscurity of risks facing such projects and the difficulties in distributing them appropriately. The paper concludes with a series of recommendations to better manage risks in such projects. 2006 Elsevier Ltd and IPMA. All rights reserved. Keywords: Procurement; Private/public partnerships; Alliances; Risk management; Infrastructure; Community 1. Introduction In most countries, the stock of public infrastructure represents an enormous asset, which effectively managed, plays a critically important role in attracting foreign investment and supporting a nation’s social, cultural and economic stability, productivity, development and prosperity. For example, in 1994, the US Public Accounts Committee predicted that continued reductions in US highway investment could, over 20 years, cause a 3.5% reduction in GDP, an 8% increase in inflation and a 2.2% increase in unemployment [1]. Typically, infrastructure projects can be divided into two broad categories; economic infrastructure and social infrastructure [2,3]. Economic infrastructure projects include bridges, drainage systems, sewage treatment plants, telecommunications networks and road, rail and air transport facilities, etc. Social infrastructure includes education, prisons, health, tourism and recreational facilities, etc. However, all infrastructure projects share several common characteristics: 1. They are generally long lived and typically involve significant technical, legal, political and economic risks, long payback periods, high gearing and negative returns in early years. For example, the consortium which built Australia’s largest ever privately funded infrastructure project, the $1.2 billion Melbourne City Link was given a 34 year concession period to operate the private toll road. 2. In fear of monopolies, public authorities sometimes introduce excessive competition, regulation and control which can stifle innovation [a major reason for going to the private sector]. For example, experience has shown that approval processes in such projects can be frustratingly lengthy and costly, government objectives can be unclear, tendering costs can be excessive, government commitment can vary, abatements can be excessive and usage rates of the final facility can be less than anticipated [4,5]. 3. Given the importance of infrastructure to the general public and the quality of service provided by private sector operators, levels of community involvement and 0263-7863/$30.00 2006 Elsevier Ltd and IPMA. All rights reserved. doi:10.1016/j.ijproman.2006.06.005 * Corresponding author. Tel.: +61 2 9385 6723; fax: +61 2 9385 5613. E-mail address: m.loosemore@unsw.edu.au (M. Loosemore). www.elsevier.com/locate/ijproman International Journal of Project Management xxx (2006) xxx–xxx INTERNATIONAL JOURNAL OF PROJECT MANAGEMENT ARTICLE IN PRESS Please cite this article as: A. Ng, Martin Loosemore, Risk allocation in the private provision of public infrastructure, International Journal of Project Management (2006), doi:10.1016/j.ijproman.2006.06.005.
ARTICLE IN PRESS A. Ng, M. Loosemore /International Journal of Project Management xxx (2006)xxx-xxx accountability are high, as are political pressures which During the operating period, the SPV receives income can interfere with the effective funding, management and based on the usage of the facility [which may be guaran- procurement of a project. For example, the Victorian teed] assuming that the service provided meets a range of Department of Treasury and Finance in Australia, key performance indicators. For example, the Colombian which annually procures over $1. 8 billion of public government agreed to reimburse the consortium which infrastructure, noted that the political and social context constructed the recent El Cortijo-El Vino Toll road in in which infrastructure projects are undertaken requires Colombia, if traffic was less than 90% of a specified level that public consultation be fully integrated into the and went even further in the new runway at Bogota's El planning process [6]. It argues that at the core of such Dorado Airport, guaranteeing a minimum revenue. Ther projects is a complex web of relationships among are normally abatement clauses in the concession contract bureaucrats, politicians, media, employees, general pub- which can penalise [sometimes excessively] the SPV for fall lic [local, national and sometimes global], labour and ing below these standards. Furthermore, there are some- special interest groups. Irrespective of any ideological times penalty points, which if accumulated to a certain preferences that Governments of the day may promote, level can lead to termination of the contract for poor per these interest groups invariably have high expectations formance. At the end of the operating period, the fully in relation to the management of issues such as the envi- operational project is transferred back to the host govern- ronment, health and safety, industrial relations and ment, usually at nominal or no cost [10]. access and equity. As Sharp [6]notes, any infrastructure project lives or dies on its reputation with these people. 3. Balancing public/private risks and rewards in PPP projec Despite the existence of many complex risks which interfere with the success of infrastructure projects, the pri- In investing in a PPP project, private sector companies vate sector has been keen to take over the traditional role aim to achieve a return on their investment in generating of the public sector in financing, procuring and managing sufficient future cash flows to cover initial capital costs such assets [5, 7, 8]. However, recent research has indicated and finance charges, thereby providing enough profit to that even on the largest Public, Private Partnership(PPP) invest in future projects and pay shareholder dividends projects, risk management practices are highly variable, In contrast, the aim of the public sector is to ensure a level intuitive, subjective and unsophisticated [4]. In this context, of service to the community which is timelier, more cost the aim of this paper is to explore the considerations to be efficient and higher quality than if the public sector had made in effective risk distribution between the public and retained responsibility. While some rather anecdotal evi private sectors on such projects. dence exists to indicate that PPP projects can better serve community objectives, than traditional public provision, 2. The structure of ppp agreements by reducing risk exposure, the debate is likely to continue until a sufficient number of projects have been studied in PPP is an evolving concept which takes many forms detail over their entire life-cycle. Nevertheless, the current around the world. However, it is essentially an arrange- arguments from each side of the fence are summarised ment by which private parties participate in, or provide below support for the provision of infrastructure-base services In contrast to traditional public procurement which 3. 1. Risks in the private provision of infrastructure involves the public sector purchasing an asset, the PPP sys tem involves the purchase of a stream of services, defined in It has been argued that concession contracts involve rel- detailed service agreement under specified terms and con- atively high waste, rework and transaction costs due to ditions [9, 12]. In simple terms, this is done via a concession lengthy and complex tendering arrangements and post-ten contract which involves a host government granting a der negotiations resulting from overly optimistic public sec- licence or concession to a private consortium [concession- tor comparators, large numbers of stakeholders and the aire, promoter or sponsor] which sets up a single purpose complex web of contracts and financial structuring needed entity known as a Special Purpose Vehicle(SPV)using con- to bind them together. It also has been argued that the com- tracts secondary to the concession, to finance, design, plexity of these arrangements increase public sector risk build, operate and maintain an infrastructure project for rather than reduce it, increases service costs for the public a set period of time known as the concession period. The and represent a barrier to entry to small companies which private consortium is normally formed by a joint venture is unfair and reduces competition [7 (V) between a range of organisations including contrac- Another argument against PPPs is that in some situa- ors, facilities managers, banks, investors and suppliers, tions, they are not economically viable for the private sec- which are willi ity and/or resources to tor without exorbitant risk-related service charges for the the project. Payments to the SPV to fund debt service nor- public. A good example in Australia are rural roads which mally commence after completion of the e construction do not provide the volume of traffic needed to justify a when the services have be made available to the public. user-pays approach which lies at the heart of concession Please cite this article as: A. Ng, Martin Loosemore, Risk allocation in the private provision of public infrastructure, International Journal of Project Management(2006), doi: 10. 1016/j. ijproman 2006.06.005
accountability are high, as are political pressures which can interfere with the effective funding, management and procurement of a project. For example, the Victorian Department of Treasury and Finance in Australia, which annually procures over $1.8 billion of public infrastructure, noted that the political and social context in which infrastructure projects are undertaken requires that public consultation be fully integrated into the planning process [6]. It argues that at the core of such projects is a complex web of relationships among bureaucrats, politicians, media, employees, general public [local, national and sometimes global], labour and special interest groups. Irrespective of any ideological preferences that Governments of the day may promote, these interest groups invariably have high expectations in relation to the management of issues such as the environment, health and safety, industrial relations and access and equity. As Sharp [6] notes, any infrastructure project lives or dies on its reputation with these people. Despite the existence of many complex risks which can interfere with the success of infrastructure projects, the private sector has been keen to take over the traditional role of the public sector in financing, procuring and managing such assets [5,7,8]. However, recent research has indicated that even on the largest Public, Private Partnership (PPP) projects, risk management practices are highly variable, intuitive, subjective and unsophisticated [4]. In this context, the aim of this paper is to explore the considerations to be made in effective risk distribution between the public and private sectors on such projects. 2. The structure of PPP agreements PPP is an evolving concept which takes many forms around the world. However, it is essentially an arrangement by which private parties participate in, or provide support for the provision of infrastructure-base services. In contrast to traditional public procurement which involves the public sector purchasing an asset, the PPP system involves the purchase of a stream of services, defined in a detailed service agreement under specified terms and conditions [9,12]. In simple terms, this is done via a concession contract which involves a host government granting a licence or concession to a private consortium [concessionaire, promoter or sponsor] which sets up a single purpose entity known as a Special Purpose Vehicle (SPV) using contracts secondary to the concession, to finance, design, build, operate and maintain an infrastructure project for a set period of time known as the concession period. The private consortium is normally formed by a joint venture (JV) between a range of organisations including contractors, facilities managers, banks, investors and suppliers, which are willing to commit equity and/or resources to the project. Payments to the SPV to fund debt service normally commence after completion of the construction – when the services have be made available to the public. During the operating period, the SPV receives income based on the usage of the facility [which may be guaranteed] assuming that the service provided meets a range of key performance indicators. For example, the Colombian government agreed to reimburse the consortium which constructed the recent El Cortijo-El Vino Toll road in Colombia, if traffic was less than 90% of a specified level and went even further in the new runway at Bogota’s El Dorado Airport, guaranteeing a minimum revenue. There are normally abatement clauses in the concession contract, which can penalise [sometimes excessively] the SPV for falling below these standards. Furthermore, there are sometimes penalty points, which if accumulated to a certain level can lead to termination of the contract for poor performance. At the end of the operating period, the fully operational project is transferred back to the host government, usually at nominal or no cost [10]. 3. Balancing public/private risks and rewards in PPP projects In investing in a PPP project, private sector companies aim to achieve a return on their investment in generating sufficient future cash flows to cover initial capital costs and finance charges, thereby providing enough profit to invest in future projects and pay shareholder dividends. In contrast, the aim of the public sector is to ensure a level of service to the community which is timelier, more cost efficient and higher quality than if the public sector had retained responsibility. While some rather anecdotal evidence exists to indicate that PPP projects can better serve community objectives, than traditional public provision, by reducing risk exposure, the debate is likely to continue until a sufficient number of projects have been studied in detail over their entire life-cycle. Nevertheless, the current arguments from each side of the fence are summarised below. 3.1. Risks in the private provision of infrastructure It has been argued that concession contracts involve relatively high waste, rework and transaction costs due to lengthy and complex tendering arrangements and post-tender negotiations resulting from overly optimistic public sector comparators, large numbers of stakeholders and the complex web of contracts and financial structuring needed to bind them together. It also has been argued that the complexity of these arrangements increase public sector risk rather than reduce it, increases service costs for the public and represent a barrier to entry to small companies which is unfair and reduces competition [7]. Another argument against PPPs is that in some situations, they are not economically viable for the private sector without exorbitant risk-related service charges for the public. A good example in Australia are rural roads which do not provide the volume of traffic needed to justify a user-pays approach which lies at the heart of concession 2 A. Ng, M. Loosemore / International Journal of Project Management xxx (2006) xxx–xxx ARTICLE IN PRESS Please cite this article as: A. Ng, Martin Loosemore, Risk allocation in the private provision of public infrastructure, International Journal of Project Management (2006), doi:10.1016/j.ijproman.2006.06.005
ARTICLE IN PRESS 4. Ng. M. Loosemore /International Journal of Project Management xxx(2006)xxx-xxx contracts [ll]. In this situation a DCM contract is likely to ees, abuses of monopoly power to the detriment of public provide better value for money to the public, while still interest, lower national security and less incentive to con- providing the private sector with an incentive to consider trol the potentially huge impact of infrastructure develop- the whole life cycle costs of an asset. It does this by com- ments on the environment [5, 9]. In recent years, paring costs with a public sector comparator, which repre- environmental and anti-globalisation pressure groups have sents a hypothetical risk adjusted cost estimate based on an become particularly vocal and active in highlighting the assumption that assets are acquired through conventional social, cultural and ecological impact of large international procurement routes and that the procurer retains signifi- PPP projects. These activists can expose participating cant risk exposure [12] organisations to unbearable reputational risks, as was illus Furthermore, in order to compensate for the largely trated in the case of Balfour Beaty and Skanska which unknown risks involved over periods of up to 35 years, recently pulled-out of the Llisu Dam project in Turkey in the private sector will inevitably demand high risk premi- fear of the adverse publicity it would bring [18] ums,as will the financial institutions and members of sup- ply chains which serve them. The result is an accumulation 3. 2. Risks in the public provision of infrastructure of risk premiums throughout the project supply chain reflecting more a public inability and/or unwillingness to Critics of the public provision of public infrastructure sibil:a ge risk rather than a more efficient transfer of respon- argue that given proper incentives, regulation and control, sibility. It is argued that this uncertainty is inevitably trans- the private sector is better placed to deliver value for ferred to the public in the form of higher charges for the money to the public [19, 11]. This derives from shareholder services delivered. There is also substantial evidence that pressures for performance and accountability, greater clar- it is difficult to predict the extent of risk in such projects, ity of objectives, higher management expertise and auton- of inappropriate risk distribution between the public and omy, lower levels of regulation and control, a competitive private sectors and of overly optimistic businesses cases environment, continuous improvement against clear key and resulting in some very high profile debacles, which performance indicators, access to a wider range of equity have cost taxpayers many millions of dollars. For example, not available to the public sector and managerial incentives in the recently completed Sydney Cross City tunnel, which and rewards for innovations. Advocates of PPP projects was completed 6 weeks ahead of schedule, initial predic- argue that these benefits are vividly illustrated in successful tions of initial usage rates of 35,000 per day, one month concession projects such as the privately funded M4 free- after opening, it was only 20,000 per day after 6 weeks. way in Sydney which was completed 6 months ahead of When the government introduced a 5 week toll-free period schedule, the third runway at Sydney Airport which was to encourage usage, patronage only rose to 55,000 per day completed 15 weeks ahead of schedule and $30 million and it is now faced with the prospect of paying compensa- under budget, offering a total saving of $200 million and tion to the project consortium [13] Similar experiences Junee Prison in Australia which was completed 3 months have been reported on other major projects such as the ahead of schedule and under budget saving an estimated Sydney Airport Rail Link and Eurotunnel project where $3 million in procurement costs government business cases were also too optimistic and It has also been argued that the shift of funding respon- ignored other competition [14]. Even today, Eurotunnel is sibility to the private sector reduces public debt and finance operating at about half its capacity and there are continu- costs, freeing up money to invest in other areas of public ous attempts to attract business from cut price ferries and interest such as education and welfare. It is also possible airlines and expand the business by increasing the freight for the public sector to reduce its in-house project manage- There are also serious questions of effectiveness in some does this release more money to invest in publie ot only arm-including attracting long-haul rail freight ment and maintenance workforce and equipment. Not only sectors. For example in the health care sector and other but due to the almost unlimited finance capacity of the highly specialised areas, the outsourcing of projects to the entire private sector, projects that otherwise might not have private sector has led to problems associated with a lack been built for some time, can be delivered many years ear- of understanding of the unique technologies, cultures and lier than anticipated. For example, in Australia, it has been politics involved in this sector [15,16]. This is particularly argued that the private involvement in the new Port Mac- the case when they are contractor-led rather than Facility quarie hospital has delivered an upgrade in health services Manager-led. For example in the UK, research shows that to the community long before the public sector alone could in the prison sector, where all SPVs are facility manage- have done so ment led, the clients feel more understood and happier than Another benefit of PPP projects is the whole of life cycle in education, where SPVs tend to be contractor-led [17]. approach it encourages in the procurement and manage- Finally, there are fears among public interest groups ment of public sector assets. By creating a single point of uch as unions that the private provision of infrastructure responsibility for an entire project from inception through effectively represents the privatisation of public services. design, construction and operation, a strong incentive It has been argued that this will result in a major loss of created to think about the implications which a design or control over working conditions for public sector employ- construction decision will have on the operating effective Please cite this article as: A. Ng, Martin Loosemore, Risk allocation in the private provision of public infrastructure, International Journal of Project Management(2006), doi: 10. 1016/j. ijproman 2006.06.005
contracts [11]. In this situation a DCM contract is likely to provide better value for money to the public, while still providing the private sector with an incentive to consider the whole life cycle costs of an asset. It does this by comparing costs with a public sector comparator, which represents a hypothetical risk adjusted cost estimate based on an assumption that assets are acquired through conventional procurement routes and that the procurer retains signifi- cant risk exposure [12]. Furthermore, in order to compensate for the largely unknown risks involved over periods of up to 35 years, the private sector will inevitably demand high risk premiums, as will the financial institutions and members of supply chains which serve them. The result is an accumulation of risk premiums throughout the project supply chain, reflecting more a public inability and/or unwillingness to manage risk rather than a more efficient transfer of responsibility. It is argued that this uncertainty is inevitably transferred to the public in the form of higher charges for the services delivered. There is also substantial evidence that it is difficult to predict the extent of risk in such projects, of inappropriate risk distribution between the public and private sectors and of overly optimistic businesses cases and resulting in some very high profile debacles, which have cost taxpayers many millions of dollars. For example, in the recently completed Sydney Cross City tunnel, which was completed 6 weeks ahead of schedule, initial predictions of initial usage rates of 35,000 per day, one month after opening, it was only 20,000 per day after 6 weeks. When the government introduced a 5 week toll-free period to encourage usage, patronage only rose to 55,000 per day and it is now faced with the prospect of paying compensation to the project consortium [13]. Similar experiences have been reported on other major projects such as the Sydney Airport Rail Link and Eurotunnel project where government business cases were also too optimistic and ignored other competition [14]. Even today, Eurotunnel is operating at about half its capacity and there are continuous attempts to attract business from cut price ferries and airlines and expand the business by increasing the freight arm – including attracting long-haul rail freight. There are also serious questions of effectiveness in some sectors. For example, in the health care sector and other highly specialised areas, the outsourcing of projects to the private sector has led to problems associated with a lack of understanding of the unique technologies, cultures and politics involved in this sector [15,16]. This is particularly the case when they are contractor-led rather than Facility Manager-led. For example in the UK, research shows that in the prison sector, where all SPVs are facility management led, the clients feel more understood and happier than in education, where SPVs tend to be contractor-led [17]. Finally, there are fears among public interest groups such as unions that the private provision of infrastructure effectively represents the privatisation of public services. It has been argued that this will result in a major loss of control over working conditions for public sector employees, abuses of monopoly power to the detriment of public interest, lower national security and less incentive to control the potentially huge impact of infrastructure developments on the environment [5,9]. In recent years, environmental and anti-globalisation pressure groups have become particularly vocal and active in highlighting the social, cultural and ecological impact of large international PPP projects. These activists can expose participating organisations to unbearable reputational risks, as was illustrated in the case of Balfour Beaty and Skanska which recently pulled-out of the Llisu Dam project in Turkey in fear of the adverse publicity it would bring [18]. 3.2. Risks in the public provision of infrastructure Critics of the public provision of public infrastructure argue that given proper incentives, regulation and control, the private sector is better placed to deliver value for money to the public [19,11]. This derives from shareholder pressures for performance and accountability, greater clarity of objectives, higher management expertise and autonomy, lower levels of regulation and control, a competitive environment, continuous improvement against clear key performance indicators, access to a wider range of equity not available to the public sector and managerial incentives and rewards for innovations. Advocates of PPP projects argue that these benefits are vividly illustrated in successful concession projects such as the privately funded M4 freeway in Sydney which was completed 6 months ahead of schedule, the third runway at Sydney Airport which was completed 15 weeks ahead of schedule and $30 million under budget, offering a total saving of $200 million and, Junee Prison in Australia which was completed 3 months ahead of schedule and under budget saving an estimated $3 million in procurement costs. It has also been argued that the shift of funding responsibility to the private sector reduces public debt and finance costs, freeing up money to invest in other areas of public interest such as education and welfare. It is also possible for the public sector to reduce its in-house project management and maintenance workforce and equipment. Not only does this release more money to invest in public services but due to the almost unlimited finance capacity of the entire private sector, projects that otherwise might not have been built for some time, can be delivered many years earlier than anticipated. For example, in Australia, it has been argued that the private involvement in the new Port Macquarie hospital has delivered an upgrade in health services to the community long before the public sector alone could have done so. Another benefit of PPP projects is the whole of life cycle approach it encourages in the procurement and management of public sector assets. By creating a single point of responsibility for an entire project from inception through design, construction and operation, a strong incentive is created to think about the implications which a design or construction decision will have on the operating effectiveA. Ng, M. Loosemore / International Journal of Project Management xxx (2006) xxx–xxx 3 ARTICLE IN PRESS Please cite this article as: A. Ng, Martin Loosemore, Risk allocation in the private provision of public infrastructure, International Journal of Project Management (2006), doi:10.1016/j.ijproman.2006.06.005
ARTICLE IN PRESS A. Ng, M. Loosemore /International Journal of Project Management xxx (2006)xxx-xxx ness and costs of a managing and maintaining a facility the project consortium if traffic flows and resulting toll during its operational life. Since over a 25-year period, income fell below a certain level [23]. these costs can be as high as 10 times initial capital costs While general risk classifications such as the above are this results in enormous savings for the public sector useful, it is also useful to consider the special risks associ- [20, 21]. For example, it has been estimated that as much ated with the PPP procurement process. After all, it is quite as 10% of the capital investments made in health facilities different to the traditional procurement process which sep in Australia are wasted due to poor facilities which create arates financing, design, construction and operational huge maintenance liabilities [22]. When one considers the responsibilities. In doing so, Standard and Poor's considers amount of money being spent, this is an enormous prob- several broad areas that can potentially affect a PPP pro- lem. For example, with the overall health related capital ject's creditworthiness. These are investment in Australia being about $3.9 billion a year, the potential savings of improving the health facilities plan- Credit risk of the public sector entity- Since the SPV ning process could be $390 million per year [22 relies on a payment stream from the government count erparty to satisfy its debt service obligations there is a 4. Risk classifications in PPP projects significant risk in the counterparty's creditworthiness Construction risks- although construction covers only Given the complexity, size and time frame of concession 34 years of perhaps a 30 year total debt exposure, the contracts, there are an enormous range of potential risks successful completion of the construction period is par- which can affect expected outcomes. Nevertheless, in very amount to servicing that debt. Delays can be disastrous simple terms, these can be classified into two main groups: and their potential is related to the design and techno- general risks or project risk logical complexity of construction; the contractor's Project risks arise from the way a project is managed or management team and approach; existing workload from events in its immediate microenvironment. They may and problems on other projects; reputation; third party include natural risks such as ground problems and weather support via bonds and guarantees and; the contractors conditions, technical problems associated with designs, experience, resources and capabilities plant and equipment, materials problems associated with Revenue structure- How certain or controllable is the suppliers, organisational problems associated with subcon revenue stream, what is the level of penalty and abate- tractors, manpower problems associated with unions, con- ment for under performance, what are the index linked tractual problems associated with Jv agreements and payment environmental problems associated with pollution, etc. Operating risk- What are the maintenance and replace- In contrast, general risks are not directly associated with ment regimes and costs? Is service provider liability for project strategies, yet can have a significant impact on its poor performance capped? Are levels of abatement outcome. These normally arise from natural, political, reg- appropriate and fair? How reliable are service providers? ulatory, legal and economic events in the general macron- Do they have a presence in the bidding consortium? vironment surrounding the project. For example, the What are the levels of competition for service providers? 2,015 MW Dabhol power plant in india was ordered to etc stop by the newly elected Maharashtra government in Financial and legal structure- Typically, PPP projects August 1995; the Tiananmen Square incident in China on have fully amortizing debt maturing in 30 years. Projects 4th June 1989 resulted in the syndication of loans for the are typically highly geared at around 80-90%. Thus the new Guanzho- Shenzen-Zhuhai super highway to be sufficiency and sensitivity of cash flows to different poten delayed until 1991 and; a 45 km BOT toll road in Shenzhen tial risks is crucial to establish how debt will be serviced was delayed because the consortium and government could To manage this, structural protective mechanisms and not agree on appropriate toll charges [9, 23]. In other exam- financial security packages can be useful such as guaran- ples of poorly managed general risks, the US$2.5 billion tees or bonds, operating accounts and reserves, etc. Malaysia North-South highway suffered a 75% cost over run largely due to inadequate allowances being made for More recently, Grimsey and Lewis have identified six inflation. Furthermore, in the 1970s, the Spanish govern- areas of risk associated with PPP projects, namely, public ment guaranteed 75% of the loans on its new highway net- risk; asset risk, operating risk, sponsor risk, financial risk work and assumed the full exchange-rate risk, a decision and default risk [14]. Public risk relates to the governments that eventually cost the Spanish Taxpayer an estimated duty to ensure that the facilities are constructed in accor USS2.7 billion. To help mitigate such risks, governments dance with legislation and codes of practices to ensure often guarantee exchange rates. For example, in a major the well being of workers and consumers. Step-in rights road project in Vietnam, assurances were given by the gov- usually exist in most PPP contracts to allow the govern- ernment bank on the right to convert Dong (Vietnams ment to intervene if this risk eventuates. Asset risk can arise Currency) into foreign currency at a certain rate of if the life of a facility proves to be shorter than anticipated exchange. Similarly, on the North-South Highway project if the costs of maintenance exceed that expected, the asset in Malaysia, the government undertook to compensate may be damaged or destroyed by a force majeure event, Please cite this article as: A. Ng, Martin Loosemore, Risk allocation in the private provision of public infrastructure, International Journal of Project Management(2006), doi: 10. 1016/j. ijproman 2006.06.005
ness and costs of a managing and maintaining a facility during its operational life. Since over a 25-year period, these costs can be as high as 10 times initial capital costs, this results in enormous savings for the public sector [20,21]. For example, it has been estimated that as much as 10% of the capital investments made in health facilities in Australia are wasted due to poor facilities which create huge maintenance liabilities [22]. When one considers the amount of money being spent, this is an enormous problem. For example, with the overall health related capital investment in Australia being about $3.9 billion a year, the potential savings of improving the health facilities planning process could be $390 million per year [22]. 4. Risk classifications in PPP projects Given the complexity, size and time frame of concession contracts, there are an enormous range of potential risks which can affect expected outcomes. Nevertheless, in very simple terms, these can be classified into two main groups: general risks or project risks. Project risks arise from the way a project is managed or from events in its immediate microenvironment. They may include natural risks such as ground problems and weather conditions, technical problems associated with designs, plant and equipment, materials problems associated with suppliers, organisational problems associated with subcontractors, manpower problems associated with unions, contractual problems associated with JV agreements and environmental problems associated with pollution, etc. In contrast, general risks are not directly associated with project strategies, yet can have a significant impact on its outcome. These normally arise from natural, political, regulatory, legal and economic events in the general macroenvironment surrounding the project. For example, the 2.015 MW Dabhol Power Plant in India was ordered to stop by the newly elected Maharashtra government in August 1995; the Tiananmen Square incident in China on 4th June 1989 resulted in the syndication of loans for the new Guanzho–Shenzen–Zhuhai super highway to be delayed until 1991 and; a 45 km BOT toll road in Shenzhen was delayed because the consortium and government could not agree on appropriate toll charges [9,23]. In other examples of poorly managed general risks, the US$2.5 billion Malaysia North-South highway suffered a 75% cost overrun largely due to inadequate allowances being made for inflation. Furthermore, in the 1970s, the Spanish government guaranteed 75% of the loans on its new highway network and assumed the full exchange-rate risk, a decision that eventually cost the Spanish Taxpayer an estimated US$2.7 billion. To help mitigate such risks, governments often guarantee exchange rates. For example, in a major road project in Vietnam, assurances were given by the government bank on the right to convert Dong (Vietnam’s Currency) into foreign currency at a certain rate of exchange. Similarly, on the North-South Highway project in Malaysia, the government undertook to compensate the project consortium if traffic flows and resulting toll income fell below a certain level [23]. While general risk classifications such as the above are useful, it is also useful to consider the special risks associated with the PPP procurement process. After all, it is quite different to the traditional procurement process which separates financing, design, construction and operational responsibilities. In doing so, Standard and Poor’s considers several broad areas that can potentially affect a PPP project’s creditworthiness. These are: Credit risk of the public sector entity – Since the SPV relies on a payment stream from the government counterparty to satisfy its debt service obligations there is a significant risk in the counterparty’s creditworthiness. Construction risks – although construction covers only 3–4 years of perhaps a 30 year total debt exposure, the successful completion of the construction period is paramount to servicing that debt. Delays can be disastrous and their potential is related to the design and technological complexity of construction; the contractor’s management team and approach; existing workloads and problems on other projects; reputation; third party support via bonds and guarantees and; the contractor’s experience, resources and capabilities. Revenue structure – How certain or controllable is the revenue stream, what is the level of penalty and abatement for under performance, what are the index linked payment periods, etc. Operating risk – What are the maintenance and replacement regimes and costs? Is service provider liability for poor performance capped? Are levels of abatement appropriate and fair? How reliable are service providers? Do they have a presence in the bidding consortium? What are the levels of competition for service providers? etc. Financial and legal structure – Typically, PPP projects have fully amortizing debt maturing in 30 years. Projects are typically highly geared at around 80–90%. Thus the sufficiency and sensitivity of cash flows to different potential risks is crucial to establish how debt will be serviced. To manage this, structural protective mechanisms and financial security packages can be useful such as guarantees or bonds, operating accounts and reserves, etc. More recently, Grimsey and Lewis have identified six areas of risk associated with PPP projects, namely; public risk; asset risk, operating risk, sponsor risk, financial risk and default risk [14]. Public risk relates to the government’s duty to ensure that the facilities are constructed in accordance with legislation and codes of practices to ensure the well being of workers and consumers. Step-in rights usually exist in most PPP contracts to allow the government to intervene if this risk eventuates. Asset risk can arise if the life of a facility proves to be shorter than anticipated, if the costs of maintenance exceed that expected, the asset may be damaged or destroyed by a force majeure event, 4 A. Ng, M. Loosemore / International Journal of Project Management xxx (2006) xxx–xxx ARTICLE IN PRESS Please cite this article as: A. Ng, Martin Loosemore, Risk allocation in the private provision of public infrastructure, International Journal of Project Management (2006), doi:10.1016/j.ijproman.2006.06.005.
ARTICLE IN PRESS 4. Ng. M. Loosemore /International Journal of Project Management xxx(2006)xxx-xxx etc. These risks can be mitigated by agreed maintenance optimal distribution there are several important and well and refurbishment scheduled, etc. Operating risk reflect established rules to follow [24]. They are, that a risk should the chance that the purchased services are not delivered only be given to someone who as agreed in terms of specification, costs or timing. Sponsor risk arises when the sPv is unable to meet its contractual. has been made fu of the risks they are taking obligations and the government is unable to enforce them Has the greatest [expertise and authority] to or recover compensation. Normally, parent company guar manage the risk ly and efficiently(and thus antees, performance bonds and sureties are used to miti- charge the lowest risk premium). gate operating and sponsor risks. Financial risk can arise Has the capability and resources to cope with the risk from prices and costs increases, financiers withdrawing, eventuating interest rates increasing or from poorly designed financial Has the necessary risk appetite to want to take the risk. structures. Finally, default risk can arise when a party is Has been given the chance to charge an appropriate pre- unable to perform its contractual obligations on time or mium for taking it to defined standards. In this case the contract will provide for remedies such as obligations to rectify, abatements, Not following these simple rules will compromise the step-in rights, termination and the transfer of completed success and efficiency of the project since it will produce assets according to a predefined valuation mechanism. higher risk premiums than necessary, increase the chance Although we have a good understanding of the risks of risks arising and the consequences if they do arise [14] associated with PPP projects, what is less known is how Further inefficiencies can arise from confused responsibility these risks change over the duration of a project [24]. While for monitoring and responding to risks; resentment for there have been some attempts to broadly define risk pro- being forced to take them and; denial, conflict and dispute files over the term of a PPP project [25], such models to avoid responsibility when they do arise. In effect, by not remain rudimentary making it difficult to produce an over- following the above rules, the public sector is merely gain all risk allocation structure with mechanisms which coordi- ing the illusion of risk transfer, since it is likely that the risk nate to ensure that all risks are appropriately managed will be transferred back to them in the form of higher risks, during all stages of a project. This should be a priority risk premiums and project problems for future research To help ensue that this does not happen, a number of standard risk allocation matrices have been produced to 5. Risk allocation in PPP projects guide appropriate risk allocation in PPP projects, most of which agree on the general allocation of risks [14, 261. As Grimsey and Lewis point out, risk allocation is PPP Grimsey and Lewis's model is typical and is presented in projects is fundamentally different to that in traditional Table I projects [14]. In the latter, the public sector purchases an While useful as a guide to government and private sec asset from private sector contractors and consultants tors, it is very important to realise the limitation of such whose liability is limited to the design and construction models and that risks must be analysed and managed of the asset. Finance and operational risks remain with a project-by-project basis. First the above table presents the public sector. In contrast, the PPP model involves the broad categories of risk and every project has a different purchase of a relatively risk-free long-term service and array of risks, which need to be thoroughly analysed and the government accepts no asset-based risk and does not understood. It is also important to recognise that the pay, or is entitled to reduce payments, abatements and appropriate distribution of risks is dependent on the compensation if the service is not delivered to the specified resources and capabilities of the parties to a contract and standards, as defined in a service agreement. They key dri- this can vary considerably. This is brought into focus wher ver of any PPP project is value for money and driven by a risk is completely outside the control of both parties this requirement, the government has to decide how the Here risks are often identified as being shared or insured risks identified above must be best distributed between against but in reality the best allocation of risk will depend the parties to a PPP project. Conceptually, as service reci- on how the private parties price the risk, whether this is pient paying only for satisfactory service, the assumption is reasonable for the public sector and how it compares to made that all project risks should be borne by the private the potential risk(in cost and probability terms) if retained ctor. In theory, the idea of transferring a risk is that some by the public sector. Finally, these are static models of risk other party is provided with an incentive to manage it effec- and risk distribution mechanisms need to reflect that risks tively. However, in reality, the government has to deter- change considerably over the life of a projed e o suggest mine value-for-money basis, what risks it should take back to achieve an optimal risk distribution. This that risk transfer is often handled poorly between parties to involves"taking back"risks which are more efficiently many PPP projects and that these types of problems are managed by it and this is normally done via three mecha- common in PPP projects. For a host of reasons, parties nisms: specified service obligations, payment mechanisms to concession projects take risks which they are not clear and contractual provisions [14]. However, in achieving an of, that they are not able to cope with, that they do not Please cite this article as: A. Ng, Martin Loosemore, Risk allocation in the private provision of public infrastructure, International Journal of Project Management(2006), doi: 10. 1016/j. ijproman 2006.06.005
etc. These risks can be mitigated by agreed maintenance and refurbishment scheduled, etc. Operating risk reflect the chance that the purchased services are not delivered as agreed in terms of specification, costs or timing. Sponsor risk arises when the SPV is unable to meet its contractual obligations and the government is unable to enforce them or recover compensation. Normally, parent company guarantees, performance bonds and sureties are used to mitigate operating and sponsor risks. Financial risk can arise from prices and costs increases, financiers withdrawing, interest rates increasing or from poorly designed financial structures. Finally, default risk can arise when a party is unable to perform its contractual obligations on time or to defined standards. In this case the contract will provide for remedies such as obligations to rectify, abatements, step-in rights, termination and the transfer of completed assets according to a predefined valuation mechanism. Although we have a good understanding of the risks associated with PPP projects, what is less known is how these risks change over the duration of a project [24]. While there have been some attempts to broadly define risk pro- files over the term of a PPP project [25], such models remain rudimentary making it difficult to produce an overall risk allocation structure with mechanisms which coordinate to ensure that all risks are appropriately managed during all stages of a project. This should be a priority for future research. 5. Risk allocation in PPP projects As Grimsey and Lewis point out, risk allocation is PPP projects is fundamentally different to that in traditional projects [14]. In the latter, the public sector purchases an asset from private sector contractors and consultants whose liability is limited to the design and construction of the asset. Finance and operational risks remain with the public sector. In contrast, the PPP model involves the purchase of a relatively risk-free long-term service and the government accepts no asset-based risk and does not pay, or is entitled to reduce payments, abatements and compensation if the service is not delivered to the specified standards, as defined in a service agreement. They key driver of any PPP project is value for money and driven by this requirement, the government has to decide how the risks identified above must be best distributed between the parties to a PPP project. Conceptually, as service recipient paying only for satisfactory service, the assumption is made that all project risks should be borne by the private sector. In theory, the idea of transferring a risk is that some other party is provided with an incentive to manage it effectively. However, in reality, the government has to determine, on a value-for-money basis, what risks it should take back to achieve an optimal risk distribution. This involves ‘‘taking back’’ risks which are more efficiently managed by it and this is normally done via three mechanisms: specified service obligations, payment mechanisms and contractual provisions [14]. However, in achieving an optimal distribution there are several important and well established rules to follow [24]. They are, that a risk should only be given to someone who: Has been made fully aware of the risks they are taking. Has the greatest capacity [expertise and authority] to manage the risk effectively and efficiently (and thus charge the lowest risk premium). Has the capability and resources to cope with the risk eventuating. Has the necessary risk appetite to want to take the risk. Has been given the chance to charge an appropriate premium for taking it. Not following these simple rules will compromise the success and efficiency of the project since it will produce higher risk premiums than necessary, increase the chance of risks arising and the consequences if they do arise [14]. Further inefficiencies can arise from confused responsibility for monitoring and responding to risks; resentment for being forced to take them and; denial, conflict and dispute to avoid responsibility when they do arise. In effect, by not following the above rules, the public sector is merely gaining the illusion of risk transfer, since it is likely that the risk will be transferred back to them in the form of higher risks, risk premiums and project problems. To help ensue that this does not happen, a number of standard risk allocation matrices have been produced to guide appropriate risk allocation in PPP projects, most of which agree on the general allocation of risks [14,26]. Grimsey and Lewis’s model is typical and is presented in Table 1. While useful as a guide to government and private sectors, it is very important to realise the limitation of such models and that risks must be analysed and managed on a project-by-project basis. First the above table presents broad categories of risk and every project has a different array of risks, which need to be thoroughly analysed and understood. It is also important to recognise that the appropriate distribution of risks is dependent on the resources and capabilities of the parties to a contract and this can vary considerably. This is brought into focus when a risk is completely outside the control of both parties. Here risks are often identified as being shared or insured against but in reality the best allocation of risk will depend on how the private parties price the risk, whether this is reasonable for the public sector and how it compares to the potential risk (in cost and probability terms) if retained by the public sector. Finally, these are static models of risk and risk distribution mechanisms need to reflect that risks change considerably over the life of a project. Unfortunately, there is considerable evidence to suggest that risk transfer is often handled poorly between parties to many PPP projects and that these types of problems are common in PPP projects. For a host of reasons, parties to concession projects take risks which they are not clear of, that they are not able to cope with, that they do not A. Ng, M. Loosemore / International Journal of Project Management xxx (2006) xxx–xxx 5 ARTICLE IN PRESS Please cite this article as: A. Ng, Martin Loosemore, Risk allocation in the private provision of public infrastructure, International Journal of Project Management (2006), doi:10.1016/j.ijproman.2006.06.005.